INTRODUCTION
Problems Statement
Child poverty approach has become an area of focus for last two decades resulting from widely acknowledgement of child focused perspective in poverty eradication (Gordon, 2003; Minujin, Delamonica, González, & Davidziuk, 2005)
Children are more vulnerable to poverty than adults due to their reliance on resources from parents and communities (White, Leavy, & Masters, 2003) Growing up in poverty can lead to adverse conditions in adulthood (Corak, 2006a) Additionally, children's basic needs, particularly for nutrition and education, differ significantly from those of adults, and neglecting these needs during childhood can have lasting negative effects on their future outcomes (Duncan).
Academic and policy makers strive to address the needs of children in poverty by utilizing empirical evidence of their experiences A multidimensional approach is essential for identifying indicators that effectively measure child poverty.
Despite Vietnam's rapid economic growth over the past decade, child poverty rates remain alarmingly high, with approximately one-third of Vietnamese children—around 7 million—classified as multidimensionally poor (Hinsdale et al., 2013) This significant deprivation highlights the urgent need for further research into children's basic needs to inform effective policy interventions However, comprehensive analyses of child poverty and its underlying causes are lacking, creating a knowledge gap that hinders targeted policy implications for children Numerous studies have explored child poverty and deprivation through a multidimensional lens in Vietnam (Roelen, 2013; Roelen & Gassmann, 2012; Roelen, Gassmann).
Previous studies, such as those by de Neubourg (2010) and UNICEF (2008), primarily concentrated on the methodologies for assessing child poverty and deprivation using a multidimensional approach However, they did not explore the specific characteristics of children who experience poverty.
There are many debates in previous researches on child poverty indicators for measurement in both monetary and multidimensional approach (Gordon, 2003)
Selecting the right methodology to assess child poverty and identify its influencing factors can lead to impactful policy decisions and interventions, enabling a more accurate understanding of specific poverty conditions.
This study aims to explore child poverty in Vietnam by employing a multidimensional approach to identify key dimensions and indicators It highlights the significant influence of parents' or caregivers' well-being on children's living conditions and seeks to understand the economic and demographic factors affecting household poverty Utilizing data from the Young Lives Survey, the research will estimate the likelihood of children under 15 experiencing poverty and identify factors that may either mitigate or exacerbate this issue.
Research Objective
The two research objectives are specified as follows:
(i) Identify the dimensions, indicators of child poverty in Vietnam through multidimensional approach
(ii) Identify the determinants of child poverty
Main Research Question
(i) What are the extents and characteristics of childhood poverty in Vietnam through multidimensional approach?
(ii) What factors determine the probability of a child being poor?
Research Scope
This study analyzes under-fifteen children using the Young Lives dataset, which categorizes them into two groups: the younger cohort (eight years old) and the older cohort (fifteen years old) The research employs selected domains and indicators of deprivation as proxies to assess multidimensional poverty The data utilized in this study is sourced from the third round of the Young Lives data collected in 2009.
Research Methodology
This study utilizes the Alkire-Foster methodology to assess child poverty in Vietnam, addressing the initial research question Additionally, a logistic regression model is employed to identify key factors that will inform the development of effective intervention policies aimed at reducing child poverty.
Thesis Structure
The thesis is structured into four chapters, following the introduction chapter; the rest content is outlined as below:
Chapter 2: Literature Review This chapter will illustrate the literature of child poverty as multidimensional, starting with definition of child poverty and multidimensional approach Theoretical studies as well as empirical results of child poverty will be reviewed then Finally, the profile of child poverty is demonstrated
Chapter 3: Data and Methodology This chapter analyses the characteristics of the data used and Alkire-Forster methodology adapted to measure child poverty in Vietnam Arguments as well as grounds for selecting domains and indicators will be reviewed then In this chapter, weight and poverty cut off follow Alkire-Foster are also outlined The analytical framework and logistic regression will be further discussed to identify determinants of child poverty
Chapter 4: Empirical Analysis of Child Poverty in Vietnam This chapter starts with the analysis of the percentage of indicators of overall deprivation Next, the Multidimensional Poverty Index, logistic regression model and their interpretation are emphasized In this chapter, a summary of findings as well as recommendation of relevant policies to reduce child poverty bases on statistical results are also presented
Chapter 5: The overall conclusion is reviewed in this section Based on the result estimation, key lessons and policy options are recommended for development and application of multidimensional approach for children, followed by the limitations of the study and suggestions for future research.
LITERATURE REVIEW
Definition
In 2006, the United Nations General Assembly (UNGA) established an international definition of child poverty, stating that children in poverty face a lack of essential material, spiritual, and emotional resources necessary for survival, development, and well-being This deprivation hinders their ability to enjoy their rights, reach their full potential, and engage as equal members of society.
The definition of child poverty, as outlined by the UNGA in 2006, emphasizes the importance of a basic rights and needs approach to effectively address the multidimensional aspects of poverty that children face This international agreement marks a significant advancement for researchers, organizations, and policymakers dedicated to improving child well-being and alleviating child poverty.
Until 1999, child poverty received insufficient attention in development studies, primarily focusing on adult and household poverty without available data or research on children In that year, a UN Expert group, including researchers like Peter Townsend and Alberto Minujin, highlighted child poverty using data on impoverished children and a monetary approach By the end of 1999, UNICEF experts stressed the importance of addressing child poverty to combat overall poverty, leading to a report published by the World Bank In 2003, the first global estimates of child poverty were established by Professor Peter Townsend and David Gordon, with the findings later presented in the State of the World’s Children 2005 report by UNICEF.
In 2004, it was reported that over one billion children were lacking in one or more basic needs, leading to a global recognition of the urgency to address child poverty Subsequently, child poverty emerged as a focal point on the international agenda, resulting in a notable increase in research within this area in the following years.
In short, under this human-rights approach, children experience poverty through various dimensions and this issue should be considered as a multi-faced phenomenon
As such, selected dimensions and indicators in child poverty should reflect basic needs for children in their daily lives
A considerable number of approaches have been applied in literature for child poverty measurement; however they could be categorized into two major groups: uni- multidimensional and multidimensional approach
The monetary approach is a widely accepted traditional method for measuring poverty, classified within the one-dimensional approach group This method utilizes an income-based poverty line to distinguish between the poor and non-poor populations.
Child poverty is a complex issue that has faced criticism for approaches that fail to address its multifaceted nature and the interactions between various factors (Sen, 1985) According to Sen (1985), effectively addressing child poverty requires a focus beyond merely increasing income and utility; it must also encompass the enhancement of children's capabilities.
The capability approach, developed by Sen (1985) and Corak (2006b), emphasizes child health as a fundamental aspect of child well-being In contrast, the Bristol approach, introduced by Gordon (2003) and endorsed by UNICEF, is rooted in human rights and aligns with the Convention on the Rights of the Child (CRC) (UNICEF, 1989) This methodology serves as a key measurement for the Global Study on Child Poverty and Disparities, facilitating comparisons of child poverty across countries and UNICEF regions It focuses on essential services and standards for children's living conditions and societal participation Similarly, the Young Lives approach aims to explore aspects of child deprivation but does not establish specific dimensions or indicators for measuring child poverty, instead seeking to analyze the underlying causes and consequences of child poverty.
This study adopts the United Nations Children’s Fund (UNICEF) definition for assessing child poverty in Vietnam, emphasizing that evaluations should not merely focus on income levels Instead, they must also consider access to essential social services, including nutrition, water, sanitation, shelter, education, and information (UNICEF, 2007a, p 1).
Dimensions and Indicators in Multidimensional Child Poverty
The selection of dimensions and indicators is crucial for accurately measuring multidimensional poverty Therefore, justifying the choice of these dimensions is essential for developing a meaningful and reliable poverty index.
Data availability is a significant challenge in selecting dimensions and indicators for a multidimensional approach to poverty However, numerous studies suggest that theoretical resources and universal standards can guide this selection process Sen (2004) emphasizes that the choice of dimensions and indicators should be based on public scrutiny and discussion He outlines key criteria for selection, highlighting that the primary focus should be the purpose of the valuation, such as targeting, monitoring, and measuring quality of life Additionally, he identifies two other important criteria: dimensions should reflect shared values and priorities of relevant groups and should be influenced by social factors.
Alkire (2008) expanded on Sen's argument by identifying five key reasons for justifying dimensions: theoretical foundations, public consensus, a continuous deliberative participatory process, data assessment, and empirical evidence reflecting people's values or expert opinions.
This view was also supported by Biggeri and Mehrotra (2011) who argued that dimensions and indicators should be conceptualized, justified by public consensus and selected through different priorities
The Bristol approach, adopted by Global Child Poverty (UNICEF), aims to measure child poverty through internationally agreed dimensions and indicators, emphasizing the importance of the Children Convention Rights and the Millennium Development Goals (MDGs) as standard measurements Gordon (2003) highlighted that these dimensions should not only be universally justified but also child-focused, addressing issues directly related to children His research identified levels of deprivation across various essential areas, including shelter, sanitation, water, information, food, education, and health, based on children's rights from the CRC In collaboration with Dr Nandy Shailen, Gordon elaborated on the rationale behind the chosen dimensions and indicators in the Bristol approach, while also critiquing commonly used measurements like the World Bank’s “US $1/day” indicator, the Asset Ownership-based Wealth Index, and the Multidimensional Poverty Index developed by Alkire and Foster (2011a).
Gordon (2003) and major studies from Bristol primarily analyze child poverty at the international level, but national measurements require context-specific dimensions, indicators, and cutoffs for effective policy intervention (Roelen et al., 2009) Alkire and Roche (2012) proposed measuring the depth, intensity, and composition of multidimensional poverty at the country level Their study on under-five child poverty in Bangladesh utilized the Alkire-Foster methodology to develop indicators based on the Bristol Approach, highlighting how thresholds and cutoffs reflect changes at both national and sub-national levels However, the study lacks a clear rationale for the selection of dimensions and indicators, merely adopting previously established deprivation thresholds, including nutrition, water, sanitation, health, shelter, and information Despite this limitation, the methodology offers valuable insights for future multidimensional child poverty research Additionally, Minujin and Delamonica (2005) conducted a detailed analysis of child poverty measurement in Tanzania, exploring regional disparities using deprivation frameworks developed by Gordon and Townsend.
The study emphasizes the importance of a multidimensional approach to effectively measure child poverty and develop targeted intervention strategies tailored to each region It incorporates a temporal analysis to assess changes in child poverty conditions, aiding in program design and targeting Additionally, the research explores the depth and sensitivity of child poverty by adjusting the thresholds of various dimensions, highlighting the necessity for flexibility in adopting these thresholds based on the unique circumstances of each country.
Table 1: Selected indicators and deprivation threshold (Alkire & Roche, 2012) Dimension Deprivation Thresholds
Children who fall more than two standard deviations below the international reference population for stunting (height for age), wasting (weight for height), or underweight (weight for age) are considered malnourished This classification is based on the algorithms established by the WHO Child Growth Reference Study (WHO, 2006).
Water Children using water from an improved source such as open wells, open springs or surface water
Sanitation Children using unimproved sanitation facilities such as pit latrine without slab, open for pit latrine, bucket toilet and hanging toilet
Health Children who have not been immunized by 2 years of age
Shelter Children living in a house with no flooring (i.e a mud or dung floor) or inadequate roofing
Information Children who have no access to a radio or television
Empirical studies on child poverty in Vietnam have increasingly adopted a multidimensional methodology Roelen et al (2009) emphasized the significance of selecting appropriate dimensions and indicators for measuring child poverty, proposing a structured process to ensure clarity in analysis Their application in Vietnam yielded consistent child poverty estimates aligned with previous research Additionally, Roelen and Gassmann (2012) identified suitable indicators and thresholds for Vietnam, revealing disparities not only between rural and urban areas but also among ethnic groups, which is crucial for targeted policy-making A 2008 report by UNICEF and the Ministry of Labor, Invalids and Social Affairs (MOLISA) introduced a new multidimensional approach, utilizing VHLSS and MICS datasets, and developed relevant indicators through expert workshops This approach highlighted the unique aspects of child poverty in Vietnam, focusing on seven dimensions: education, health, shelter, water and sanitation, child work, leisure, and social inclusion and protection The study's methodology for selecting domains and indicators, informed by expert opinions and public consensus, established a robust framework for measuring child poverty, particularly emphasizing the importance of water and sanitation.
Child poverty Profile
The multidimensional approach to measuring child poverty has significantly advanced our understanding of the various aspects of children's experiences in poverty However, there is growing academic concern regarding the identification of risk factors that contribute to the deprivation of these dimensions among children.
Developing a child poverty profile is essential, as highlighted by previous research Notten, Neubourg, Makosso, and Mpoue (2012) emphasized the importance of creating such a profile in Congo Brazzaville to identify the most vulnerable groups and pinpoint areas for effective policy intervention Their study focused on examining the poverty headcount rather than measuring overlapping deprivations.
Children aged 6 to 11 are the most vulnerable to deprivation, particularly concerning enrollment-based indicators Dercon (2012) highlighted that child deprivation is influenced by a complex interplay of factors, including individual, household, and community elements Therefore, it is crucial to consider these various influences to avoid oversimplifying the interpretation of child poverty and deprivation dimensions.
The Young Lives project is a longitudinal study that analyzes the causes and consequences of child poverty, highlighting various risk factors that impact child welfare outcomes (Boyden, 2005) This data is essential for understanding the complexities of child poverty and its effects on children's lives.
Despite the limited number of empirical studies focusing on this concept in specific countries or regions, Adetola and Olufemi (2012) utilized the Alkire Foster methodology to assess multidimensional child poverty in rural Nigeria They subsequently employed the resulting Multidimensional Poverty Index (MPI) as a benchmark to distinguish between poor and non-poor children.
The study examined dimensions and indicators such as Safe Drinking Water, Sanitation, Housing, Health, and Nutrition to identify determinants of multidimensional poverty Using logistic regression, the analysis revealed that health and sanitation significantly impact the overall multidimensional index Additionally, the findings indicated that factors like parents' education, employment in the service sector, and the availability of health facilities decrease the likelihood of multidimensional child poverty.
A longitudinal study conducted by (2005) examined children's poverty in Vietnam, Ethiopia, Peru, and India using Young Lives data The key finding reveals that economic growth alone does not fully address child poverty; instead, it can exacerbate inequalities among children in certain situations Furthermore, the study highlights that childhood deprivation has long-term effects on individuals as they mature, underscoring the need for policy implications aimed at mitigating these impacts on children.
Research indicates that micro-level determinants, such as child, household, and community characteristics, play a crucial role in understanding the factors that influence children's poverty status Consequently, empirical findings suggest that targeted policy implications can be more effective in alleviating child poverty.
Conceptual framework
The conceptual framework is developed based on child poverty dimension from the study of UNICEF (2008) apply for Vietnam and relevant characteristics literature, which is demonstrated as follows:
This chapter reviews the definition of child poverty and its multidimensional measurement, focusing on six key dimensions: enrollment, health, shelter, water and sanitation, child work, and leisure The multidimensional poverty index is influenced by child, household, and community characteristics Subsequent chapters will delve into the methodology for measuring child poverty through a multidimensional lens.
DATA
The Young Lives Round 3 (2009) dataset provides comprehensive data on child poverty across four countries: Ethiopia, India, Peru, and Vietnam Established in 2001, this longitudinal study tracks the development of 12,000 children and their families over five rounds, each lasting three years In Vietnam, data was collected from five provinces—Lao Cai, Hung Yen, Da Nang, Phu Yen, and Ben Tre—between September 2009 and January 2010, involving 2,939 children and households The quantitative data covers various aspects, including household characteristics, infant health, nutrition, education, child labor, and social capital However, the relatively small sample size limits its effectiveness for monitoring child poverty at the national level The complete dataset is available for download from the UK Data Service.
Missing values often pose challenges in survey data analysis When information is absent for certain indicators or variables, the corresponding households are excluded from the calculations Consequently, the final count of observations used for computation is 2,898.
METHODOLOGY
To address the first research question regarding the measurement of multidimensional child poverty, this study will utilize the Alkire & Foster method (2011b), which incorporates headcount and dual cutoff measurements We will assume a matrix \( D = [d_{ij}] \) representing the achievements of \( n \) children across \( d \) dimensions, where \( d_{ij} > 0 \) indicates the \( i \)-th child's achievement in the \( j \)-th dimension A vector \( z_j = z(1, 2, 3, \ldots, n) \) will be established to set deprivation cutoffs for each indicator, determining if a child is deprived in that dimension Specifically, if a child's achievement falls below the cutoff, they are classified as deprived From the achievement matrix \( D \), we will derive a deprivation matrix \( g_o = [g_{oij}] \), where \( g_{oij} = 1 \) if \( d_i < z_j \) and \( g_{oij} = 0 \) otherwise This allows us to construct a column vector \( c \) of deprivation counts, where each entry \( c_i = [g_{oij}] \) represents the deprivation status of the \( i \)-th child.
The "intermediate" approach defines an individual as poor if they experience deprivations in at least \( k \) dimensions, where \( k \) is a cutoff value between 1 and \( d \) This method identifies children as living in absolute poverty when they face two or more deprivations (with \( k=2 \)) and in severe poverty if they experience at least one deprivation (with \( k=1 \)) The next step involves calculating the proportion of children classified as multidimensionally poor by determining the headcount ratio.
Where q=q (y;z) is the number of children in the set E k , as identified using the Z k dual cutoff method Average shared deprivation is found by following formula:
The adjusted headcount ratio M o (y,z) is then measured by:
H represents the percentage of children living in poverty, highlighting the incidence of multidimensional poverty Meanwhile, A indicates the average proportion of weighted deprivations that children experience simultaneously, reflecting the intensity of multidimensional child poverty.
The Multidimensional Poverty Index (MPI) can be analyzed by breaking it down into contributions from specific sub-groups within the population (Alkire & Foster, 2011b) For instance, consider two sub-groups, represented by distributions x and y, which together form a matrix of overall population achievement Let n(x) denote the number of children in sub-group 1, n(y) represent the number of children in sub-group 2, and n(z) equal the total number of children, calculated as n(x) + n(y).
The population subgroup decomposability can be calculated follow this formula:
Then one can calculate the contribution of each group to overall poverty:
This decomposition method is effective in breaking down the overall multidimensional poverty measure and reveals the contribution of each indicator or sub group to MPI
2.2 Rationale of dimensions and indicators selection
The article discusses six dimensions and sixteen indicators related to children's well-being, based on available data and previous studies The selected dimensions include education, health, shelter, water and sanitation, child work, and leisure Four outcome dimensions reflect children's human capital, specifically school attendance, nutrition, leisure, and child work The remaining two dimensions represent environmental conditions and living standards, encompassing factors such as electricity, roofing, flooring, water, cooking fuel, and toilet facilities Justifications for each dimension and indicator, as well as the rationale for including or excluding certain indicators, will be elaborated in the subsequent sections.
Education plays a crucial role in reducing poverty, as recognized globally, including by the United Nations CRC and the Millennium Development Goals, which highlight it as a fundamental human right essential for children's wellbeing In Vietnam, the Law on the Protection, Care and Education emphasizes that all children, regardless of their background, have the right to access education, ensuring equal opportunities for those from disadvantaged environments Consequently, enrollment rates and the age at which children begin formal schooling are key indicators of educational deprivation.
Enrollment rate is a key indicator influencing children's educational outcomes, frequently highlighted in child poverty research and international agreements For instance, Gordon (2003) explored varying levels of educational deprivation among children in developing countries, emphasizing primary enrollment rates as crucial metrics in the Millennium Development Goals (MDGs) 2 and 3 Additionally, a comprehensive report by MOLISA and UNICEF focused on Vietnam's educational context, utilizing enrollment rates to assess improvements in school attendance from preschool through secondary education This study leverages Young Lives data to evaluate educational deprivation, defining a child as deprived if they are not currently attending school.
In addition to enrollment rates, the age at which children begin formal schooling significantly impacts their academic performance (Angrist & Krueger, 1992) In Vietnam, the Law of Education mandates that children must start primary school at the age of six, making those who begin schooling later at a disadvantage in this regard.
Malnutrition in early childhood significantly weakens resistance to infections and increases mortality and morbidity, closely linking it to child poverty (Gordon, 2003) It severely impacts childhood development (Grantham-McGregor et al., 2007) and contributes to poor academic performance, which can result in low productivity in adulthood and perpetuate poverty in future generations The World Bank has identified poor nutrition as a primary factor in perpetuating poverty among children.
The Convention on the Rights of the Child (CRC) emphasizes the importance of children's health as a fundamental right, ensuring that they have access to adequate health and healthcare services (UNICEF, 1989) In Vietnam, the Law on the Protection, Care, and Education of Children reinforces this principle by highlighting the necessity for children to receive appropriate preventive and healthcare from their parents or caregivers (UNICEF, 2008).
Mortality, morbidity, and anthropometry are key indicators of children's health While mortality is included in the MDGs to assess child health, it is unobservable in the Young Lives dataset due to the focus on children as the unit of analysis Morbidity, particularly through immunization, serves as a valuable measure of child poverty in health, as highlighted by Gordon (2003), who emphasized the cost-effectiveness of immunization against childhood diseases Additionally, Alkire and Roche (2012) utilized the immunization indicator in their analysis of child poverty in Bangladesh, although they did not clarify their rationale for selecting this dimension This immunization indicator has also been employed in various studies to assess child deprivation (Qi & Wu, 2014).
Roelen & Gassmann, 2012; Trani, Biggeri, & Mauro, 2013; UNICEF, 2008)
The Young Lives data does not include an immunization indicator, necessitating the development of an alternative measure Consequently, stunting, assessed through the Body Mass Index (BMI-z score), has been chosen as the substitute indicator This measure is strongly endorsed for evaluating the nutritional status of children, as highlighted by Behrman, Glewwe, and Miguel.
2007) That is, under-nourished children are highly in risk of infection and diseases than other in normal development and maybe lead to the morbidity and mortality
The health of children is significantly influenced by the quality of healthcare services available in their communities (Akin, Griffin, Guilkey, & Popkin, 1986) Certain childhood illnesses necessitate prompt and careful attention from hospitals or healthcare centers, as any delays or inadequate care can have serious repercussions on a child's health Importantly, the decision to seek appropriate treatment is closely linked to the economic status of the parent or caregiver Consequently, a child is considered deprived if they do not receive medical attention when ill due to these economic constraints.
Since poor children have difficulties in going to health facility for ill treatment, health insurance is an effective tool for providing health care to children (Nguyen,
In Vietnam, the government has provided health insurance since 1992, ensuring that children under 6 years old receive free coverage For children aged 6 and older, there are two health insurance options available: school health insurance and free health insurance for those from low-income families A child is considered deprived if they lack health insurance.
Food security significantly impacts children's health, particularly regarding their daily food intake Adequate meals are essential, as they have been recognized in previous research as a social norm for assessing child poverty (Pantazis, Townsend, & ).
Gordon, 2000; Qi & Wu, 2014) Under this indicator, the child is deprived if he or she does not take enough food frequently in the last 12 months
Summary
This chapter provides a detailed overview of the data and methodology used in the study A sample of 2,898 observations from the Young Lives dataset was utilized The research methodology for measuring the Multidimensional Poverty Index (MPI) is based on the recent approach developed by Alkire and Foster Six dimensions, represented by sixteen indicators, were selected through a comprehensive literature review, previous studies, public consensus, international agreements, and available data An equal weighting system was employed for estimation, and logistic regression was applied to identify the impact of child, household, and community characteristics on child poverty.
EMPIRICAL ANALYSIS
Overview Child Poverty in Vietnam
The advancement of economic policies in Vietnam has significantly contributed to reducing child poverty, benefiting approximately 26 million children Recent statistics indicate that their living conditions have improved markedly over the past two decades, as evidenced by both monetary and multidimensional measures.
The monetary approach indicates that the VHLSS 1988 reported approximately 47.2% of children living below the GSO-WB poverty line By 2010, this rate remained significantly high at 29.2% (Hinsdale et al., 2013) Nonetheless, the issue of child poverty was still more widespread than the statistics suggested (UNICEF, 2009).
Child poverty measurement has limitations, as it is often defined by households living in poor conditions This approach fails to address the unique basic needs of children, which differ significantly from those of adults, and does not accurately reflect the deprivation that children experience (UNICEF).
In 2008, with UNICEF's support, Vietnam implemented a new multidimensional approach to assess child poverty, establishing specific indicators for child health, nutrition, shelter, social inclusion, and protection Utilizing the VHLSS and MICS datasets, it was found that approximately 30% of children, or around seven million, live in multidimensional poverty Key areas requiring attention include nutrition, water and sanitation, leisure, and health, with one-third of children experiencing stunting and half lacking access to hygienic sanitation facilities The findings also highlighted a significant urban-rural divide in both monetary and multidimensional poverty measurements, with ethnic children facing higher risks—61% and 62% poverty rates compared to 13% and 22% for Kinh children in both assessments.
Figure 2: Monetary and multidimensional child poverty in Vietnam 2008 divided by Urban, Rural and Ethnics
The report highlights significant regional disparities in poverty rates, with the Mekong River delta exhibiting the highest levels of multidimensional poverty at 53% and monetary poverty at 16% Following this, the central highlands and northern midlands rank second and third in terms of child poverty rates.
Figure 3: Monetary and Multidimensional child poverty in Vietnam, 2008 divided by Regions
Indicators Deprivation
Figure 4 highlights the percentage of children facing deprivation in various indicators according to Young Lives round 3 The findings reveal that child labor, toilet facilities, and cooking fuel are the most critical areas of deprivation, with nearly 59% of children engaged in work to support themselves Following closely are the deprivation rates for toilet facilities and cooking fuel, at 55% and 54%, respectively Other significant indicators, albeit with lower deprivation rates, include water access, walling, and health insurance, which rank fourth, fifth, and sixth.
The leisure and enrollment hold the lowest percentage at almost 0% These comparisons indicate that child work, toilet facility and cooking fuel are indicators that need important attention
Figure 4: Proportion of children deprived in each indicators
Figure 5 illustrates the percentage of boys and girls among children affected by various deprivation indicators Notably, the percentage of girls facing deprivation in enrollment rates and access to health facilities is significantly higher than that of boys, with a difference of approximately 25 percentage points Conversely, the nutrition indicator reveals a higher deprivation rate for boys compared to girls The remaining indicators exhibit minimal disparity between the two genders.
Figure 5: Indicators deprivation by gender based on Young Lives round 3
Figure 5 illustrates significant disparities across various regions, with notable differences among indicators The Mekong River Delta and Central Area exhibit the most severe deprivation indicators In the Mekong River Delta, the highest rates of deprivation are observed in flooring, walling, and water access Meanwhile, in the Central Area, approximately 60% of deprivation rates are attributed to the indicators for starting formal education and leisure activities.
Figure 5: Indicators deprivation by region based on Young Lives round 3
Figure 6 illustrates the percentage of children experiencing deprivation across various indicators Notably, a small fraction of children are not deprived in any indicators, while 18% face deprivation in three indicators, marking the highest rate Additionally, deprivation in two and four indicators ranks as the second and third most common, respectively The data highlights that the largest proportions of children are deprived in three and four indicators.
Figure 6: Proportion of children deprived in various numbers of indicators
Poverty cut off and MPI estimation
The selection of the poverty cutoff in the AF methodology remains a topic of debate, as it significantly impacts the identification of multidimensional poverty among children A child may experience deprivation in several indicators, but if their deprivation cutoff is below the established poverty cutoff level, they may not be classified as multidimensionally poor According to Alkire and Foster (2011a), determining the optimal poverty cutoff "k" is influenced by the subjective perspectives of researchers and policymakers, as well as the social context and study objectives Setting a cutoff that is too low may categorize a large number of children as poor, while a cutoff that is too high could lead to an unreliable multidimensional poverty index, resulting in misleading policy implications Therefore, it is crucial to establish an appropriate poverty cutoff tailored to the specific context and goals of the study.
Table 4: Multidimensional child poverty estimate on various cut off point
Cut off point Head Count Ratio
Adjusted Head Count Ratio (Mo=H*A)
Table 4 presents the estimated poverty index based on the poverty cutoff value The findings indicate that a cutoff point of 0.3 is significant, as it defines children as multidimensionally poor when they experience deprivation in at least 30 percent of the weighted indicators This threshold aligns with the headcount ratio reported by MOLISA and UNICEF, which shows 37% using MICS and 31% using VHLSS in their research on multidimensional child poverty in Vietnam.
In a study by Alkire and Roche (2012), it was found that the average deprivation among children is 3.67 for k=3, indicating that those identified as poor experience an average deprivation of 36 percent based on a weighted sum of indicators The potential deprivation that a poor individual may face, relative to all possible deprivations, is 11 percent Additionally, the analysis reveals a negative correlation between cut-off points and both the headcount ratio and adjusted headcount ratio; as the cut-off point decreases, these ratios increase Conversely, average poverty exhibits a positive relationship with higher cut-off points, suggesting that lower cut-off points lead to a reduction in the average deprivation experienced by poor children.
Censored Headcount and Related contributions to MPI
In this section, the related contributions of the various indicators and regions to overall multidimensional poverty are illustrated in Table 5 and Table 6
The analysis from Table 5 reveals that the child work indicator has the highest contribution to poverty at 91% with a cutoff point of 0.3, followed closely by toilet facilities at 90% In contrast, cooking fuel has the lowest contribution at 83% Other indicators, such as walling, water, and health insurance, also require significant attention This suggests that policy efforts should focus on improving shelter dimensions, particularly toilet facilities and cooking fuel, to effectively reduce child poverty Additionally, the health insurance indicator is crucial, as it accounts for half of the overall multidimensional poverty.
Table 5: Contribution of indicators to MPI
Table 6 illustrates the regional decomposition of the Multidimensional Poverty Index (MPI) using a cut-off point of k=0.3, indicating 30 percent deprivation in the weighted sum of indicators The Northern uplands contribute the most significantly to poverty, with an index of 25%, followed by the Mekong River Delta at 13% In contrast, the Central coastal region exhibits the lowest poverty incidence, with only 6 percent of children classified as multidimensionally poor.
The adjusted headcount ratio, which measures the depth of poverty, shows similar rankings to the headcount ratio Among impoverished children, the Red River Delta and Central Coastal regions exhibit the same minimal level of deprivation On average, each multidimensionally poor child in these areas experiences a deprivation of 35 percent based on the weighted sum of indicators.
Table 6: Decomposition of Multidimensional Poverty indices by region
Determinants of Child Poverty
This section examines the individual, household, and community characteristics that influence the risk of children experiencing multidimensional poverty The table below displays descriptive statistics for the variables chosen to identify the key factors affecting multidimensional child poverty.
Table 7: Descriptive statistics of Child Poverty determinants variables
Determinants of multidimensional child poverty
X 1 dum_gender, male , dummy 1476 dum_gender, female , dummy 1422
X 6 m_edu1, Primary or less, dummy 892
X 16 hhincome (Logarithm of consumption per capital) 2898
X 18 region2, Red River Delta, dummy 646
X 20 region4, Mekong River Delta, dummy 614
Table 8 presents the logistic regression estimates identifying the determinants of child poverty, with a poverty cutoff point of 0.3 used to establish the Multidimensional Poverty Index (MPI) as the threshold for classifying children into poor and non-poor categories.
Table 8: Logistic Regression estimates of determinants of child poverty
Marginal effects β 0 _cons 9.675*** 9.675 15921.000 β 1 dum_gender 0.419*** 0.097 1.520 0.070
Research indicates that female children are more likely to experience multidimensional child poverty compared to their male counterparts The analysis shows a statistically significant positive relationship between a child's gender and the likelihood of poverty, with a p-value of less than 0.01 Specifically, the odds ratio of 1.520 suggests that when the gender shifts from male to female, the odds of higher child poverty increase by approximately 54% Additionally, the marginal effect of 0.07 indicates that female children have a 7% higher probability of falling into poverty compared to male children.
The education level of parents significantly influences child poverty rates Analysis reveals that children whose fathers have attained secondary education are 29% less likely to be in poverty compared to those whose fathers have only primary education This reduction is even more pronounced for children whose mothers have completed high school or higher, with decreases of 50% and 52%, respectively Additionally, the impact of a mother's education is notable, with a 1 percentage point increase in her education correlating with a higher probability of reducing child poverty The highest probability of alleviating child poverty, at 15.3%, is observed among mothers with post-secondary or higher education These findings align with previous research by Adetola and Olufemi (2012) and UNICEF (2008), highlighting that education enhances human capital, which in turn affects wages and income.
The employment status of a mother plays a crucial role in reducing the likelihood of her child falling into poverty In contrast, the job status of the father does not show significant effects Specifically, children of wage-employed mothers are 23% more likely to experience poverty compared to those with unemployed mothers This finding aligns with Ruhm (2004), which highlights the strong correlation between parents' occupations and the adverse conditions faced by their children.
UNICEF (2008) emphasizes that in Vietnam, households with professional, skilled sales, or service staff are crucial in reducing the risk of child poverty.
Household income is a significant factor influencing poverty levels This study employs the natural logarithm of annual household expenditure per capita as a proxy for total household income, utilizing a double log model The findings indicate that a 1% increase in total consumption per capita correlates with a 23.4% reduction in the probability of being poor, highlighting the strong link between household income and child poverty, while controlling for other variables.
Regional effects significantly influence the risk of poverty among children living outside the Northern uplands The analysis shows that children in other regions, specifically the Red River Delta, Mekong River Delta, and Central Coastal areas, have a lower probability of being multidimensionally poor compared to those in the Northern uplands This finding aligns with UNICEF (2008), which identified the Mekong River Delta and North West as regions that increase the likelihood of child poverty.
Summary
This section of the thesis presents findings from a statistical analysis based on Young Lives Data, highlighting that child labor, toilet facilities, and cooking fuel are the primary indicators of deprivation A poverty cutoff of 0.3 was utilized to construct the multidimensional poverty index using the Alkire-Foster methodology, with a breakdown method illustrating the contributions of various elements to overall poverty The results indicate that indicators within the shelter domain significantly contribute to the multidimensional poverty index, with the upland northern region showing the highest contributions The chapter concludes with a logistic regression analysis that assesses the factors influencing children's risk of falling into poverty, revealing that household characteristics are strongly correlated with this risk The marginal effects suggest that household income and parental education play crucial roles in reducing overall poverty.