INTRODUCTION
Problem Statement and Significance of Research
Child labor and health shocks are critical issues in developing countries, with significant implications for research and policy Governments recognize the necessity of protecting children and fostering an environment conducive to their development (UNESCO, 2008) Reducing child labor is a primary objective for many international organizations and governments Health shocks have been shown to negatively affect household outcomes, including children's well-being (Beegle et al., 2004; Dilion, 2012; Alam & Mahal, 2014) By exploring the interplay between health shocks and child labor, this paper aims to provide insights that can enhance child development in Vietnam and similar contexts The following discussion will offer a comprehensive overview of these issues, underscoring the importance of this research.
Health shocks pose significant risks to households in developing countries, leading to detrimental outcomes such as income reduction, increased expenditures, and labor supply imbalances Long-term effects may include psychological issues and chronic illnesses (Alam & Mahal, 2014) These unpredictable idiosyncratic risks can severely strain household budgets (Wagstaff, 2007; Bandara et al., 2015; Mitra et al., 2016) In Vietnam, the crude death rate rose from 5.521% to 5.815% between 2002 and 2014, primarily due to cardiovascular diseases and diabetes (World Bank, 2017; WHO, 2015) Additionally, health expenditure in Vietnam increased from 5.2% to 6.9% of GDP, reaching nearly 191 trillion VND in 2014, with private spending accounting for over 52% of total healthcare costs Households experienced an average monthly healthcare expenditure increase of $11.4 from 2002 to 2014, totaling about $116 annually, highlighting the ongoing burden of catastrophic health expenses.
High healthcare spending in Vietnam, which stands at 2% compared to the global average of 45.5% (World Bank, 2017), can lead to significant financial burdens that risk pushing households into poverty Various studies indicate that health risks negatively impact earned income, increase out-of-pocket expenditures, reduce labor productivity, and lower individuals' BMI (Wagstaff, 2006; Van Minh et al., 2012; Bales, 2013) Consequently, it is clear that health shocks disrupt household stability and warrant attention in developing countries like Vietnam.
In response to health shocks, households adopt various coping strategies to stabilize their lives, including utilizing savings, trading livestock, selling assets, accessing credit, and altering labor supply, which often involves child labor (Bandara et al., 2015; Bonfer & Wright, 2016) When faced with budget constraints and labor shortages, parents may resort to sending their children into the labor market, as children can contribute to household income and help alleviate the financial strain caused by illness or death in the family (Basu & Van, 1998) Children often take on more work to support their families, particularly in low- and middle-income countries where the demand for unskilled labor remains high, especially in agriculture Consequently, the incidence of child labor is expected to rise in households experiencing health shocks (Brown et al., 2002).
Child labor significantly harms both childhood experiences and future opportunities for children (ILO, 2013) When children prioritize work over education and leisure, their overall development and human capital suffer Additionally, child labor is often linked to inadequate nutrition and compromised survival Many children are forced to work excessive hours beyond their capacity, leading to severe vulnerabilities As a result, some children are unable to attend school, while others miss out on essential childhood experiences.
A study by Beegle et al (2004) highlights the detrimental impact of child labor on educational achievement in Vietnam The findings indicate that child labor not only hampers academic success but also diminishes the child's potential human capital, which is crucial for future economic development.
Child labor persists despite numerous international conventions and domestic laws aimed at its eradication The Vietnamese government has implemented various measures to reduce child employment, including ratifying the Worst Forms of Child Labor Convention (C182) and the Minimum Age for Admission to Employment Convention (C138) in 2000 Domestically, the amended Labor Code of 2012 prohibits child labor for those under 15, with specific exceptions outlined in Circular No 10/2013/TT/LDTBXH, which identifies occupations and locations where adolescent labor is banned Additionally, Decision No 1023/QD-TT, issued by Vietnam's Prime Minister in June 2016, launched a national project aimed at preventing and minimizing child labor from 2016 to 2020, focusing on combating illegal child labor and providing development opportunities for children Despite these efforts, approximately 2.83 million children, predominantly from rural areas, continue to engage in economic activities in Vietnam.
According to the Viet Nam National Child Labor Survey conducted in 2012, approximately 1.315 million children are engaged in hazardous labor The survey reveals that among children aged 5 to 17 who are not enrolled in school, about 21% are forced into paid work and 9.2% are unable to invest in their education Despite the government's various efforts to safeguard children's rights, the issue of child labor continues to be a significant problem.
The exploration of child labor and health shocks in Vietnam reveals compelling narratives worthy of study Numerous studies have examined the effects of health shocks on household outcomes and the coping strategies employed by families (Wagstaff, 2007; Mitra et al., 2015) Additionally, significant research has been conducted on the intersection of children and child labor, highlighting the complexities of these issues (Rosati & Tzannatos, 2000; Edmonds & Pavcnik, 2002; Beegle et al.).
This paper explores the connection between health shocks and child labor, addressing a significant gap in existing research It investigates how health-related crises impact child labor dynamics while also examining the role of household assets and access to credit as mechanisms that may assist families in managing these challenges.
This article explores the impact of asset holdings and access to credit on child labor in Vietnam, particularly in the context of health shocks experienced by households It highlights the buffering effect that financial resources can have in reducing the reliance on child labor during challenging times The findings aim to provide recommendations for supporting families, ultimately fostering the protection of childhood and mitigating the prevalence of child labor in the region.
Research Objectives and Research Questions
This study examines the effects of health shocks on child labor in developing countries, with a focus on Vietnam's dataset It specifically investigates how children's working hours are affected when households face health-related crises, such as the illness or death of a family member Additionally, the research explores potential strategies that families may employ to mitigate the impact of these health shocks on child labor To fulfill its objectives, the study aims to answer two key research questions.
(i) The first question: if a family experienced health risk events such as death, illness of household members, the child labor will increase or not?
(ii) The second question: whether other mechanisms such as accessing the credit,asset holdings can reduce the negative effects of health shock on child labor?
Scope of the study
This study examines the connection between child labor and health shocks, highlighting critical issues for developing countries Utilizing a longitudinal dataset from the Young Lives project in the UK, the research analyzes data from Vietnam collected in 2006 and 2009, focusing on children aged 4 to 16 years.
Structure of Thesis Design
This study is structured into several chapters: Chapter 2 reviews relevant theories and empirical research on child labor in economics, the impact of health shocks on household outcomes, and the interplay between health shocks and child labor Chapter 3 outlines the research methodology, detailing the theoretical and estimation models, along with an introduction to the Vietnam Young Lives dataset Chapter 4 presents the statistical description of the data, estimation results, and discussions Finally, the concluding chapter summarizes key findings and insights.
LITERATURE REVIEW
Impacts of Health Shocks on Household Outcome
Health shocks have significant negative impacts on household outcomes, particularly among low-income families, leading to adverse economic and non-economic effects (O'Donnell et al., 2005; Wagstaff, 2007; Alam & Mahal, 2014) These shocks often result in reduced income and an imbalance in labor supply, which can increase the incidence of child labor This section aims to provide a comprehensive understanding of the consequences of health shocks on households and explore the coping mechanisms that link health shocks to child labor in the subsequent section.
Health shocks refer to adverse events affecting the health of family members, such as illnesses, injuries, or even the death of a loved one (Alam & Mahal, 2014) These significant health events can have profound implications for families, influencing their emotional well-being and financial stability.
The term "shock" refers to significant negative impacts, often associated with severe events such as the death of adults or the diagnosis of terminal illnesses (Bandara et al., 2015) While these extreme situations are clear examples, even mild illnesses within a family can disrupt daily life, requiring increased care and support for the affected individuals Consequently, children may be compelled to take on additional responsibilities or jobs to help their families cope with the challenges posed by illness.
Numerous studies have examined the effects of health shocks on household outcomes, particularly in low- and middle-income countries (LMICs) Alam and Mahal (2014) conducted a comprehensive review of the empirical literature, highlighting the significant economic impacts that health shocks can have at the household level Their findings indicate that health shocks can profoundly affect households in various ways.
Out-of-pocket (OOP) health payments can lead to catastrophic spending and impoverishment for households, significantly impacting their labor supply and overall income Health shocks, such as the death of a parent or adult family member, illness, or disability, are critical indicators that affect household finances Additionally, changes in self-reported health and specific illnesses like cancer or HIV contribute to non-medical expenditures, further straining family resources.
Empirical studies indicate that health shocks in low- and middle-income countries (LMICs) significantly increase out-of-pocket (OOP) health payments, adversely affecting household consumption and leading to impoverishment, particularly among poor families For these households, the proportion of OOP health expenses relative to total income is considerably higher than that of wealthier families, resulting in more severe consequences Enhanced public health services and insurance programs can alleviate some of these financial burdens for low-income households, potentially reducing OOP health expenditures In Vietnam, research by Van Minh et al (2012) reveals a 2.5% increase in poverty rates when households face health risks, while Wagstaff (2007) notes that adult mortality correlates with a 27% rise in medical expenses in the preceding month.
Health risks can significantly impact family income and labor supply, particularly in cases of adult mortality Illness or death within a household often leads to a decline in productivity, as affected individuals may be unable to work, and other family members may need to take on caregiving roles This disruption typically results in decreased wages, lower farming returns, or diminished business earnings Research, such as Beegle's 2005 study in Tanzania, indicates that families experience a wage reduction of 66-75% for men within six months following the death of an adult aged 15 or older, highlighting the profound economic effects of such losses.
The impact of health crises on labor productivity is significant, as evidenced by a study in Bangladesh where the death of a household member resulted in an 8.63% reduction in hours worked in the past week Similarly, research by Bales (2013) in Vietnam indicates that health shocks, defined as adult members being bedridden for 14 days or more within a year, lead to a 7.7% decrease in annual workdays Furthermore, findings from Wagstaff (2007) reinforce the notion that health-related issues adversely affect workforce participation in Vietnam.
1 households experience the death of working age member in urban areas in two years,lead to drop by
Research by Bales (2013) utilizing VHLSS data reveals that severe illness, adult death, and disability onset significantly impact household welfare in Vietnam, leading to reduced labor supply and non-farm employment income While some studies suggest negative effects of health risks on household income, others find no correlation The outcomes appear to vary based on health shock measurements and the type of labor force involved For instance, Yomano & Yayne (2014) report an insignificant relationship between adult deaths and off-farm income overall, yet identify a significant decline linked to the death of a male household head This inconsistency is also reflected in Khan's (2010) findings.
Other health shock effects are on non-medical consumption Using data from Vietnam, Wagstaff (2007) finds the negative effect on food expenditure However, Bales
A study utilizing VHLSS data in 2013 reveals a decrease in labor supply and non-farm employment income, while non-medical payments remain unaffected by health risks Additionally, it concludes that households with access to credit markets do not fully utilize consumption insurance to manage health shocks, highlighting that household characteristics significantly influence consumption smoothing.
Health shocks can significantly impact households by leading to reduced income, increased out-of-pocket health expenses, and an imbalance in labor supply Additionally, these shocks can contribute to the prevalence of child labor The following section will explore how households cope with health shocks, providing insights into the decision-making processes related to child labor.
Response of Household with Health Shocks
To manage health shocks, individuals often resort to various strategies such as cutting back on both food and non-food expenditures, selling assets or livestock, utilizing savings, and borrowing from formal or informal sources Additionally, intra-household labor substitution plays a significant role in coping with these challenges (Yilma et al., 2014; Alam).
& Mahal, 2014; Mitra et al, 2014; Bonfrer & Wright, 2016) It takes into account that under pressure of health shocks, child labor is also a coping strategy without other mechanisms (Basu & Van, 1998).
Beegle et al (2006) argue that using assets is an important way to cope with negative impacts of transitory income shocks Besides, using data in Vietnam,
Wainwright and Newman (2011) highlight that households utilize liquid asset holdings to mitigate the effects of economic shocks However, when faced with idiosyncratic shocks such as health issues, these households may suffer negative consequences, leading to a depletion of their savings, livestock holdings, and insurance coverage.
According to Baland and Robinson (2000), in a perfect financial market, households have access to credit at competitive interest rates, allowing them to smooth their consumption Gertler et al (2009) found that in Indonesia, households utilize the formal capital market to manage health-related expenses, with those located near commercial banks or microfinance institutions faring better in coping with illness Additionally, research in Bangladesh by Islam supports these findings.
Microcredit packages are effective in alleviating the impact of income fluctuations on household consumption, as noted by Maitra (2012) Government and credit organization support is crucial for households in need of these financial resources In India, Mohanan (2013) found that over 70% of families facing health issues resort to borrowing for medical expenses Additionally, a household's wealth significantly influences its ability to cope with health shocks, with wealthier families more capable of managing medical costs through asset liquidation or using assets as collateral for credit Higher-income households, with stronger financial foundations, are better equipped to handle health-related risks, while lower-income families struggle to afford medical expenses or hire additional help Bandara et al (2015) further emphasize that assets are essential in reducing the impact of household deaths on children's total work hours during both income and non-income shocks.
Households often utilize informal coping mechanisms to manage health risks without resorting to child labor, relying on support from extended family, friends, and neighbors for both financial and practical assistance This can include low or no-interest loans, transfers, and in-kind support such as food and seeds, as well as job introductions (Yilma et al., 2004) Such strategies are particularly prevalent in rural areas, where access to formal resources is limited and strong social networks are common Families unaffected by health shocks can benefit from these supportive relationships.
In countries like Vietnam, particularly in rural areas, community support plays a vital role in assisting individuals facing health fluctuations The close-knit nature of these communities fosters a strong network of help, making it essential for those in need to receive care and assistance from their neighbors and friends.
In Vietnam, households facing health shocks often resort to strategies such as taking loans, selling assets, and cutting education expenses (Mitra et al., 2015) Labor supply adjustments are crucial for managing these health crises, as highlighted by Wagstaff (2007) In rural areas, livestock trading serves as a vital mechanism for stabilizing household consumption Access to microcredit enables families to retain their livestock, preserving their production capabilities and ensuring future income through livestock farming (Islam & Maitra, 2012).
However, some coping strategies might be difficult to apply in reality, poor households are less likely to against health shocks with based on savings or assets (ILO,
In developing countries, access to formal credit remains limited, particularly in rural areas and among impoverished families Many poor households lack the means to implement coping strategies during health crises, such as selling or mortgaging land and assets Health shocks can hinder their ability to repay debts, making it challenging to secure loans from financial institutions Research indicates that adult illness or death further diminishes lenders' confidence in borrowers' repayment capabilities Consequently, households may resort to selling assets to cover medical expenses, leading to a lack of collateral for future loans This often forces them to seek informal credit, which comes with higher interest rates and can result in long-term debt accumulation The burden of medical payments can lead families to cut expenditures, diversify income sources, push children into the workforce, and reduce investments in their children's education.
Another alternative strategy which households apply to deal with health shocks is
Health shocks significantly influence household labor supply, as demonstrated by Bazen and Salmon (2008) through the "added worker effect hypothesis." Their findings reveal that a father's short-term illness or treatment increases child labor, while chronic illness prompts mothers to work more Similar patterns are observed when other family members experience health issues In Indonesia, Gertler and Gruber (2002) note that a health risk faced by the household head leads to increased work hours for other members Kadiyala et al (2009) highlight in their Ethiopian study that Prime Age Adult Mortality (PAM) adversely affects children's welfare due to imbalanced labor supply and expenditure PAM also results in lost working hours for caregivers, diminishing childcare quality and pushing children into various activities.
Health shocks significantly affect households by deteriorating their living standards and disrupting labor supply To mitigate these negative impacts, mechanisms like asset holdings and access to credit can be beneficial Additionally, child labor may serve as a coping strategy during health crises, providing both income and a substitute for adult labor The next section will explore this relationship in greater detail.
Health Shocks and Child Labor
Health shocks are unique disruptions that originate within households, significantly affecting both income and labor availability This article explores the direct and indirect impacts of health shocks on child labor, highlighting four main channels (Dinku, 2017) First, health shocks can lead to an increase in child labor as children often need to care for sick family members Additionally, households may face an imbalance in their labor supply due to health shocks, which can result from the loss of adult workers—who are typically the primary earners—and a decline in overall productivity due to the poor health of remaining laborers.
In some cases, family members may resign from their jobs to provide care for a loved one suffering from illness Additionally, children may need to contribute by taking on responsibilities such as farming, household chores, and gathering firewood.
Health shocks can significantly impact households, leading to increased medical expenses that strain financial resources As a result, parents may require their children to work more frequently, shifting their focus from education and leisure to labor This situation not only reduces household income due to decreased employment wages and farming profits but also forces families to trade assets or take loans, jeopardizing future earnings Consequently, children may face longer work hours, and families may invest less in their education, ultimately diminishing future human capital and increasing children's participation in the labor market.
Child labor often serves as a coping mechanism to replace adult labor in the market, but it is not a perfect substitute due to varying work characteristics and technologies (Basu & Van, 1998; Raijan, 1999) The involvement of children in labor depends on their skills and health, and their work may include domestic tasks such as chores, collecting firewood, fetching water, and caring for younger siblings While child labor is prohibited in certain sectors, it remains prevalent in informal markets like agriculture and business Children can assist adults in various tasks, allowing them to balance work with education and leisure For instance, research by Ray (2000) highlights the complementary roles of mothers and daughters in household chores in Pakistan, while data from Peru shows that higher male wages correlate with reduced working hours for girls Similarly, Diamond and Fayed (1998) found comparable patterns among female adults and children in household settings.
Health shocks significantly impact child labor, as noted by Alam and Mahal (2014) When family members experience mild illnesses, they may continue working, albeit with reduced productivity In contrast, when a family member is bedridden, household activities are disrupted, often requiring caregivers This situation tends to increase children's participation in the labor market more than in cases of mild illness.
The death of a primary wage earner after prolonged medical treatment can significantly impact families, leading to increased financial strain (Salmon, 2008; Dillion, 2012) In such situations, dependent family members, including children, may be forced to enter the labor market prematurely, often sacrificing their education to help alleviate the financial burden caused by lost income and medical expenses.
Despite numerous studies examining the effects of health shocks on household labor supply, there is limited focus on child labor Research in Bangladesh by Bazen and Salmon (2008) reveals that parental health issues lead to an increase in child labor, with father’s illness correlating with overall increased child work and mother’s recent illness also contributing to higher child labor rates Dillon (2012) further supports these findings, noting that illness among siblings can increase a child's agricultural work by approximately 4 hours weekly, while adult female and male illnesses result in children spending an additional 1.6 and 2.6 hours per week, respectively, on caregiving and business tasks In Tanzania, Bandara et al (2014) find that health shocks, including the death of family members, significantly increase total work hours for both male and female children This paper also highlights that health shocks tend to elevate agricultural work hours while reducing indoor work, and it suggests that family assets can mitigate the adverse effects of health shocks.
Households facing income or non-income shocks often resort to employing child labor as a coping mechanism, as noted by Basu & Van (1998), Gertler & Gruber (2000), and Dillon (2012) However, even when families have access to credit or possess assets, child labor can still be utilized to manage these shocks effectively The influence of health shocks on child labor is influenced by the level of asset ownership and credit availability Beegle et al (2003) highlight that access to credit can alleviate the adverse effects of income shocks on child labor, demonstrating this through an interaction of collateral assets and crop loss events to assess the buffering impact of credit.
Research by Bandara et al (2015) indicates that asset holding significantly mitigates the effects of income and health shocks on child labor.
Health shocks significantly impact childhood development and human capital, as evidenced by the prevalence of child labor When families face health risks, their ability to provide essential support for children diminishes For instance, the death of a mother can result in a lack of caregiving, hindering children's growth Additionally, financial strain may reduce investments in education and nutrition, further compromising child development Psychological effects also play a crucial role in this context Research by O'Donnell et al (2005) highlights the long-term consequences of child labor on health outcomes in Vietnam.
Omitted variables bias can arise from the simultaneous relationship between child labor and health shocks caused by adverse events like disease, drought, and floods For instance, during a drought, families may struggle with limited access to fresh water, negatively impacting household health due to dehydration or stress, while also increasing the time children spend collecting water or resources for livestock To address this issue, it is essential to include additional variables related to other shocks (Bandara et al., 2015) Moreover, the death of a household member is considered a permanent shock, with some effects remaining unobserved Unobserved household characteristics may simultaneously influence both family health and child labor Farrell and Fuchs (1982) illustrate that parents with low expectations for future returns on their children's education may also neglect their health, leading to increased health risks Therefore, incorporating parental characteristics as variables can help mitigate the omitted variable bias (Bazen & Salmon, 2008; Bandara et al., 2015; Dinku, 2017).
RESEARCH METHODOLOGY
Research Methodology
In terms of household decisions for child labor, the paper employs a basic model of Basu and Van (1998) and is developed by Kruger et al (2007) and Bandara et al
In 2015, the study explores how parents make household decisions, particularly regarding child labor and education, while aiming to maximize household utility It investigates whether financial constraints, stemming from health shocks, influence the decision to send children to work The analysis begins with a utility function that considers both consumption and the human capital development of children in single-parent households.
Where � �� is denoted as consumption of household i in period t; ℎ �� is the human capital of children; σ is the elasticity of substitution and constant with 0 < σ < 1; α is constant parameter, α > 0.
The study posits that household consumption is derived from income generated by both adult and child labor, with parents fully participating in the workforce According to the established utility function, household consumption (c) will only be positive when family income is also positive.
Where � ��� , � ��� are the time which the child and parent spend to work respectively;
� ��� , � ��� are wages of the child and parents respectively.
This study begins by examining a basic model that assumes households lack both asset holdings and credit access Subsequently, it will sequentially introduce asset holdings and credit access to analyze their effects on the relationship between child labor and health shocks.
(1) Assume that households have neither asset holdings nor access to credit
The paper begins with households with the assumption the absence of asset holdings as well as access to credit Parental income is measured as function of three main
2 factors, including parental wage � ��� � ��� , household characteristics 𝑋 �� , and is influenced by health shocks HS ��
The allocation of a child's time between work and education is crucial for human capital development In this context, the variable representing time spent on schooling is denoted as \( T_s \), while the total available time for children is referred to as \( T_c \) Balancing these elements is essential for optimizing a child's growth and learning opportunities.
The human capital of the child also is identified from a function of technological
𝜎 1 component β and the child's time allocated for schooling: ℎ �� �� ���
In other words, child’s human capital is generated only from the time of study of the child with a technology Besides, child labor contributes to the household income with wage
� ��� Households will make their decisions based on the utility maximization with a consumption constraint condition Accordingly, the utility function is written as below:
Subject to the budget constraint:
Defining λ as the multiplier on the full-income constraint, first-order conditions for � �� and � �� are presented respectively as follows:
Moving on the second first-order condition characteristic, households stand to make decision sending children to work or investing in the child’s education In particular, if
When the marginal value of returns from human capital is lower than that of child labor, parents are more likely to choose sending their children to work instead of enrolling them in school Conversely, if the returns from education surpass those from child labor, parents will prioritize their children's schooling.
�� � � , themarginal value of investment in human capital is higher than the child labor counterpart, it argues that the child will spend
1 The more common function of human capital is present: ℎ �� = �� ��� , where ν should be different with σ in above function This paper assumes ν=σ for easily to calculate in the formulation.
� all his/her time to schooling On the other hand, if
�� � � , parents will be indifferent in decisions between sending the child to work and schooling Meaning children will distribute the time both work and study.
Assuming constant parental and child wages, and with the child wage remaining unchanged, the analysis leads to a derived function that outlines the household decision-making process based on the first-order condition.
Parental income from labor significantly influences children's time allocation for schooling and child labor, as indicated by the equation Additionally, health shocks and household characteristics play a crucial role in determining child labor dynamics.
Higher parental income and increased investment in a child's human capital are anticipated to reduce child labor in households without asset holdings Additionally, health shocks are expected to negatively affect the child's involvement in work.
(2) Assume that households have asset holdings.
The paper will explore how households with asset holdings can potentially reduce child labor Specifically, when families face health risks, they can utilize their assets to meet essential consumption needs, thereby decreasing their reliance on child labor By having these assets, households can adjust their budget constraints to better manage their financial challenges.
The equation \( A_{t+1} = A_t(1 + r) \) illustrates the relationship between household assets \( A_t \) at time \( t \) and the interest rate \( r \) This paper assumes a constant growth rate of household assets, integrating this assumption with the first-order conditions outlined in scenario (1) to derive the relevant function for asset growth.
(9) Where � is a constant element, and expected smaller than 0, meaning the high level of asset holdings can reduce child labor.
(3) Assume that households have both hold assets and access to credit.
The last scenario, the paper will consider a case where households have both hold assets and access credit Hence, the budget constraint can be given as follows:
+ 𝐴𝐶 �� − (1 + �)𝐴𝐶 ��+1 (10) Where r is an interest rate and 𝐴𝐶 �� is credit which household i can access
(borrowing) in time t The solution to the first-order condition to the household decision problem, the function for case (3) is given by:
Where � is a constant element and expected smaller than 0.
In labor supply models that measure labor by work hours, the issue of truncated samples arises, as only individuals actively participating in the labor market are observed, limiting the sample size relative to the full population To address this limitation and improve parameter estimation, employing a sample selection model is essential (Greene, 2012) Specifically, in the context of child labor, observed work hours reflect only those children who are employed, while those not engaged in work are recorded as zero hours Ordinary Least Squares (OLS) estimates can lead to inconsistent and biased results due to unobserved factors included in the error term, particularly since children with zero work hours may not be randomly selected but influenced by intentional decisions or external circumstances (Heckman, 1979; Hine, 2012) Consequently, the Heckman selection model is preferred as it effectively addresses the limitations of OLS in this scenario.
In the field of household decision-making regarding consumption and labor supply, researchers often utilize a two-step process, as highlighted by Mulligan and Rubinstein (2004) and Eckstein & Lifshitz (2011) Hien (2012) applies the Heckman selection model to investigate the interplay between child labor and land size, treating child labor as a two-step process to address datasets with significant "zero" observations The study employs a sample selection model, also known as the Tobit Type II model, to examine the relationship between health and child labor.
2 shocks and child labor The study uses Heckman's selection model is developed by Heckman (1979) to estimate the parameters of the selection model.
This methodology employs a two-step process, beginning with a participation estimate for the probit function through maximum likelihood to determine γ, which represents the observed error term Subsequently, the inverse Mills ratio (�̂ i) is calculated for each observation The second step involves a determinant function that utilizes least squares regression to estimate β and βλ, based on the regressors and the previously estimated �̂ from the first step This detailed selection model effectively integrates these components for robust analysis.
The participation function is evaluated using a probit model that assumes a bivariate normal distribution, focusing on two scenarios: participation and non-participation This analysis is grounded in a set of independent variables.
0 𝑖 � ∗ ≤ 0 The probability of participation, meaning
= 1 with the condition � ′ is equal to Φ(� � �) and equal to 1 - Φ(� � �) otherwise.
The inverse Mills ratio, derived from the estimation of the participation function, highlights the issue of selection bias A statistically significant inverse Mills ratio indicates that the selection model, such as the participation function, is valid, reinforcing the relevance of the two-step process in the analysis.
Where 𝜙(� � �) is the probability density function while Φ(� � �) presents for the cumulative distribution function.
For the second step of the estimate process, the determinant function of outcome
� � is estimated as a regression of observed � � if � � is equal to 1.
� � = � ′ � + � if � = 1Above two steps of process under the assumption that:
Vietnam Young Live dataset Overview
RESULTS AND DISCUSSION
Descriptive statistics
Table 2 reveals the daily time allocation of children in the dataset, highlighting that girls dedicate significantly more time to domestic chores, averaging 0.72 hours per day, compared to boys, who spend around 0.53 hours on similar tasks Additionally, boys engage in both paid and unpaid work slightly more than girls The findings also indicate that child labor is consistently higher in rural areas than in urban settings across all work activities.
Children in rural areas engage in work activities for an average of 1.28 hours daily, which is nearly twice the amount of time spent by their urban counterparts.
Table 2: The child’s work hour following gender and the type of site
Variable Obs Mean Obs Mean Obs Mean Obs Mean
The analysis of child labor in relation to age reveals significant differences in how children allocate their time As illustrated in Figure 1, older children are more likely to participate in work compared to their younger counterparts.
Figure 1: The graphical relation between the child work hour and the child age
Source: Compiled from Young Live Data
Observations Mean Std.Dev Min Max
Table 3 illustrates the correlation between child labor and age, highlighting distinct differences in work hours among various age groups Children over 14 years old work approximately 2.93 hours per day, totaling nearly 20.51 hours per week, indicating a significant level of child labor among older youth.
Table 3: Distribution of child labor following the child age
Range of the child age (the year old)
Sources: Compiled from Young Live Data
In a study of household health shocks, it was found that illness among family members occurred in 25.36% of the 965 observations, highlighting a significant frequency of health-related events In contrast, the occurrence of a household member's death was notably rare, primarily due to missing variable data Overall, the analysis indicates that households frequently experience health challenges, including both illness and death.
In Vietnam, 25.36% of households, or one in every four, have experienced health risk events, highlighting the significant impact of health shocks on the population This statistic underscores the need for a deeper examination of the consequences associated with such health challenges.
Table 4: Description of health shocks
Frequency Percentage compare with the total of observations (excluding missing observations)
Death or Illness of household member 984 25.36
Sources: Compiled from Young Live Data
Table 5 illustrates the correlation between health shocks and child labor, revealing that children from households without health shocks tend to work slightly more hours Specifically, children in households experiencing health shocks average approximately 1.32 hours of work per day, compared to just 1.09 hours for those in unaffected households.
Table 5: The statistical description health shocks and child labor
The household experiences the health shock 984 1.32 2.07 0 18
The household do not experience the health shock 2896 1.09 1.65 0 17
Sources: Compiled from Young Lives Data
This paper defines child labor as any work performed by a child for more than one hour per day, and subsequent sections will discuss this definition Table 6 provides a description of the variables used in the analysis The gender distribution of children in the sample is relatively balanced To account for the long-term effects of past health shocks, the study considers the survival status of both parents, although the incidence of parental death remains low, below 1% for both mothers and fathers in the sample.
In terms of parental education, both mothers and fathers have an average highest grade of around 7, indicating a similar level of educational attainment Notably, over 16.78% of fathers have achieved a highest grade of 12 or above.
A study involving 3,880 individuals reveals that approximately 13.86% have mothers with a low level of education, which is largely attributed to the rural setting of the sample, where nearly 80.1% of participants reside In terms of parental occupations, around 45% of mothers and 51% of fathers are primarily engaged in non-agricultural sectors, while about 35% of mothers and 40% of fathers work in agriculture.
The average value of durable assets held by households is approximately 15.87 million dong, with 54% of households having taken out loans in the past year Furthermore, 18% of households are classified as poor according to the Vietnam MOLISA standards for the Hunger Eradication and Poverty Reduction Program.
Table 6: Description of using variables
Variable Obs Mean Std Dev Min Max
Father Works in Non-agriculture 3880 0.51 0.5 0 1
Mother Works in Non-agriculture 3880 0.45 0.5 0 1
Sources: Compiled from Young Lives Data
The testing the correlation among variables is included in Appendix.
Regression results
The paper utilizes Heckman’s selection model to analyze the relationship between child labor and health shocks through a two-step process The inverse Mills ratio is crucial for assessing the model's suitability, as indicated in Table 7, where the second column details the child labor participation function and the third column provides insights into the degree of child labor The results demonstrate that the inverse Mills ratio supports the validity of the Heckman’s selection model, as the hypothesis of its irrelevance cannot be rejected Additionally, the Wald test confirms the appropriateness of the model.
Table 7: Results of Heckman’s selection model
Father Works in Non-agriculture 0.26*** -0.12
Mother Works in Non-agriculture 0.04 -0.20
Inverse Mills ratio (lambda) -0.89*** Wald chi2(20) = 141.97
Health shocks have a statistically significant positive impact on child labor participation, indicating that when families face health crises, children are more likely to enter the workforce This finding aligns with existing literature, particularly in the context of Bangladesh, where a bivariate probit model analysis shows that a father's illness correlates with increased child labor attendance (Bazen and Salmon, 2008).
Table 8: Marginal effect on the child labor participation function
Father Works in Non-agriculture 39.10% 60.40% 79.10%
Asset Holdings if Health Shocks are not happened 27.90% 47.60% 67.90% Asset Holdings if Health Shocks are happened 24.60% 43.80% 64.60%
Loans if Health Shocks are not happened 27.90% 47.60% 67.90%
Loans if Health Shocks are happened 33.60% 54.10% 73.60%
1 Marginal effect of health shock at median of vnloans = 1 and ln(durable_asset)=8.97
The analysis of the marginal effect of health shocks on child labor participation reveals significant insights, as detailed in Table 8 The marginal effect is assessed at three initial probability levels: 30%, 50%, and 70% At a 30% baseline, health shocks result in a modest increase of 0.4%, raising the probability of child labor participation to 30.4% This finding highlights the mitigating influence of asset holdings, which help lower the likelihood of child labor Specifically, when the parameter for health shocks is set at 0.71, the probability of child labor participation rises to 59.5% from the initial 30% However, when factoring in interaction variables, the overall impact of health shocks diminishes significantly The study underscores the critical role of asset holdings in buffering against the adverse effects of health shocks on child labor.
3 the research of Bandara et al (2015) However, they consider the dependent variable as the work hour of children is not a dummy of the child work participation as this step.
Health shocks do not significantly influence the decision regarding children's work hours, indicating that the extent of child labor is not affected by these health issues Instead, various other factors play a more crucial role in determining how many hours children work While health shocks may affect the initial stages of decision-making, they do not have a direct impact on the overall degree of child labor.
This paper examines the long-term effects of health shocks experienced prior to the interview, specifically focusing on how the presence of parents influences their children's employment outcomes following parental death However, the findings reveal that the impact is not statistically significant.
As children grow older, they are more likely to engage in work, with participation rates increasing by 23.5% from an initial 30% for each additional year of age This indicates that older children have significantly more opportunities for employment Additionally, there is a quadratic relationship between age and work hours, with each year of age resulting in an increase of 2.82 hours of work per day from the median age of 8 This trend aligns with findings from Gebru & Bezu (2014), which highlight that older children are more inclined to participate in domestic tasks such as collecting firewood, water, and fodder Overall, older children face greater pressure from employment compared to their younger counterparts.
Older children often dedicate a significant amount of their time to work instead of engaging in study or leisure activities This shift can adversely affect their educational outcomes and overall health, ultimately impacting the development of their human capital (Beegle et al., 2004).
Research on child labor reveals significant differences in employment participation based on gender Specifically, male children are less likely to engage in the labor force compared to their female counterparts, with a notable negative coefficient of -9.1% in the initial employment probability, which stands at 30% for girls This highlights the gender disparity in child labor dynamics.
Research indicates that girls may engage in less labor compared to boys, particularly in the context of agricultural shocks, as shown in a study by Kruger et al (2007) in Brazil Gebru & Bezu (2014) further highlight that girls participate in collection activities 48% less than boys, suggesting a gender disparity in resource collection These findings imply that gender influences child labor participation differently, with female children more likely to engage in employment, although the link between gender and working hours remains unclear It is also important to note that child labor encompasses a broad range of activities rather than being limited to specific types.
The study utilizes parental education as a proxy for parental income, positing that higher educational attainment among parents correlates with reduced child labor due to the luxury axiom effect (Basu & Van, 1998) The findings indicate that only the father's education significantly influences child labor, highlighting the father's prominent role in Vietnamese households, where he is typically regarded as the head of the family In contrast, maternal education does not show a significant effect on child labor in this context.
Parental employment significantly influences child labor, particularly the father's job status Children are more likely to work when their fathers are employed, whether in agriculture or non-agriculture, compared to when fathers are unemployed or engaged in household chores This suggests that with an employed father, children may take on more domestic responsibilities or assist with work due to increased household demands Notably, the impact of a father working in agriculture is greater, with a marginal effect of 10.2%, compared to 9.1% for those in non-agricultural sectors This aligns with findings by Parikh & Sadoulet (2005), highlighting the relationship between parental employment and child labor participation.
3 between the parental occupation and child labor in Brazil.
Household size significantly influences household characteristics, with larger families typically facing higher expenditures As the number of individuals in a household increases, the demand for both food and non-food items rises, often necessitating that children enter the labor market to contribute financially or assist adults Additionally, larger households often include dependent members like young children and grandparents, requiring more caregiving time Consequently, older children may take on domestic responsibilities while their parents work longer hours outside the home The marginal effect of household size indicates that for every additional person in a household, the likelihood of a child participating in work increases by 2.3%, correlating with an increase of 0.19 hours worked, holding other variables constant.
Poverty significantly influences child labor, with children from poor households more likely to engage in work Research indicates that the probability of child labor participation and the number of hours worked increase notably for children in impoverished families Specifically, children from poor households experience a 5.3% rise in labor participation compared to their peers, translating to an additional 0.43 hours of work per day.
The study investigates the impact of crop shocks on child labor, revealing that households facing such shocks may see a 5.4% increase in child participation, rising from an initial 30% probability However, the analysis indicates that crop shocks do not show a significant relationship in the subsequent phase of the model.
Access to asset holdings and credit is anticipated to decrease child labor, which often serves as a coping mechanism for households facing risks This study employs interaction variables to analyze how asset holdings and credit access buffer against child labor The findings indicate that asset holdings significantly contribute to reducing child labor, aligning with the results of Bandara et al (2015) Specifically, in the participation function, asset holdings lead to a 24.6% reduction in child labor probability when households experience health shocks, compared to a 27.9% probability without such shocks, highlighting the critical role of wealth in mitigating child labor.
CONCLUSION AND POLICY IMPLICATIONS
This study examines the link between child labor and health shocks using the Vietnam Young Lives Dataset from 2006 and 2009, employing Heckman's selection model to address sample truncation and bias issues The analysis reveals that health shocks negatively influence child labor participation, increasing the likelihood of involvement, while not significantly affecting the number of hours worked Additionally, the impact of health shocks on participation is moderated by asset holdings, with the probability of child labor participation rising from 30% to 30.4%.
Asset holdings are crucial for Vietnamese households in managing health shocks, significantly affecting child labor participation rates Specifically, asset holdings lower the likelihood of child labor participation from an initial 30% to 27.9% when health shocks are absent, and to 24.6% when such shocks occur This indicates a buffering effect of assets against health-related challenges, although this effect does not extend to the number of hours children work In fact, if the value of assets doubles while other factors remain constant, the daily work hours for children decrease to 16 hours.
Rural households, especially those in poverty, experience a significant increase in child labor participation and hours worked The father typically holds the primary authority in household decisions, including those about child labor Notably, the Mekong River Delta region exhibits the lowest levels of child labor, particularly when families face health crises In contrast, children in other regions are more likely to participate in child labor compared to those in the Mekong River Delta.
The study suggests that health shocks significantly increase child labor, particularly among children in rural and impoverished households, where low-skill labor and low income are prevalent Additionally, household assets are crucial in mitigating the adverse effects of health shocks on child labor; families with limited assets are more vulnerable Consequently, policies aimed at supporting programs that reduce health shocks and enhance asset holdings in these communities are essential for protecting children from labor exploitation.
Addressing poverty is crucial, as enhancing access to public health services and offering comprehensive health insurance can significantly mitigate the adverse effects of health shocks on households By doing so, families can alleviate the financial strain caused by health crises, enabling them to better safeguard their children from potential risks.
In Vietnam, fathers play a crucial role in family dynamics, and programs aimed at enhancing fathers' understanding of child development can significantly help reduce child labor Older children often feel compelled to work due to societal expectations, which can disrupt their education as they face increasing academic pressures Therefore, it is essential for parents to prioritize a balance between their children's education and work commitments to ensure their overall well-being and development.
This paper acknowledges several limitations, including the dataset's constraints, variable selection, and research scope The absence of continuous variables necessitates the use of a larger dummy variable, and the small percentage of member deaths in the sample leads to a generalized discussion of health shocks, lacking in-depth analysis Therefore, the study could benefit from exploring multiple dimensions of the subjects, particularly by examining specific types of health shocks and their relationship with child labor for a more comprehensive understanding.
Ahmed, I (1999) Getting rid of child labour Economic and Political Weekly, 1815-1822.
Alam, K., & Mahal, A (2014) Economic impacts of health shocks on households in low and middle income countries: a review of the literature Globalization and health, 10(1), 21.
Alam, S A (2015) Parental health shocks, child labor and educational outcomes:
Evidence from Tanzania Journal of health economics, 44, 161-175.
Apps, P F., Kabátek, J., Rees, R., & van Soest, A (2012) Labor supply heterogeneity and demand for child care of mothers with young children.
Bachman, S L (2000) A new economics of child labor: Searching for answers behind the headlines Journal of International Affairs, 545-572.
Baland, J M., & Robinson, J A (2000) Is child labor inefficient? Journal of Political
Bales, S (2013) Impact of health shocks on household welfare in Vietnam—Estimates using fixed effects estimation Institute of Health Policy & Management (HEFPA) Working Paper, 18.
Bandara, A., Dehejia, R., & Lavie-Rouse, S (2015) The impact of income and non- income shocks on child labor: evidence from a panel survey of Tanzania World development, 67,
Basu, K (2006) Labor Laws and Labor Welfare in the Context of the Indian Experience.
In Poverty, Inequality and Development (pp 183-204) Springer US.
Basu, K., & Tzannatos, Z (2003) The Global Child Labor Problem: What do we know and what can we do? The world bank economic review, 17(2), 147-173.
Basu, K., & Van, P H (1998) The economics of child labor American economic review, 412-427.
Bazen, S., & Salmon, C (2010) The impact of parental health on child labor: the case of Bangladesh Economics Bulletin, 30(4), 2549-2557.
Becker, G S (1965) A Theory of the Allocation of Time The economic journal, 493-
517 Beegle, K (2005) Labor effects of adult mortality in Tanzanian households.
Economic Development and Cultural Change, 53(3), 655-683. credit.
Beegle, K., Dehejia, R H., & Gatti, R (2003) Child labor, income shocks, and access to
Beegle, K., Dehejia, R H., & Gatti, R (2006) Child labor and agricultural shocks
Bonfrer, I., & Gustafsson-Wright, E (2016) Health shocks, coping strategies and foregone healthcare among agricultural households in Kenya Global public health, 1-22.
Brown, D K., Deardorff, A V., & Stern, R M (2002) The determinants of child labor:
4 theory and evidence School of Public Policy, University of Michigan.
Cain, M (1982) Perspectives on family and fertility in developing countries Population
Cigno, A., Rosati, F C., & Guarcello, L (2002) Does globalization increase child labor? World Development, 30(9), 1579-1589.
Diamond, C., & Fayed, T (1998) Evidence on substitutability of adult and child labour The Journal of Development Studies, 34(3), 62-70.
Dillon, A (2012) Child labour and schooling responses to production and health shocks in northern Mali Journal of African economies, ejs025.
Dinku, Yonatan M "Health Shocks and Child Time Allocation Decisions of Households: Evidence from Ethiopia."
Dumas, C (2013) Market imperfections and child labor World development, 42, 127-142.
Eckstein, Z., & Lifshitz, O (2011) Dynamic female labor supply Econometrica, 79(6), 1675- 1726.
Edmonds, E V (2007) Child labor Handbook of development economics, 4, 3607-3709
Edmonds, E V., & Pavcnik, N (2005) Child labor in the global economy The Journal of
Edmonds, E., & Pavcnik, N (2002) Does globalization increase child labor? Evidence from Vietnam (No w8760) National Bureau of Economic Research.
Edmonds, E., & Turk, C (2004) Child labor in transition in Vietnam Economic Growth,
Poverty and Household Welfare in Vietnam, Washington DC: World Bank, 505-550.
Emerson, P M., & Knabb, S D (2007) Fiscal policy, expectation traps, and child labor Economic Inquiry, 45(3), 453-469.
Farrell, P., & Fuchs, V R (1982) Schooling and health: the cigarette connection Journal of health economics, 1(3), 217-230.
Gebru, B., & Bezu, S (2014) Environmental resource collection: implications for children's schooling in Tigray, northern Ethiopia Environment and Development Economics, 19(02), 182-200.
Gertler, P., & Gruber, J (2002) Insuring consumption against illness The American
Gertler, P., Levine, D I., & Moretti, E (2009) Do microfinance programs help families insure consumption against illness? Health economics, 18(3), 257-273.
Hien, V M (2012) Investigating the “Wealth Paradox” on Child Labor: A Case Study of rural areas in Vietnam Preliminary version.
International Labour Organization (ILO) (2013) World report on child labour: Economic vulnerability, social protection and the fight against child labour Geneva, Swizerland: Author.
Islam, A., & Maitra, P (2012) Health shocks and consumption smoothing in rural households: Does microcredit have a role to play? Journal of development economics, 97(2), 232-243.
Islam, A., & Maitra, P (2012) Health shocks and consumption smoothing in rural households: Does microcredit have a role to play? Journal of development economics, 97(2), 232- 243.
Jacoby, H G., & Skoufias, E (1997) Risk, financial markets, and human capital in a developing country The Review of Economic Studies, 64(3), 311-335.
Kadiyala, S., Quisumbing, A., Rogers, B., & Webb, P (2009) The impact of prime age adult mortality on child survival and growth in rural Ethiopia World Development, 37(6), 1116- 1128.
Khan, F U (2010) Economic consequences of health shocks and coping strategies:
Evidence from urban poor households in Bangladesh Erasmus University.
Kruger, D I (2007) Coffee production effects on child labor and schooling in rural Brazil Journal of development Economics, 82(2), 448-463.
Levison, D., Anker, R., Ashraf, S., & Barge, S (1998) Is child labor really necessary in Indias carpet industry? [Unpublished] 1994 Presented at the Annual Meeting of the Population Association of America Miama Florida May 5-7 1994.
Mitra, S., Palmer, M., Mont, D., & Groce, N (2015) Can households cope with health shocks in Vietnam? Health economics.
Mohanan, M (2013) Causal effects of health shocks on consumption and debt: Quasi- experimental evidence from bus accident injuries Review of Economics and Statistics, 95(2), 673- 681.
Morduch, J (1999) Between the state and the market: Can informal insurance patch the safety net? The World Bank Research Observer, 14(2), 187-207.
Mulligan, C B., & Rubinstein, Y (2004) The closing of the gender gap as a Roy model illusion (No w10892) National Bureau of Economic Research.
Nakamura, A., & Nakamura, M (1991) Children and female labour supply: A survey of econometric approaches In Female Labour Market Behaviour and Fertility (pp 213-236)
O'Donnell, O., Rosati, F C., & Van Doorslaer, E (2005) Health effects of child work: Evidence from rural Vietnam Journal of Population Economics, 18(3), 437-467.
Parikh, A., & Sadoulet, E (2005) The effect of parents' occupation on child labor and school attendance in Brazil.
Ranjan, P (1999) An economic analysis of child labor Economics letters, 64(1), 99-105.
Ranjan, P (2001) Credit constraints and the phenomenon of child labor Journal of development economics, 64(1), 81-102.
Ray, R (2000) Child labor, child schooling, and their interaction with adult labor: Empirical evidence for Peru and Pakistan The World Bank Economic Review, 14(2), 347-367.
Rosati, F C., & Tzannatos, Z (2000) Child labor in Vietnam The World Bank: mimeo.Shimamura, Y., & Lastarria-Cornhiel, S (2010) Credit program participation and child schooling in rural Malawi World Development, 38(4), 567-580.
Strauss, J., & Thomas, D (1995) Human resources: Empirical modeling of household and family decisions Handbook of development economics, 3, 1883-2023.
The Minister of Labor, War Invalids and Social Affairs (2013) Circular No 10/2013/TT-
BLDTBXH T he lists of jobs and workplaces in which the employment of minor persons is prohibited.
Van Minh, H., Phuong, N T K., Saksena, P., James, C D., & Xu, K (2013) Financial burden of household out-of pocket health expenditure in Viet Nam: findings from the National Living Standard Survey 2002–2010 Social science & medicine, 96, 258-263.
Viet Nam’s Institute of Labour, ILO (2012) The Viet Nam National Child Labour Survey
Vietnam’s Prime Minister (2016) Decision No 1023/QD-TT Approval of Prevention
Program to Minimize Child Labor During the Period 2016 – 2020.
Wagstaff, A (2007) The economic consequences of health shocks: evidence from Vietnam Journal of health economics, 26(1), 82-100.
Wainwright, F., & Newman, C (2011) Income shocks and household risk-coping strategies: evidence from rural Vietnam Institute for International Integration Studies Discussion paper, (358).
Yamano, T., & Jayne, T S (2004) Measuring the impacts of working-age adult mortality on small-scale farm households in Kenya World Development, 32(1), 91-119.
Yilma, Z., Mebratie, A., Sparrow, R., Abebaw, D., Dekker, M., Alemu, G., & Bedi, A S.
(2014) Coping with shocks in rural Ethiopia Journal of Development Studies, 50(7), 1009-1024.
Figure A1: Distribution of children from 5-17 years old participating in Economic Activities (EA) in Vietnam (2012)
Source: Viet Nam National Child Labour Survey (2012)