Determinants of child health in krong buk district, daklak province

THE UNIVERSITY OF ECONOMICS
HO CHI MINH CITY
VIETNAM

INSTITUE 0 F SOCIALS TUD IES
THE HAGUE
THE NETHERLANDS

THE VIETNAM-NETHERLANDS PROJECT FOR M.A PROGRAM
IN DEVELOPMENT ECONOMICS

DETERMINANTS OF CHILD HEALTH
IN KRONG BUK DISTRICT, DAK LAK PROVINCE

BY

E>OAN TI-Q: THU HUONG

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS IN ECONOMICS OF DEVELOPMENT

SUPERVISOR: DR. HO NGQC PHUONG

HO CID MINH CITY, 2003


ACKNOWLEDGEMENT
I would like to thank the Vietnam - Netherlands Project for MA Program m
Development Economics for offering me their scholarship to attend this program.
I would like to send my gratefulness to all the lecturers and the staff of the program for
their helping and providing me with knowledge during my study.

I would like to thank Dr. Ho Ngoc Phuong, my supervisor, Dr. Nguyen Van Ngai and
MA Nguyen Htiu Loc for their useful advice, comments, and their kindness in my
thesis process.
I would like to thank the local government of Krong Buk district, Buon Ho town and

l

\

Chu Bao commtme of Dak Lak province, the Krong Buk hospital, the Representative
of Ministry of Health in Ho Chi Minh City for providing me with their useful
documents.
I would like to thank Mr. Nguyen Ngoc Thinh and his family, Miss H'Brat Nie and her
family for their help during my survey in Krong Buk district.
I would like to thank all my classmates, especially Chung Thuy, Phuong Chi, Minh
Phuong, Bich Ngoc, Tan Viet and Van Dung, who are always my companions in my
study.
Finally, my greatest gratefulness belongs to my family, especially my mother. With
their love and financial suppmt, I am able to overcome my two-year hard study.
Ho Chi Minh City, September 2003.

ii


CERTIFICATION
I certify that the substance of this thesis has not already been submitted for any degree
and is not being current submitted for any other degree.

I certify that to the best of my lrnowledge, any help received in preparing this thesis,

and all sources used, have been aclrnowledged in this thesis.

DOAN THI THU HUONG
Date: September 08 2003.

ii,i


ABSTRACT
In 2001, Dak Lak, a province locating in Central Highlands of Vietnam, still had the
highest rates of undernutrition of children under five years old in comparison with
other provinces. The National Institute of Nutrition concludes that one of the reasons
causing child malnutrition is that child nutritional knowledge of the parents and
putting this knowledge into practice within a household is not improved much.
Therefore, this thesis aims to study whether the improvement in parent's education
influences child health status in Ede group, a native ethnic minority in Dak Lak
province, or not. Beside that, other characteristics of the household and the child will
also be examined to find out the determinants of child malnutrition in Ede group.
Based on the results, this thesis suggests some policy implications as well.

iv


TABLE OF CONTENTS
CHAPTER 1 : INTRODUCTION ........................................................................ ;................................ 1

1.1 PROBLEM STATEMENT............................................................................................................. 1
1.2 OBJECTIVES OF THE STUDY .................................................................................................... 3
1.3 RESEARCH QUESTIONS ............................................................................................................ 3
1.4 RESEARCH HYPOTHESIS .......................................................................................................... 4

1.5 RESEARCH METHODOLOGY ................................................................................................... 4
1.6 THESIS DESIGN ........................................................................................................................... 5
CHAPTER 2 : LITERATURE REVIEW ............................................................................................. 7

2.1 DEFINITIONS ......................................................................................................................... :..... 7
2.1.1 Human capital.. ....................................................................................................................... 7
2.1.2 Household ............................................................................................................................... 7
2.1.3 Health ..................................................................................................................................... 8
2.1.4 Indices of nutritional status ..................................................................................................... 8
2.1.4.1 Height for age .......................................................................................................................................... 8
2.1.4.2 Weight for height .................................................................................................................................... 9
2.1.4.3 Weight for age ......................................................................................................................................... 9

2.1.5 Nutritional conditions ............................................................................................................. 9
2.1.5.1 Stunting .............................................................................................. :...................................................... 9
2.1.5.2 Wasting .................................................................................................................................................. 10
2.1.5.3 Underweight .......................................................................................................................................... 10

2.2 HUMAN CAPITAL THEORY .................................................................................................... 10
2.3 THE ECONOMICS OF HOUSEHOLD BEHAVIOUR .............................................................. 12
2.4 ECONOMIC MODELS OF HOUSEHOLD BEHAVIOUR RELATING TO HEALTH ............ 16
2.4.1 Household production functions and reduced-form demands for health and nutrients ........ 16
2.4.2 Child health production model ............................................................................................. 20
2.4.3 Bargaining model of household behaviour ........................................................................... 23
2.4.4 Dynamicproductionfunctions .............................................................................................. 24

2.5 EMPIRICAL STUDIES ............................................................................................................... 25
2.5.1 Papua

N~w

Guinea ............................................................................................................... 25

2.5.2 Mozambique ........................................................................................................................... 26

2.5.3 Vietnam ................................................................................................................................. 27
2.6 CONCLUSION ............................................................................................................................ 28
CHAPTER 3 : OVERVIEW OF CHILD HEALTH CARE IN KRONGBUK BUK DISTRICT,
DAI( LAK PROVINCE ........................................................................................................................ 31

v


3.1 AVERAGE POPULATION AND HEALTH BUDGET .............................................................. 31
3.2 MALNUTRITION RATE OF CHILDREN UNDER FIVE YEARS OLD ................................. 32
3.3 INFANT MORTALITY ............................................................................................................... 33
3.4 IMMUNIZATION ........................................................................................................................ 33
3.5 CONCLUSION ............................................................................................................................ 35

CHAPTER 4: MODEL SPECIFICATION AND RESEARCH METHODOLOGY ..................... 36
4.1 MODEL SPECIFICATION .......................................................................................................... 36

4.1.1 Variablejustification ............................................................................................................ 36
4.1.2 Model specification ............................................................................................................... 41
4.2 RESEARCH METHODOLOGY ................................................................................................. 42

4.2.1 Data requirement .................................................................................................................. 42
4.2.2 Calculating height for age index ........................................................................................... 42
4.2.3 Choosing sample size ............................................................................................................ 43
4.2.4 Method of sampling............................................................................................................... 44

4.2.5 Analysis methods ................................................................................................................... 44
4.3 CONCLUSION ............................................................................................................................ 45

CHAPTER 5: EMPIRICAL ANALYSIS OF CHILD HEALTH IN KRONG BUK DISTRICT,
DAK LAK PROVINCE ........................................................................................................................ 46
5.1 DESCRIPTIVE ANALYSIS ................................................................ ,....................................... 46

5.1.1 Univariate analysis ............................................................................................................... 46
5.1.1.1 Measures oflocation ....................................................................... .".................................................... .46
5.1.1.2 Measures of dispersion ........................................................................................................................ .47

5.1.2 Bivariate analysis ................................................................................................................. 48
5.1.2.1 The relationship between malnutrition and sex and age of children ................................................. 51
5.1.2.2 The relationship between malnutrition and family location ............................................................... 54
5.1.2.3 The relationship between malnutrition and parents' education ......................................................... 55
5.1.2.4 The relationship between malnutrition and number of children in a household ............................... 58
5.1.2.5 The relationship between malnutrition and household's expenditure ............................................... 59
5.1.2.6 The relationship between malnutrition and mother's possibility of speaking Vietnamese .............. 60
5.1.2.7 The relationship between malnutrition and the interval between two births .................................... 61

5.2 REGRESSION ANALYSIS ......................................................................................................... 62

5.2.1 Regression result ................................................................................................................... 62
5.2.2 Diagnostic testing ................................................................................................................. 63
5.2.2.1 Testing for omitted variables and incorrect functional form .............................................................. 63
5.2.2.2 Testing for heteroscedasticity ............................................................................................................... 64
5.2.2.3 Testing for multicollinearity ................................................................................................................. 64
5.2.2.4 Testing for normality ............................................................................................................................ 64

5.2.3 Interpretation of the regression results ................................................................................. 65

vi


5.2.3.1 Effects of child's characteristics .......................................................................................................... 65
5.2.3.2 Effects of household's characteristics .................................................................................................. 65
5.2.3.3 Effects of community's characteristics ................................................................................................ 67

5.3 CONCLUSION ............................................................................................................................ 67
CHAPTER 6 : CONCLUSION AND RECOMMENDATION ......................................................... 69
6.1 CONCLUSION ............................................................................................................................ 69
6.2 POLICY IMPLICATIONS ........................................................................................................... 69
6.3 LIMITATION OF THE STUDY .................................................................................................. 71
APPENDIX ............................................................................................................................................ 72
BIBLIOGRAPHY ................................................................................................................................. 85

vii


LIST OF FIGURES
Figure 2.1 Parents exhibit equal concern between children ............................................ 14
Figure 2.2 Parents exhibit greater concern for child A ................................................... 14
Figure 3.1 Malnutrition rate of children under 5, Krong Buk district ............................ 32
Figure 3.2 Fully vaccinated coverage among children under 1, Krong Buk district ...... 34
Figure 5.1 The distribution of residuals .......................................................................... 65
Figure A2.1 The distribution of HAZ .............................................................................. 79
Figure A2.2 The distribution of MOAGEBI ............................................................................... 79
Figure A2.3 The distribution ofFAEDU ........................................................................ 80
Figure A2.4 The distribution of MOEDU .................................................................................... 80
Figure A2. 5 The distribution ofFAHEI... ..................................................................................... 80

Figure A2. 6 The distribution of MOREl ...................................................................................... 81
Figure A2. 7 The distribution ofEXPMON .................................................................................. 81
Figure A2.8 The distribution ofNUMCHI ..................................... :............................................ 81

viii


LIST OF TABLES
Table 2.1 Summary of empirical studies ......................................................................... 29
I

Table 3.1 Average population and Health budget, ITable 3.2 Infant mortality, ITable 3.3 Immunization coverage among children under 1, Krong Buk district ............ 33
Table 4.1 Variable explanation ....................................................................................... .40
Table 4. 2 Malnutrition classification system.................................................................. .43
Table 5.1 Description of variables .................................................................................. .46
Table 5.2 Distribution of nutritional indicators (Z-scores) ............................................ .48
Table 5.3 Means of height for age Z-score, weight for height Z-score, and weight

for age Z-score of children under five years old by gender, parent's
education, expenditure quintile, regions and age group in month ................. .49
Table 5.4 Prevalence oflow levels of nutritional indicators by sex ................................ 51
Table 5.5 Prevalence of low levels of nutritional indicators by age categories .............. 52
Table 5.6 Prevalence oflow height for age (stunting) by sex and age group ................. 53
Table 5. 7 Prevalence of low levels of nutritional indicators by family location ............. 55
Table 5.8 Prevalence of low levels of nutritional indicators

by father's years of schooling ........................................................................ 56
Table 5.9 Prevalence of low levels of nutritional indicators

by mother's years of schooling ....................................................................... 51
Table 5.10 Prevalence of low levels of nutritional indicators by number of children .... 58
Table 5.11 Prevalence oflow levels of nutritional indicators by expenditure ................ 59
Table 5.12 Prevalence of low levels of nutritional indicators by mother's possibility

to speak Vietnamese ....................................................................................... 60

ix


Table 5.13 Prevalence of low levels of nutritional indicators by distance from the
previous birth ............................ :..................................................................... 61
Table 5.14 Summary of regression result.. ...................................................................... 63
Table A3.1 The variance-inflating factor ......................................................................... 82

X


ACRONYMS
Coef.

Coefficient

GSO

General Statistical Office

HAZ

Height for age Z-score

ISID

Integrated and information department

MoH

Ministry of Health

Std. Err.

Standard Error

Unicef

United Nations Children's Fund

WAZ

Weight for age Z-score

WHZ

Weight for height Z-score

WHO

World Health Organization

xi


CHAPTER 1 : INTRODUCTION
1.1 PROBLEM STATEMENT

The ultimate target of economic development is human development. Particularly, the
objective of development is "to increase the capabilities of people to lead full,
productive, satisfying lives" (Griffin and McKinley, 1994:1). The benefits that people
receive from economic development are not only money income, material goods and
services but also a long life and good health. Those benefits originate from the stock of
capital which can be divided into three components: the stock of natural capital, the
stock of manmade physical capital and the stock of human capital. Among them,
I

human capital is now identified to play an important part in the development process.
In human capital formation health has a positive affected position. (Griffin and
McKinley, 1994).
According to Ponce et al (1998: 257), child nutrition is an important objective of
human capital since the well-being of a country's populationis partially reflected by
the nutritional status of its children. Good nutrition promotes the development of
cognitive capacity in the early childhood and school perfonnance. In long-run,
improved child nutrition results in higher labor productivity and longer life expectancy
in adulthood. Since the family, particularly the mother, is the basic provider of health,
the education of women is a significant detenninant of the health status and
development of children (Unicef, 1994).
The tenth Commonwealth Health Ministers' conference concluded that women played
a crucial role in 'taking necessary steps to target public health education programmes'
(Rao, 1996: 97). In addition, according to Kickbasch (1992), 'improving education
contributes to health more efficiently than various expensive medical care projects'

(Rao, 1996: 105). Hartmann (1992) wrote in his book that educated women control


family planning more effectively. Therefore, educating women is one important way to
tackle health problems (Rao, 1996).
Investing in human development in general .and in health and education of children in
particular has been followed by the Vietnamese government since 1945. As a result,
Vietnam has gained a high average rate of literacy of 88%, enjoyed high life
expectancy and low infant and young child mortality (Unicef, 1994: 1). Vietnam has
reached "superior levels of infant mortality and fertility" at its level of per capita GDP
in comparison with other countries in the South East Asia. This achievement is the
result of an extensive health care delivery network combining with a well developed
primary health care system, a large supply of health workers, and a well-organized
national public health programs such as the Extended Program on Immunization
(World Bank, 2000: 9). However, child malnutrition rates in

Vi~tnam

are still among

the highest. The National Institute of Nutrition and United Nations Children's Fund
(Unicef) have conducted a recent national survey of nutrition and found that "thitty
nine percent of under five year old children are malnourished in terms of weight for
age and thirty four percent are undernourished in terms of height for age" (World Bank
et al, 2001: 27).

In 2001, Dak Lak, a province locating in Central Highlands of Vietnam, still had the

highest rates of undernutrition of children under five years old in terms of stunting
(53%) and underweight (45.3%), in comparison with other provinces (MoH, 2001). In

Dak Lak, there are forty three ethnic groups living together. Among them, Ede people
is the native (Tran, 2002). Its population ranks the second after the Kinh (GSO, 2001).
However, according to Nguyen (1996), the height and the weight of the Ede children
are less than that of the Kinh children, showing the nutritional status of stunting and
wasting.

2


According to the report of the National Institute of Nutrition, one of the reasons
causing child malnutrition is that child nutritional knowledge of the parents and
putting those knowledge into practice within a household is not improved much (MoH,
2003). As stated above, educated mothers are proved to have positively effect on
family's health. Therefore, the problem emerges that whether there is the impact of
mother's education on Ede child health in Oak Lak province. This is the goal of this
thesis. Beside that, other characteristics of the household and the child will also be
examined to find out the determinants of child malnutrition in Ede group.
1.2 OBJECTIVES OF THE STUDY

As the problem statement has mentioned, this thesis is done to report empirical
evidence from Dak Lak province. The relationship studied is maternal education and
health status of young children. Hopefully, the results of the thesis may help to support
the fact that improving the education of women may be an efficient way to attain child
health development. Besides, the child health situation in Ede group presented in this
paper will be good information for policy makers' work.

1.3 RESEARCH QUESTIONS

In order to accomplish the purpose of my study, I formulate research questions that
will guide my research later:

Is there a positive effect of mother's years of schooling on the improvement of
height for age Z-score of young children in Ede group in Dak Lak province in
2003?
Is there a positive effect of whether Ede mothers can speak Vietnamese on the
improvement of height for age Z-score of young children in Ede group in Dak
Lak province in 2003?

3


Is there a positive effect of expenditure or income per capita on the improvement
of height for age Z-score of young children in Ede group in Dak Lak province in
2003?
What are other determinants of the situation of stunting of young children in Ede
group in Dak Lak province in 2003?
1.4 RESEARCH HYPOTHESIS

As mentioned in my problem statement, women play a very important role in making
decisions and implementing them within a household. Moreover, because restricted
resource of a household limits women's ability to make family planning, education
will improve capacity of the mothers. Therefore, my research hypothesis is that an
increase in mother's years of schooling will have positive effect on the improvement
of height for age Z-score of young children in Ede group in Dak Lak province in 2003.
The mother's ability to speak Vietnamese and expenditure or income per capita also
have the same impact as mother's years of schooling does. Besides, other determinants
such as father's education, parents' height, mother's age at birth, number of children in
a household and family's allocation will affect child health.
1.5 RESEARCH METHODOLOGY

Child health is measured in terms of height for age Z-score which is calculated by the

EPI-INFO software. Both descriptive statistics and econometric technique will be
employed to serve for the purpose of the thesis. Descriptive statistics method is used to
analyze the current situation of child health care and other economic factors that can
influence it, basing on the secondary and tertiary data. The ordinary least square (OLS)
estimation will be exploited to run regression on variables mentioned in the analytical

4


framework. The results of the ordinary least square estimation with the descriptive
analysis are the answers for my research questions.
Because the data from the Vietnam Living Standard Survey 1997-1998 has been used
by Vo (2001) in her thesis of the role of parent's education and preventive measures on
child health in Vietnam, I have to conduct my own household survey to collect
primary data for my thesis. The survey is done in ILak province. Thanks to the local government, my survey was finished successfully,
with full information of 100 households locating in Chu Baa commune and Buon Ho
town. With the help of Ministry of Health, Institute of Nutrition, Economics
Department of Tay Nguyen University and Krong Buk hospital, secondary and tertiary
data from official reports are used to analyze child health care in Dak Lak province in
recent years.
1.6 THESIS DESIGN
The organization of my thesis includes five chapters.

Chapter 1 is the introduction. The importance and purpose of the study are written in
the

probl~m

statement and objectives of the research. Besides, research questions,

research hypothesis and methodology used are also in this chapter.
Literature review is the second chapter. It presents the concepts relating to child health,
human capital theory, and economics of household behaviour, health models,, and
empirical studies.
Chapter 3 expresses the overview of child health care in Iprovince in recent years.

5


Chapter 4 describes the way I set up my own model, basing on the literature review,
and methods that I apply to analyze data.
Chapter 5 will start from descriptive analysis about child nutrition situation in Krong
Buk district, basing on the survey's data. Then regression analysis will be presented.
The final chapter, chapter 6, is the conclusion and policy implication of the study,
basing on the results of descriptive analysis and OLS regression.

6


CHAPTER 2 : LITERATURE REVIEW
Before reviewing the literatures related to the thesis research, it is necessary to
understand some basic concepts. Then the literature review will examine the general
theory of human capital development. Next, this part will investigate the more specific
theory relating to health, theory of household behaviour. After that it will refer to some
models linking to health. Finally, a search of empirical studies on child health care will
be done.
2.1 DEFINITIONS
2.1.1 Human capital

According to Pearce (1992: 188), the nature of human capital is that 'investments are
made in human resources so as to improve their productivity'. Whether such
investment that is economically worthwhile depends on the excess of benefits over
costs by a sufficient amount. Investment in men covers from formal schooling, postschool training to family care in the pre-school years, the acquisition of improved
health, and investments in labor market information via job search.
2.1.2 Household

'A household is a unit consisting of people with a common budget who organize some
aspects of work and consumption together' (Himmelweit et al, 1998: 149). This
definition is a combination of two other definitions of household proposed by Roberts
(1991) and Messer (1990). The Roberts's definition focuses on the activities occurring
within a household and their importance to society as a whole while the Messer's
definition concentrates on the internal relations of a household, reflecting in the use of
household's budget of human and material resources. Combining these two definitions
I

7


results in the concept of a household that is widely used in most economic researches
(Himmelweit et al, 1998: 150).
2.1.3 Health
In 1948, the World Health Organization (WHO) defined health as 'A state of complete
physical, mental and social well-being and not merely the absence of disease or
infirmity' (Abel-Smith, 1994: 4). However this concept does not show how health can
be measured. Besides, the ambiguity emerges with the term 'physical well-being'.
Who determines it, the patient or the doctor? How can mental well-being be measured
without social well-being? (Abel-Smith, 1994: 4).
Also commenting on this definition, McGuire et al (2000: 21) writes that if we rely on

WHO's definition to judge health, "few of us will ever be healthy". However• he
admits that the definition includes some good points such as 'the absence of disease
and infirmity' and the presence of subjective feelings.
2.1.4 Indices of nutritional status
2.1.4.1 Height for age

Height for age index reflects the past undernutrition status or chronic malnutrition
(Cogill, 2001) because it measures a child's cumulative nutritional status since birth.
Each time a child suffers from inadequate nutrition, he or she will stop growing or
grow more slowly. When the child recovers from such experiences, he or she cannot to
catch up the height that would have been reached as usual. Therefore the child may be
shorter than a child without malnutrition, conditional on his or her age (Ponce et al,
1998). This measurement is available for children 0-4 years old (Heltberg and
Johannesen, 2002). The disadvantage of this index is that it cannot measure short-term
changes in malnutrition.

8


2.1.4.2 Weight for height

Weight for height is an indicator of cun-ent or acute undemutritiqn or wasting. In the
short run, a child who is bearing a reduction in nutrition will lose his or her weight
quickly though his or her height does not have enough time to be affected. Thus this
indicator is used to 'examine short-term effects such as seasonal changes in food
supply or short-term nutritional stress brought about by illness' (Cogill, 2001: 11).
This index has an advantage that it can be applied when we have difficulty in
determining exact ages. However it reveals nothing about inadequate nutritional intake
in the past.
2.1.4.3 Weight for age

Weight for age· is an index to define whether a child is underweight at a specific age.
This index can show us the past and present undernutrition status because a child with
low height for age or low weight for height will also have low weight for age. In
general, weight for age is less useful than height for age and weight for height as it is
unable to distinguish between short term and long term malnutrition (Cogill, 2001;
Ponce et al, 1998).
2.1.5 Nutritional conditions

2.1. 5.1 Stunting

Stunting is identified by low height for age. This status points out the past growth
failure because low height for age is a consequence of 'a slowing in the growth of the
fetus and the child and resulting in a failure to achieve expected length as compared to
a healthy, well nourished child of the same age' (Cogill, 2001: 11). Causes of stunting
include chronic insufficient protein and energy intake, frequent infection, sustained
inappropriate feeding practices and poverty.

9


2.1.5.2 Wasting

Wasting identifies the condition of current or acute malnutrition of a child because of
failure to gain weight or actual weight loss. The weight of a wasting child is
significantly 'below the weight expected of a child of the same length or height'
(Cogill, 2001: 12). Wasting is defined by weight for height index. Causes of wasting
include inadequate food intake, incorrect feeding practices, disease, and infection or,
more frequently, a combination of these factors.

2.1.5.3 Underweight

Underweight is reflected by low weight for age index. It is 'a composite measure of
stunting and wasting' (Co gill, 2001: 11 ). People use this indicator to evaluate changes
in the magnitude of malnutrition over time.
2.2 HUMAN CAPITAL THEORY
In the beginning Of the human capital theory, education and training was paid atte:ption
most by some economists such as Mincer (1958), Schultz (1960) and Becker (1962).
Most research concentrates on investment in education and its effect on future income.
After formal education in school, on-the-job training and the building-up of experience
are other forms of investment in human capital. Although the core of human capital
theory is education, health is recognized as an important component as well.
According to Schultz, improved health and longevity, the reduction in child mortality
and greater resources devoted to children in the home also enhance human capital
(Eatwell et al, 1991). Human capital formation through education and health is one of
critical determinants of the success or failure of development, in which achieving high
growth rates and reducing income inequalities is the target (Zulkifli, 1999). However
health was paid less attention than education because health was often considered to be
t

an exogenous variable (Kooreman and Wunderink, 1997: 180). Besides, measurement

10


of health is more difficult than measurement of formal education, and this partly
explains why there are not much empirical studies in health as those in education
I

(Herrin, 1999).

There is a controversy that income growth finally lead to improvements in public
health. However, there is an argument that health programs must come first as they
will upgrade labor productivity and the personal sense of well-being (Hogendom,
1992: 317). This issue of reverse causality is also claimed by Langhammer when he
writes in his paper: "while health improves income prospect, income growth itself
improves health" (Langhammer, 1999: 3). Preston, in his research, suggests a crosssection relationship in parabolic form between income and life expectancy. Given
level of real income per capita, people tend to live longer as time goes by (Gilles et al,
1996). Better health in tum can contribute to development. Gilles et al (1996) argue
that the quality of human resources will be improved by health services now and. in the
future. Healthy workers benefit the economic growth immediately. Better child health
and nutrition promote future productivity growth. Health expenditures also lengthen
the future expected working life. Nevertheless the results of some researches
measuring the effects of better health on labor productivity do not always support the
expected positive effects of better health on productivity. People maintain their
productivity although they may not feel well as they find ways to adapt to wide ranges
of ill health and malnutrition.
Finding the relationships between health and income growth, or/and between health
and productivity, human capital theory helps to explain why individuals or the
governments spend money on health and other human services. The final target of
those people is to raise their income and productivity. The tool that human capital
theory uses to analyze health is the cost-benefit analysis. The family or the government
would invest in health if the discounted benefit was larger than the discounted cost.
However, the cost-benefit analysis measure expenditures on health but it does not

11


measure the time or effort that a household or a community devotes to health
investment. Besides, it is hard to quantify and verify empirically both returns to
investment in health and the economic effects of better health (Gillis et all, 1996).

Thus in the mainstream of human capital investment, studies in health have not been
paid much attention by economists.
As the main producer of health is a household and health status is determined by the
multiplicity of factors that go together, human capital theory does not provide
researchers with enough instruments to go insight the essence of health and its
determinants. This limitation requires another economics theory that allows
approaching to health in the level of household and that can capture other determinants
of health except expenditure. Thus, the economics of household behaviour was used as
an additional method.
2.3 THE ECONOMICS OF HOUSEHOLD BEHAVIOUR
Individuals formulate a household because living in a household with others generates
higher utility than living alone does. The benefits from living together are the love and
companionship, the existence of economies of scale in a household, and the possibility
of specializing (Kooreman and Wunderink, 1997). However, if the utility gained
within the household is less than the utility attainable outside the household,
individuals will choose to break a household down. The threshold of the utility of
breaking down is called the threat point or the fall-back position.
The economics of household behaviour, or another name - intrahousehold resource
allocation, was rooted from the 'New Household Economics'. From the point of view
that household is not only the consumer of the goods and services produced in the
private and public sectors but also the producer of home production, Gary Becker
emphasizes the productive role of households. The essence of his approach is that the
\

commodities that generate utility for the household are a result of the household

12


production, basing on a combination of time, goods purchased in the market, and

goods produced at home, in accordance with a single set of preferences. Since then,
economists have used this approach to study the determinants of education, health,
fertility, child fostering, migration, labor supply, home production, land tenure, and
crop adoption (Haddad et al, 1997). There are two ways to approach intrahousehold
allocation: unitary approaches and collective approaches.
Using unitary approaches, many economists such as Becker and Tomes (1976),
Behrman, Pollak, and Taubman (1982), and Behrman (1996) examine models of
parental investment in children (Haddad et al, 1997). Parents maximize their joint
welfare function, of which determinants are their own consumption, the adult income
of each child, and the size of transfers made to each child. This maximization is
constrained by parental budget and earnings production function for each child. The
child's earnings production function is defined over a function of human capital
investments made in a child and that child's initial endowment. This model is named
as "the parental altruism model". From the parental altruism model emerges a problem
that how fair the parents make their concern to each child in investment in children.
This restriction gives out other two models, wealth model and separable earningstransfers model.
The wealth model assumes that parents consider only to their children's total level of
wealth. Therefore, parents pay attention to each child equally and transfers are made in
such a way as to equalize children's wealth. In the other hand, the separable earningstransfers model allows different parental transfers to children, depending on each
child's endowment. If the parents' utility is enhanced with the equalization of their
children's future earnings, they will invest more in the child that has lower initial
endowment. On the contrary, if parents' aim is to maximize the total future earnings of
their children, the child with the best future prospects will receive more investments
than the others do.

13


Figure 2.1 and figure 2.2 describe the relationship between household's utility and
heath outcome of children with different concerns from parents.

Figure 2.1: Parents exhibit equal concern between children ·

Health related
outcome of
child B

Health related outcome
of child A
Source: Patton (1993: 7)

Figure 2.2: Parents exhibit greater concern for child A

Health related
outcome of
child B

Health related outcome
of child A
Source: Patton (1993: 8)

14