African Development Bank Seminar - African Economic Research Consortium

22-24 November, Tunis

Investing in people:
Education and health

Pr Abdoulaye DIAGNE
Economic and Social Research Consortium (CRES),
Faculty of Economic Sciences and Management (FASEG),
Cheikh Anta Diop University of Dakar (UCAD)

Dakar, 6 November 2006

 Contents

INTRODUCTION................................................................................................................................................. 3
SECTION 1: WHY INVEST IN EDUCATION AND HEALTH: THEORETICAL JUSTIFICATIONS ... 3
A/ ECONOMIC EFFECTS ....................................................................................................................................... 4
B/ SOCIAL EFFECTS ............................................................................................................................................. 6
SECTION 2: WHY INVEST IN EDUCATION AND HEALTH: EMPIRICAL EVIDENCE ...................... 8
A/ THE ECONOMIC IMPACT OF INVESTING IN HUMAN CAPITAL ............................................................................ 8
1. Effects on personal income........................................................................................................................ 8
2. Effects on growth....................................................................................................................................... 9
B/ THE SOCIAL IMPACT OF HUMAN CAPITAL INVESTMENT ................................................................................ 13
1. Effect of human capital on poverty reduction.......................................................................................... 13
2. Impact of the human capital of parents on that of children..................................................................... 15
3. Impact of a mother’s education on demographic variables .................................................................... 16
SECTION 3. EDUCATION AND HEALTH STATUS IN AFRICA ............................................................. 18
GENDER AND ZONAL DISPARITIES .................................................................................................................. 21
SECTION 4: LEVERS FOR ACCELERATING INVESTMENT IN PEOPLE: ......................................... 23
EDUCATION AND HEALTH POLICIES....................................................................................................... 23
CONCLUSION ................................................................................................................................................... 29
BIBLIOGRAPHY ............................................................................................................................................... 30
ANNEXES ........................................................................................................................................................... 35

2
 Introduction

Development strategies and policies have always sought to identify the key
variable on which governments can act to speed up the creation of wealth and thus
bring about a quick improvement in the wellbeing of their citizens. Hence,
emphasis has kept shifting from physical capital accumulation to labor, natural
resources, institutions, etc. This paper puts together a set of theoretical arguments
and empirical evidence to underpin the thesis that investing in human capital
through education and health, is the absolute prerequisite for ensuring sustainable
economic growth, behavior change and a democratic society. It also demonstrates
that the current context of resource scarcity, even pauperism, dogging low-income
countries is not a major obstacle blocking access to quality education and health
for the poor. Indeed, the volume and quality of human capital accumulated by a
country depends, in the long run, on the policy choices made in education and
health.

This paper is organized as follows: Section 1 presents the theoretical reasons for
investing in education; Section 2 reviews empirical studies on the interactions
between education, health, growth and poverty; Section 3 presents the status of
human development from the education and health standpoints with focus on
African countries; Section 4 demonstrates that any progress made in these areas
depends essentially on the type of policies implemented. The Conclusion outlines
all the lessons learnt from all the theoretical and empirical arguments developed.

Section 1: Why invest in education and health: theoretical justifications

Education and health are certainly the two biggest investments any individual or
society can make since they generate considerable and mutifaceted economic and
social benefits. Usually, a more educated and healthy person has access to better
jobs and earns a higher income than one who is less educated or unhealthy. More
educated societies have better health indicators and enjoy a higher level of
development.

In summary, education and health increase human productivity, raise life

expectancy and facilitate community life. They have a major impact on economic

3
growth and increase individual and collective wellbeing. However, the direct
benefits of education and health should not blind us to the positive interactions
between them.

A/ Economic effects
Job search, productivity and remuneration
Better educated and healthy persons are more likely to get employment and are
less likely to lose their jobs, if they are gainfully employed. The labor participation
rate increases with the level of education attained by individuals. Persons with the
highest qualifications earn the highest incomes. Analyses of the effects of
education have often striven to attribute differences in personal incomes to
workers’ unequal endowment in human capital. Accordingly, the traditional theory
of human capital (Mincer, 1974; Becker, 1975) explicitly posits that workers are
remunerated on the basis of their marginal productivity, and that more educated
workers logically earn higher wages because they are supposedly more productive.

While the effect of education on personal incomes and economic growth has been
established for a long time, that of health on the same variables was demonstrated
a lot later. Over time, it gradually became clear that healthy workers are more
productive. Since they are seldom absent, healthy workers enable employers to
reduce labor turnover, thereby increasing labor productivity; accordingly they earn
higher wages than sick workers. Various studies on the economic effects of certain
diseases like malaria, AIDS or tobacco-related illnesses have highlighted the
exorbitant cost of such diseases to corporations, workers and the community
(Banque mondiale, 2005; Russell, 2004).

Human capital and economic growth

In early growth models such as that of Solow (1956), long-term growth was
made dependent on technological advance, without any real analysis of its origin,
although mention was made of the important role of education. Research on
growth factors took a new dimension with the works of Romer (1986) and Lucas
(1988) who focused on new theories of endogenous growth. Through more rigorous
analysis, they demonstrated that countries which made a greater effort in
educating their citizens ended becoming richer. This result was obtained either by
establishing an analogy between human capital (education) and physical capital

4
(i.e. considering it as a factor whose accumulation raises the level of production),
or by assuming that human capital directly affects the growth rate through
technological advance (Gurgand, 2003).

Although the role of education has been established, there is still need to
identify the mechanisms through which it affects economic growth. Nelson and
Phelps (1966) had already stated that since education is essentially the capacity to
understand new information and adapt one’s behavior accordingly, economies that
have the greatest number of highly-skilled workers will more rapidly adopt and
implement the most efficient technologies. According to them, it is the level of
education that raises the economic growth rate by speeding up the assimilation of
technological progress. Another mechanism was identified later by Romer (1990)
who examined the relevance of research and development (R&D) activities. He
established that accumulating a stock of knowledge makes it possible to increase
efficiency in generating wealth through labor and capital. Raising the level of
education, and consequently the number of highly-skilled workers who can
participate in such accumulation of knowledge, will increases the pace of
discoveries and consequently multiply economic growth possibilities.

Aghion and Cohen (2004) articulate these two mechanisms by making a

distinction between imitation economies and innovation economies. They
demonstrate that imitation economies, which are technologically backward and
have a high potential for assimilating technologies designed elsewhere, should
invest mainly in educational levels (primary and secondary education) that are
conducive to imitation and application of new technologies. In contrast, innovation
economies have to contribute to technological innovation and, to that end, have a
large highly-qualified workforce (with university education). In both types of
economy, the determinant factor in productivity growth is the proportion of the
labor force that attains a high level of education.

Until recently, all analyses of economic growth determinants laid emphasis on

the role of education, while health, the other dimension of human capital was
ignored. It was only towards the late 1990s that a lot more studies were conducted
on the relationship between health and economic growth. One of their main
characteristics is that they use different indicators to measure health.

5
 Using life expectancy at birth as a measure of health status, Bloom and Malaney
(1998) and Bhargava et al. (2001) demonstrated that health is an important factor
of economic growth since it extends the life expectancy of the labor force. Other
studies analyzed the effects of health on growth by using demographic indicators
(fertility rate, fertility pattern and mortality rate) or examining the consequences
of AIDS on economic development. They showed that a decline in fertility enabled
more women to take up employment and helped to improve the health and
nutrition of children. Having fewer children reduces the number of dependent
persons in the family and makes it possible to invest more in the human capital of
children through health and education which are fundamental to economic growth.

B/ Social effects
The social effects of education and health include dimensions such as an
improvement in quality of life and the reduction of gender and social (rich/poor)
disparities. Education and health also play a role in reinforcing cohesion (political
stability) and democracy and help to integrate the country into the world economy.

Education and improvement of the quality of life

Education facilitates access to and understanding of medical information.
Educated persons are better able to share their resources between the various
inputs of health production. Moreover, they make better use of consultations and
medicines (Grossman, 1995).
Well-educated women have higher incomes often have husbands who are also
educated; this can improve their husbands’ health status as well. For example,
men married to educated women have the lowest mortality rate (Egeland, 2002)).
There is a very close relationship between education and an improvement in
women’s reproductive health. Many studies have shown that educated women have
are more likely to get good quality antenatal care and use contraceptive methods
that prevent unwanted pregnancies. They marry late and have fewer children than
uneducated women.

Education and ITCs

With the globalization of information, communication and economic activity,
knowledge has become a major factor of production. Production depends much on

6
capital and labor, but depends a lot more on technology nowadays. Since
technological advance is determined by the level of human resource development,
competitiveness now depends increasingly more on a well-trained labor force than
on physical capital and natural resources. The dissemination of new technologies
calls for a skilled workforce that is capable of assimilating and applying new
knowledge. In summary, countries wishing to succeed economically in the current
context of globalization need to build up a critical mass of human resources with
solid scientific and technical training that ensures rapid assimilation and
adaptation of the latest technological innovations to the local context. Only
education, in terms of massive enrolment and quality, can provide such expertise.

Externality of education
Apart from the abovementioned economic and social effects, a person’s
education can
positively (or negatively) affect the productivity of other individuals (in addition to
his own) or change their behavior. This effect is often viewed as a sort of
“collective know-how”; that is, information that is beneficial to the community
and which results from the activity of one (or several) better educated individual(s)
who cannot keep it to himself (themselves). For example, if an educated farmer
changes his farming methods (use of improved seeds, proper application of
fertilizer or pest control products), the technological innovations he adopts will
enable him to raise his income, and possibly encourage his (probably illiterate)
neighbors to imitate him and thereby raise their own incomes as well. If that were
to happen, then the benefits accruing to society from the training of one
innovative farmer will be far more than his gains as an individual. Furthermore, the
impact of education on fertility could also be extended to uneducated women.
McNay, Arokiasamy and Cassen (2003) have noted that the growing use of modern
contraceptive methods by uneducated women is the main cause of the decline in
fertility currently observed in India. Apparently, it is the dissemination of such
methods by educated women among uneducated women at home or within the
community that accounts for this phenomenon.

7
 Section 2: Why invest in education and health: empirical evidence

A/ The economic impact of investing in human capital

1. Effects on personal income
The existence of a significant and positive statistical correlation between a
person’s number of years of study, his health status and his income on the labor
market is one of the best established stylized effects in economics.

By applying the Mincerian equation to data collected in Botswana, Siphambe

(2000) demonstrates that returns increase in tandem with level of education. Alba-
Ramirez and San Segundo (1995) estimate average returns per additional year of
study at 8.4%.
According to Psacharopoulos and Patrinos (2002) in their latest update, worldwide
personal income growth per additional year of study is close to 10% on average,
with wide disparities. In general, the impact of education is greater for poorer
classes whose average level of education is lower; it stands at 9.9% in Asia, 11.7%
in sub-Saharan Africa and 12% in Latin America, compared to 7.5% for the OECD
countries; 10.9% for low-income countries compared to 7.4% for high-income
countries; 9.8% for women compared to 8.7% for men. Private returns from the
entire primary education sector (26.6%) are higher than for secondary education
(17%) or higher education (19%), with the same regional, income and gender
variations: 37.6% for primary education in sub-Saharan Africa compared to 11.3%
for secondary education in the OECD. Such private returns come with corresponding
social returns that include the cost of educational policies and benefits other than
a wage increase. Since there is no systematic estimation of benefits,
Psacharopoulos and Patrinos have proposed returns calculated from private
benefits and total costs (public and private), which are lower than private returns,
but have essentially the same characteristics. The main conclusion of this study is
that, education and especially primary education constitute, is a particularly
efficient investment for developing country governments.

Barro (1996) examines the direct and indirect effects of health on productivity
and demonstrates that an improvement in health will raise labor productivity for a
given amount of working hours, physical capital, level of education and experience.
Apart from this direct effect, an improvement in health brings down mortality and

8
morbidity rates, and consequently reduces the depreciation rate for human capital
(composed of education and health itself). Hence, an improvement in health raises
demand for human capital and has an indirect positive effect on productivity.
The microeconomic analysis of the effects of education on personal income reflects
the macroeconomic analysis of the effects of education on aggregate income and,
in particular, on per capita income growth.

2. Effects on growth
The main factors of endogenous growth are research, human capital and public
expenditure. Apparently health and education are the only sectors that integrate
all the three factors.
It has now been proven that human capital is a crucial factor of economic growth
and recent data even indicates that it is associated with other benefits such as an
improvement in health and wellbeing. Most of the empirical methods currently
used to measure the impact of human capital (education) on growth are based on
models of the 1950s and 1960s.

The relationship between human capital and economic growth is often

appreciated through regressions in the values of several countries, while
incorporating explanatory variables that relate to education, physical capital and
other variables characteristic of the social and institutional environment of the
country. Many studies have used such analyses and covered both developing and
industrialized countries.

Mankiw, Romer and Weil (1992) have used the aggregate production function of
Solow’s model, adding human capital to physical capital and labor. Their estimates
concern 98 countries on which they have data for 1960-1985. They measure human
capital accumulation for this period using the average product of the secondary
enrolment ratio for persons aged 12-17 and the proportion of those aged 15-19
within the population that is above 15 years of age. They consider this product as
an approximation of the proportion of the labor force that is undergoing secondary
education. First of all, they estimate per capita GDP determinants for 1985, which,
in addition to human capital accumulation, include the average physical capital
investment rates and population growth rates over the 1960-1985 period.

9
 These three variables account for the variance of more than 80% in per capita
GDP among countries; human capital has a significant and highly positive impact by
generating about 30% of value-added, a figure close to the percentages attributed
to physical capital and labor. Next, they use a convergence equation in which per
capita GDP growth from 1960 to 1985 depends on the same three variables as the
per capita GDP level for 1960. Again, human capital is still found to play a
significant and positive role with its value-added share inching down to about 25%.

Using models that focus on the adaptation and innovation functions of

education, Benhabib and Spiegel (1994) find that there is a correlation between the
economic growth rate of an economy, the stock of education that boosts its
innovation capacity, and a variable that reflects its capacity to catch-up with
technological advance. To prove their point, the authors calculated the catch-up
potential by determining the gap between the highest per capita income in the
world and the per capita income of each country considered. The impact of
education is then estimated to be proportionate to this gap: a positive coefficient
indicates a rise in education which translates into higher economic growth and
catch-up potential.

According to the growth model of Benhabib and Spiegel, the impact of

education on growth depends on a country’s level of development. The level of
development, in turn, depends heavily on economic and institutional conditions
(Dessus, 1998; Maurin et al., 2003). Dessus observes that the impact of education
on growth is greater on countries with more open economies. Maurin et al., show
that education is particularly valued in firms facing international competition since
it boosts the capacity to adapt in a highly competitive environment.

Barro and Sala-i-Martin (1995) present other estimates of the contribution of

human capital to growth based on the same data and period as previous studies.
Per capita GDP growth depends, inter alia, on the number of years of primary,
secondary and higher education accumulated by the people and on education
expenditure, as a percentage of GDP at the beginning of the period. While the
primary education coefficient is not significant, those of secondary and higher
education are quite high and positive: a rise in the standard deviation of the
average number of secondary school years (representing 0.9 year) is associated

10
with a 1.5% rise in growth rate, with the corresponding figures for higher education
being 0.2 year and 1% respectively. In contrast, the education of women is in
negative correlation with growth: a rise in the standard deviation of the average
number of secondary school years (0.9 year) is associated with a decline of 0.8%
per annum.

This conclusion does not support the thesis that educating women is one of the
keys to economic growth. However, these additional results show that educating
women is important for other development indicators like fertility and infant
mortality. Educating women has a negative impact on fertility and infant mortality
(Schultz, 1989; Behram, 1990; Barro et Lee, 1994). It can be inferred from the
above that educating women can encourage economic growth indirectly by curbing
infant mortality and high fertility. The result will be more resources accumulated
as savings to be invested in the basic social sectors that drive growth.

This study by Barro and Lee confirms the idea that there is a strong correlation
between enrolment – especially at the primary education level – and economic
growth. Countries that experienced an enrolment boom in the 1950s and 1960s
recorded higher economic growth rates in the 1970s and 1980s, even though other
growth factors have also been taken into account. A World Bank study (1993a)
showed that sustained investments in primary education a few decades ago were
crucial to the success of the eight fastest growing East Asian economies (the
Tigers). The historical evidence is so convincing that many economists have posited
that economic growth gathers momentum once human capital accumulation
reaches a certain threshold.
Focusing on the quality of education, Hanushek and Kimko (2000) directly
measure the “quality” of the labor force through surveys on knowledge of
mathematics and science. They use a set of international tests conducted since the
1960s by IAEP (International Assessment of Educational Progress) and IEA
(International Association for the Evaluation of Educational Achievement) to
produce indices which they apply to the labor force by matching the date of
conduct of the tests to the age structure of the workforce. By so doing, they
obtained a limited series of 31 countries. They observe that countries whose
candidates had the best scores in the tests equally had the highest economic
growth rates over the 1960-1990 period. Paradoxically, in the article by Hanushek

11
and Kimko, the number of years of study in a country, its national expenditure on
education and its teacher-pupil ratio do not account for differences in economic
performance. Only the primary enrolment ratio and appurtenance to an Asian
country influence performance.

Health and growth

Health currently plays a fundamental role in improving the productivity and
wellbeing of the population and in boosting economic growth. Empirical studies on
the impact of health on growth are based on a methodology that entails using the
panel data of certain countries and regressing the per capita GDP growth rate to
the initial level of health and of the variables that are supposed to influence the
equilibrium (economic policy, institutional, educational and other variables).

In most studies, life expectancy is used as a measure of health status.

Researchers suggest that this indicator is a major predictor of economic growth. A
five-year increase in life expectancy improves the growth rate by 46% (Barro and
Sala-i-Martin, 1995). Bloom and Melaney (1998) used a panel of 77 countries and
over a 25-year period, applying the simple linear regression method. They studied
the impact of health on growth and variables relating to demography (population
and labor force growth rates), education (secondary enrolment), natural resources,
external openness and institutional variables. For health, they obtained a positive
impact, with a 21% rise in the GDP growth rate, after a 5-year increase in life
expectancy. Bloom, Sachs and Williamson (1998) used different attendant variables
to obtain similar results, namely the positive impact of health variables on growth.

Bhargava et al. (2001) used a panel of 92 countries to study the influence of

human capital, represented (for health) by the adult survival rate on the one hand,
and the same rate weighted by the per capita GDP logarithm, on the other hand.
The other explanatory variables are the degree of economic openness, the
investment rate and fertility. The first health variable showed a positive impact
while the second produced a negative impact. This result confirms those of other
studies.

Long-term demographic and economic data on 45 developing countries shows

that high fertility aggravates poverty by slowing down growth and reducing the

12
consumption share of the poor. A high population growth rate has a negative
impact on the per capita GDP growth rate. Moreover, a high growth rate translates
into more resources needed to raise children, less economic flexibility for mothers
and limited productive investment capacity for households and the State.
Conversely, low fertility can boost growth through various channels. Recent
research has shown that fertility usually declines when more women are educated
and the health situation improves (Barro et Lee, 1994). Reducing mortality through
education, or its direct effects, will help to reduce fertility and thus boost growth
(de la Croix, Doepke, 2003). This shows that growth can be boosted by the direct
effects of an improvement in health.

Several East Asian countries and a few developing countries have benefited
from the effects of low fertility on growth. For example, it is estimated that low
fertility in Brazil contributed up to 0.7% to economic growth per year.
Almost all the studies conducted on growth have shown that improvements in
health (life expectancy, health expenditure, low fertility, child mortality, etc.)
have a positive impact on growth. All these studies tend to confirm this trend,
although they usually have variations resulting from the country samples chosen,
the periods covered, the control variables used and other factors.

B/ The social impact of human capital investment

1. Effect of human capital on poverty reduction
While there is no obvious relationship between growth and income inequality
from the macroeconomic standpoint1, econometric analyses show that per capita
GDP growth is crucial to poverty reduction2 . Meanwhile, it has been clearly
established, theoretically and empirically, that education and health play a key
role in boosting incomes and economic growth, thereby making it possible to
reduce poverty (Barro 1991; Chu et al. 1995; Tanzi et Chu, 1999).

Investing in the education and health of the poor brings economic benefits in
terms of productivity, income and economic growth. That is why poverty reduction
strategies particularly focus on these two sectors. They interact in different
spheres. More educated workers enjoy better health and earn higher incomes.

1
Cogneau, D. 2002; "New Poverty Reduction Strategies", pp 53-82, Economica.
2
Cling et al. 2002). "New Poverty Reduction Strategies", Economica.

13
 The strong correlation between poverty, health and education is widely
recognized today. Persons living in absolute poverty are five times more likely to
die before the age of 5 and twice more likely to die between the ages of 15 to 49
than those belonging to higher income groups. Similarly, Gupta et al. (2001) use
instant cross-sectional data on 50 developing and transition countries to show that
the health status of the poor is substantially worse than that of the non-poor. On
average, child mortality is about 6 times higher for the poor than for the non-poor.
In sub-Saharan Africa, the mortality rate is seven times higher for the poor than for
the non-poor (Gupta et al.). According to an OXFAM report, the incidence of
poverty in Niger is 70% for families with uneducated parents, compared to 56% for
families with parents that have primary education.

The above results show that education and health play a leading role in
promoting economic development and improving the wellbeing of the poor. Indeed,
the relation between an individual’s education and his productivity (evaluated in
income terms) is one of the best established stylized effects in economics.
According to the World Bank’s World Development Report (1990) on poverty,
various studies conducted in the 1970s and 1980s proved that education raised
incomes and labor productivity in the agriculture and informal sectors. The report
indicates that the incidence of education investment on formal sector salaries is
still significant: in Indonesia, the rate of return in primary and secondary education
is above 10%; it stands at about 9% in Tanzania, and is higher in Madagascar: 10%
for formal and informal sector workers with primary education alone. With regard
to effects on labor productivity, the transition from 5 to 9 years of study in
Indonesia made it possible to raise production by 10%; in contrast, this effect
remains ambiguous in Madagascar’s agricultural sector.

Besides, the positive effects of good health on agricultural activity and salaried
employment have been observed systematically. The same report also mentions
the effects of better nutrition on agricultural labor productivity but gives no
information on the effects of healthcare. Meanwhile, several studies conducted in
Madagascar and Indonesia show the positive impact of healthcare access on
incomes. Disabled persons participate less in the labor market and earn less when
they find work because they often work in the informal sector. A survey in two

14
Indonesian provinces shows that if the consultation rate in health centers declines
the average salary for men and the female participation rate will decline. The
findings of these surveys all agree: any individual who is unable to afford treatment
for a curable disease suffers a great loss in income, which may be quite substantial
if he has to stop work.
In the long run, healthcare also has an impact on incomes: according to
Demographic and Health Surveys (DHS) conducted in 20 African countries, children
who are immunized, or whose mother’s had medical assistance during pregnancy
are in better health (taller). Hence, the capacity for future gains is correlated with
school performance, which in turn depends on health status.

2. Impact of the human capital of parents on that of children

Many studies have shown very close correlations between parents’ education
and the education and health of their children. However, a mother’s education
usually has a greater impact than that of the father (Leight, 1998; Behram et al.,
1997). Better educated mothers stand a greater chance of securing highly-paid
employment and are thus better able to handle schooling expenses. They are also
more aware of the benefits of education for their children. Moreover, educated
women have children who are in excellent health. Health has external effects on
education: a child’s school performance partly depends on his health (Martorell and
Habicht, 1986). Well-nourished pupils perform better.

The children of well-educated parents are more likely to attain a high level of
studies (Behram and Rosenzweig, 2005; Black et al., 2005). More educated parents
are better informed on the importance of their children’s health to which they pay
more attention. There are other advantages apart from these health benefits,
namely: better grades and, later on, higher salaries. In contrast, when parents are
illiterate their children enjoy less good health and this increases school
absenteeism and encourages early school drop-outs. Their children usually have
poor grades (Case, Fertig ; Paxson, 2005), and a limited capacity to earn higher
salaries. Besides, sick children will likely grow into sick adults (Cae et al., 2002).
The DHSs show that in most of the 20 African countries surveyed, a mother’s
education has a positive impact on the health and schooling of her child.

15
3. Impact of a mother’s education on demographic variables
Empirical studies conducted in many developed and developing countries
establish a strong negative correlation between a mother’s education and fertility,
and a positive correlation between a mother’s education, her child’s health and
her own health.

The impact of a mother’s education on population variables

The existence of a strong negative correlation between education, especially
women’s education, and fertility is an indisputable stylized fact. According to
Schultz (1995), empirical studies laid emphasis on “supply” factors (access to
contraception, family planning) as opposed to the Becker approach which
emphasizes “child demand”. Regardless of the approach used, the fact remains
that educating women has a negative and substantial impact on fertility. Studies
that separately measure the impact of various levels of education, showed that this
effect is only felt from the fourth year of study - that is, at (or towards) the end of
primary education when the child is able to read and write correctly.

Schultz estimates the aggregate determinants of fertility by combining data

from the UN, the World Bank and other international institutions on 80 countries
over 1972, 1982 and 1988 periods. He regresses the total fertility rate (TFR) over
the number of years of education for women and men, infant mortality and other
variables. He finds that a one year increase in women’s level of education reduces
the TFR from 0.161 to 0.551 children according to estimates, and that this effect
declines if infant mortality is factored into the analysis.

Ainworth et al. (1996) analyze the demographic patterns of 14 sub-Saharan

African countries3 in a study conducted on DHS data collected in the mid-1980s and
the early 1990s. These countries are characterized by a high fertility rate which
ranges from 5.0 (Botswana) to 7.4 (Niger) children, a low average level of
education for women (2 to 6 years, with a median of 0 in 6 countries), and very
limited access to modern contraceptive methods (less than 10% of women). They
find that a woman’s number of years of schooling has a significant negative effect
on fertility in all countries (except Burundi where the coefficient does not appear

3
The countries are Botswana, Burundi, Cameroon, Ghana, Kenya, Mali, Niger, Senegal, Tanzania, Togo,
Uganda, Zambia and Zimbabwe.

16
to be significant) although this effect is not linear: it increases with level of
education. In most countries, the effect becomes significant from the end of the
primary education cycle: the coefficient associated with four to six years of
schooling is significant in 8 countries (Botswana, Kenya, Mali, Niger, Senegal,
Tanzania, Uganda, and Zimbabwe); those associated with seven to ten years, and
11 or more years, are significant in all countries (except Niger for the last
category). They also find that the education of husbands also has a negative impact
on fertility and a positive impact on modern contraceptive use, although the effect
is limited compared to that of women’s education. According to the authors, the
disparity in these variable effects between countries stems from the differences in
school quality, access to contraception and family planning programs, and the
functioning of the labor market in sub-Saharan Africa.

Like Ainworth et al., other authors introduced variables on education quality

such as cognitive skills. Thomas (1999) finds that introduction of the results of
language and mathematic tests for 3rd or 4th year classes in South Africa, into a
regression of fertility over education quantity reduces by one-third the coefficient
associated with the latter, which still remains significant. They conclude that the
quality and quantity of women’s education both have a negative impact on
fertility. Similar results were obtained by Olivier (1999) for Ghana.

Impact of mother’s education on her own health and that of her child
A mother’s education has an effect on both her health and that of the child. Our
focus here is on the impact on child health.
Child health has today become a key economic development indicator. Of the eight
Millennium Development Goals adopted in 2000 by the 189 member-countries of
the UN, at least four relate directly to the health and nutritional status of children
(Todaro et Smith, 2005). Apart from being a development indicator, child health is
associated with other factors such as adult health, school performance,
productivity and income (Case et al., 2005). Of all the potential determinants of
child health, economists have focused on the education of mothers. Well-educated
mothers usually have healthy children since they have greater knowledge of the
care and nutrition suitable for the child and can provide a healthier and safer
environment for their children (Glewwe, 1999; Currie and Morettie, 2003).

17
 Studies that analyze the relationship between a mother’s education and child
health mainly focused on infant mortality and malnutrition.
Cadwell (1979) had already demonstrated that there is a very close link between a
mother’s education and infant mortality in Nigeria. This has since been confirmed
by Rosenweig and Schultz (1982) in their study on India. This impact increases
when household income, productive household possessions (radio, TV,
réfrigérateur, etc), and the household environment (availability of potable water,
toilets, access to electricity, etc.) are also taken into account. Besides, according
to an OXFAM report (Education Now: Break the Cycle of Poverty), the fact that a
mother has primary level education reduces infant mortality risks by 50% in the
Philippines; and these risks are reduced three times over if she has received
primary education.

A mother’s education also has a significant positive impact on the

anthropometric measurements of her children, notably “height by age” or “birth
weight”, which are excellent indicators of the child’s long-term development.
Thomas, Strauss and Henriques (1990) thus show that in the urban areas of
northern Brazil, children whose mothers completed primary school are 2.5 times
taller than those whose mothers are illiterate.
Moreover, a mother’s education positively affects child health production inputs
such as the number of antenatal consultations with a doctor, the probability of
obtaining better care for the baby and immunization for the mother. However,
these studies say little on how parents’ education, especially that of the mother,
affects children’s health. Desai and Alva (1998), Currie and Moretti (2002)
demonstrate that the impact of the mother’s education is predicated on the
stability of the marriage (the more educated the mother is, the more educated the
father will be also and have higher income). Blunch (2004), Kovsted et al. (2003)
demonstrated that a mother’s education facilitates her access to the media and,
consequently, to information on the health of her children.

Section 3. Education and health status in Africa

When they gained political independence in the 1960s, most sub-Saharan

African countries had education development indicators that left this region
trailing far behind the other continents. In 1960, the primary enrolment ratio was

18
only 36%, or about half the ratios for Asia (67%) and Latin America (73%). This ratio
was 38% in French-speaking territories (50% in the Belgian colonies and barely 30%
in French colonies) and 40% in English-speaking territories (Ko-Chih Tung, ).

The transition rate from one cycle to another remained low while the drop-out
rate remained high right up to the late 1980s. The secondary enrolment ratio in
Africa was 3% in 1960, compared to 14% in Latin America and 21% in Asia. In 1990,
the secondary education ratio in Africa was low, representing only 14%. The same
applied to higher education with 4% enrolment in all education structures.
In all, progress was made in the education sector in Africa in 1960-1990: total
enrolment in all three cycles of education (primary, secondary and higher
education) rose by 5% on average (UNDP, 1992).

However, since the Jomtien Conference in 1990, much progress has been made in
primary education, especially in sub-Saharan Africa. Indeed this region made
progress towards universal primary education (UPE) during the 1990-2001 decade
and during its last three years (1998-2001). Enrolment ratios have improved
significantly as compared to previous decades but remain low relative to other
continents. Despite the performance recorded, there are still wide disparities
between urban and rural areas and between men and women. Moreover, the
quality of education is mediocre. Public credits to the education sector in sub-
Saharan African countries are not used effectively and efficiently (Gupta,
Verhoeven, Tingson, 1999).

Performance since 1990

Sub-Saharan African countries made much progress in education in the 1990-
2001 decade. Some managed to reach the target of basic primary education, while
others came close, with disparities in school performance indicators in both cases.
The gross admission rate (GAR) rose during the decade from 70% in 1990 to 81% in
1998, representing an increase of 15%. This result is remarkable since it came in a
context of high population growth. The apparent admission rate for girls also
improved from 65.3% to 73.5%, but remains low compared to the rate for boys.

Primary enrolment for school-age children, measured by the net enrolment

ratio (NER) also increased. The NER rose from 54% in 1990 to 58% in 1998, and 63%

19
in 2001. Despite this increase, there were still more than 40 million out-of-school
primary school-age children (CONFEMEN, 2005). There was a slight improvement in
the gross enrolment ratio (GER) from 73% in 1990 to 75% in 1998. Only a few
countries registered a GER of 100% and above and a NER exceeding 90% (Cape-
Verde, Mauritius and Seychelles, among others). The disparity in results can be
attributed to differences in economic situation (favorable or unfavorable), political
situation (instability/stability) and government policies.

Countries experiencing conflicts and/or economic crises have a GER that is

below 100% and a NER that falls below 70%, or even 50% in the case of Eritrea,
Ethiopia and Niger. Meanwhile, government policies based on new EFA guidelines
may help to raise these rates. For example, the introduction of free primary
education in Malawi in 1994 doubled the GER between 1990 and 1998 from 64% to
128% (CONFEMEN, 2005).
The primary school drop-out rate has varied considerably in the sub-Saharan
African region. Many children drop out without completing the primary cycle which
is indispensable for acquiring basic knowledge either because of high school fees or
the need to work to supplement meager household incomes. The survival rates (up
to year 5) for primary school pupils was variable over the 1990-1998 period. In half
of the countries, the survival rate in primary year 5 was below 67%. Children living
in stable and prosperous countries have a greater chance of reaching the 5th year.
In Botswana, Cape Verde, Ghana, Mauritius, Namibia, Senegal, Seychelles, Uganda,
Zambia and Zanzibar, more than 80% of pupils reach this level, while 50% or less of
primary enrolment never reaches the 5th year in countries like Angola, the
Comoros, Congo, DRC, Ethiopia, Madagascar, Malawi and Mozambique (Ko-Chih
Tung,). In sub-Saharan Africa, only Seychelles and Mauritius have succeeded in
attaining the target of universal primary education.

Enrolment ratios still remain very low in the higher levels of education.
Although several countries have partly made secondary education compulsory, a
good number of primary school leavers do not pursue secondary education. In 2001,
the region’s GER was only about 27% for secondary education, although it exceeded
50% in South Africa, Botswana, Cape Verde, Gabon, Mauritius, Namibia and
Seychelles. In higher education, the enrolment was below 2.5% in half of the
countries that had data (CONFEMEN, 2005).

20
 Gender and zonal disparities
Education inequalities between boys and girls, and between rural and urban
areas are rampant. Enrolment is substantially lower for girls than for boys in
primary education where girls make up more than 50% of out-of-school children. In
primary schools, there are only 86 girls for every 100 boys. These disparities are
even wider in at the higher levels, with half of the countries having a gender parity
index (GPI) that is below 0.79 in secondary education. More than 60% of illiterate
adults in the region are women; the average GPI for illiteracy rates is 0.77, with
values that fall below 0.5 in some countries (Benin, Burkina Faso, Mali, Niger)
which also have the lowest literacy rates (CONFEMN, 2005). However, in certain
countries, there are improvements in gender parity. Seychelles and Maurice have
survival rates of 100% for girls. With regard to first-year primary enrolment, the
number of girls enrolled is higher than the number of boys in countries like
Mozambique and Mauritius.

There is also a major disparity between urban and rural areas in the supply of
education services that most sub-Saharan African countries have not yet corrected.
Gross enrolment ratios in urban areas, especially in the cities, are far higher than
in rural areas. Given this imbalance in the supply of services, it is absolutely
necessary for sub-Saharan African countries to substantially increase their first year
enrolment for girls and for rural areas where the majority of the poor live, in order
to boost economic growth.

Quality of education
Human capital production is not only measured from the number of children
having access to education, as determined by the EFA target, but also from the
quality of education received which makes it possible to enjoy the individual and
societal benefits accruing therefrom. However, in many countries, education has
developed without much regard for quality.

Education quality issues can be approached in two ways: first of all, by

considering a quality school to be one endowed with a lot of financial and human
resources (well-trained teachers, appropriate teaching equipment and aids in great
quantity, limited class enrolment, etc.); and secondly, by considering a quality

21
school to be one whose pupils have an excellent level of learning after spending a
certain number of years in the system.

Much research has shown that the quality of education on the continent,
especially sub-Saharan Africa, is mediocre. Most pupils complete their schooling
without mastering the knowledge and skills prescribed in the curriculum. The
CONFEMEN Educational System Analysis Program (CESAP) conducted a study on
school knowledge acquisition in six French-speaking African countries. The results
showed that 14% to 43% of 5th year primary school pupils had “insufficient”
knowledge of French and Mathematics. The CONFEMEN study(2000) on five French-
speaking African countries (Burkina Faso, Côte d’Ivoire, Cameroon, Madagascar and
Senegal) measure school knowledge acquisition in these countries and examines the
factors that characterized knowledge acquisition by pupils. In the five countries
concerned, the average value of pupils’ knowledge of mathematics and French in
primary classes 2 and 5 is 50.6, with 50 for Burkina Faso and Côte d’Ivoire, 58 for
Cameroon, 56 for Madagascar and 40 for Senegal. In 7 Southern African countries
that participated in the SACMEQ study (1995-1998), 37% of pupils in class 6 had
attained a “desirable” level in reading, while 22% to 65% were still at the
“minimum” level.

Health status
The health situation in developing countries, particularly in Africa, can be
assessed by measuring the progress made towards attainment of the health-related
Millennium Development Goals (MDGs). It should be remembered that three of the
eight MDGs relate to health: reduce infant mortality by two-thirds between 1990
and 2005 (MDG 4); reduce maternal mortality by two-thirds between 1990 and 2015
(MDG 5); halt and begin to reverse the incidence of HIV/AIDS, malaria and other
major diseases by 2015 (MDG 6).

Table 1A shows the proportion of the population living on less than $1 dollar a
day and the under-five mortality rate in 42 developing countries. This proportion
represents 75% of the local population. The data focuses on the year 2002 and
comes from the Demographic and Health Survey (DHS) conducted among the
households in these countries. The correlation between poverty and poor health is
obvious: for 79% of the population living above the poverty line (1$ per day and per

22
capita), the rate is 88-89 deaths per 1000 live births; for 27% living below the
poverty line, the rate is 156 deaths per 1000 live births, or about 75% higher than
that of those living above the poverty line. Another trend that emerges from the
table is the wide gap between sub-Saharan African (SSA) countries and other groups
of developing countries. The number of deaths is 204 pupils in SSA compared to 98
in Latin America and the Caribbean, 145 in the Middle East and North Africa and
127 in South and South-East Asia.

Another health status indicator that can be used is MDG 5 on the maternal
mortality rate. The world average is 400 per 100,000 live births, and it naturally
eclipses wide disparities between country groups. In SSA the average is 1000 per
100,000 live births, which is twice the rate of South East Asia, four times that of
Latin America and the Caribbean and 50 times that of developed countries (Graph
1A). The international community wants to reduce the mortality rate by 75%
between 1990 and 2015. In 200-2005, the overall fertility rate was 2.65 children,
corresponding to half of the rate for 1950-55. The result has been a decline in the
number of births in certain countries. It is projected that births in South Asia will
remain at 40 million per year between 1990 and 2015, although there are variations
between countries. Births are expected to decline in India from 27 to 24 million,
thereby contributing to a 9% decline in deaths. In SSA, births will continue to rise
by 43% during the same period (Ronsmans, Graham, 2006). Such fertility puts an
extra burden on overstretched maternity services, and so increases obstetrical risks
and consequently raises the number of maternal deaths. While there is hope of
attaining MDG 5 in other regions of the world, such is not the case in SSA.

Section 4: Levers for accelerating investment in people:

Education and health policies

The preceding section presented a brief description of the education and health
situation in developing countries. The conclusion is that much progress has been
made in all developing country groups since 1990 although performance has been
very variable. We will now examine the origin of these differences.

Over a long period, the wide variations recorded cannot be attributed to

differences in living standards alone. They stem from differences in the health and

23
educational policies pursued while economic policy differences account for
variations in per capita GDP growth rates that have a clear impact on health and
education. Education and health policies have a decisive influence on the
bottlenecks and progress in human capital accumulation in developing countries.
This crucial role played by policies will be illustrated through certain aspects:
financing, efficiency and effectiveness.

Financing
The financing of the education and health systems has a decisive impact on their
performance. In developing countries, especially low-income countries, the volume
of resources raised is an important determinant of access to education and health
services. The Macroeconomic Commission of the World Health Organization (WHO)
estimates that the amount of public spending that each low-income country has to
make to ensure availability of basic health services is US$34 per capita. But then,
most low-income African countries often allocate less than $10 per capita for
health and some like Ethiopia, Burundi and Democratic Republic of Congo (DRC)
spend less than $3 (Table 1A).

On the whole, the health sector is supposed to receive at least 10% of recurrent
public spending. Similar standards also exist in the education sector. Available
statistics show that most low-income countries are far from attaining these
minimum thresholds which constitute benchmark indicators for assessing the
relative volume of a country’s investments in education and health.

However, the volume of financing alone is not enough to determine the quantity
and quality of a country’s health and education services. The breakdown of
financing according to its various sources is equally important. In the health sector,
for example, there are four sources: taxes, social insurance, private insurance and
direct or cash (out-of-pocket) payments. According to the privileged financing
mechanism, the amount of health expenditure varies with an individual’s income
level. In OECD countries, the financing structure has major implications on equity
because it can lead to major differences in vertical equity (measure whereby
unequal income-earners contribute unequally to health expenditure) and
horizontal equity (measure whereby equal income-earners contribute unequally to
health expenditure) (Doorslaer, O’Donnell 2006).

24
 In low-income countries, the situation is different since the financing structure
depends on the use of health services and, consequently, on the distribution of
health services among individuals. If direct payments predominate, poor
households will not be able to afford health expenses above a certain threshold
amount, which is relatively low besides; this will not be the case for households
that belong, for example, to the last two quintiles.

By studying the case of 13 territories holding 55% of the Asian population,

O’Donnell, Doorslaer et ali (2005) analyzed the healthcare financing structure.
Certain lessons can be drawn from their study: (i) Private insurance plays a small
role compared to public and direct payments which together represent at least 30%
of health expenditure in all territories, except Japan. (ii) The share of direct
payments declines with a rise in development levels. In the poorest countries
(Nepal, Bangladesh, Kyrgyz, Punjab, and Sri Lanka), patients settle their health
expenses almost exclusively through out-of-pocket payments. (iii) The more
affluent households pay more but are the greatest beneficiaries of healthcare
services. (iv) Disaster health expenditure – that is, which forces households to
spend at least 10% of their budget on health contingencies – plunges millions of
people into poverty. Territories that depend most on direct payments have the
highest prevalence of direct payments and it is mainly non-poor households that
resort to this form of payment.

These conclusions show that the health situation of the majority of the
population depends on the balance existing between direct payments and pre-
payments (taxes, social security) within the health system. That is why the
financing policy is an important springboard for improving health or educational
status in low-income countries and constitutes a major reason for the differences
observed between low-income countries.

Palmer, Mueller, Gilson, Mills, Haines (2004) reviewed all the research
conducted since 1995 on the impact of different financing strategy experiments in
low-income countries. The following lessons can be drawn from such experiments:
(i) User fees constitute an obstacle to the use of health services. (ii) Community
prepayment schemes and micro-insurance systems provide opportunities for

25
improving access but have a rather limited scope. (iii) Conditional money transfers
certainly increase access to health services but may also encourage perverse
inclinations. (iv) The context within which a financing scheme functions is
important: conditional money transfers are experimented in Latin America while
payment of user fees is experimented in Africa. But then, to what extent such
experiments be transposed to a group of countries other than those where they
have already been applied?

Efficiency
A major characteristic of the efficient functioning of an educational system is its
capacity to adequately manage student flows all through their study cycle. Hence,
class repeating and dropout promote inefficient use of public and private resources
allocated to the education sector; they also translate into a rapid decline in access
rate as students progress in the various cycles. Africa has mid-cycle repeater rates
that far exceed those observed in other countries of the world. These rates are
particularly high in French-speaking African countries (20% on average) compared
to the rates in English-speaking African countries and in Asian countries whose
rates are below 10% (Mingat and Suchaut, 2000). Meanwhile, repeating encourages
dropout and frequent repeating and dropouts in the course of a cycle leads to
considerable loss of resources. About 40% of public resources are squandered in
French-speaking African countries, compared to 25% in English-speaking countries
(Mingat and Suchaut, 2000). Moreover, Saoussen (2006) finds that African countries,
on average, have the lowest education sector efficiency scores, compared to the
other sample countries. Public expenditure in these countries registered only 15%
efficiency. This low rate is reportedly due to the relatively high burden of public
spending in Africa in 1990-2002, representing about 4.8% of GDP, compared to 3.5%
in Asian countries and 4% in Latin American countries.

Effectiveness
Another way of measuring the capacity of an education or health system to make
good use of available resources is to compare the volume of allocated resources
with obtained results. By limiting analysis to the quantitative dimension of school
coverage, the average duration of schooling corresponding to 1% of GDP spent can
be used as an indicator of the effectiveness of an educational system. The amount
of routine public expenditure allocated to education, as a percentage of GDP,

26
measures the volume of resources received. The average duration of schooling is an
indicator of the quantitative performance obtained with these resources.
Graph 1 shows that many countries could have used the same volume of public
resources to provide their citizens with more years of schooling.

Graph 1: Share of education in GDP and number of years of schooling in some African countries

Source: World Bank, 2003.

For example, in Africa, countries like Zambia, Eritrea and Burundi where able to
provide their citizens with more years of schooling than Senegal which spent a lot
more than these three countries in relative terms. Other countries that allocate
the same share of their GDP to education as Senegal are able to provide their
people with a higher level of education. From an assessment of the relationship
between the number of years of schooling and the share of GDP devoted to
education, estimated for all African countries, it is calculated that with the 3.8% of
GDP spent annually on education, the number of years of schooling in Senegal
should have been 4.7 and not 4.1 in 2002 (A. Diagne 2004).

The relatively high level of unit costs, notably the salaries of teachers and health
staff, in a context of limited resources seriously undermines the results of the
27
education and health systems. However, the experiences of low-income countries
in this domain are very diverse and give rise to some interesting lessons. When
compared to per capita GDP, the salaries of teachers and health staff in African
countries are well above salaries paid in other countries with the same level of
development. While the salaries of primary school teachers in Asia and Latin
America fall below 3 units of GDP, they remain high in Africa with 3.6 for English-
speaking Africa, 6.3 for French-speaking Africa and 8.2 for Sahelian countries
(Mingat and Suchaut, 2000). These high salaries substantially reduce resources that
could be allocated to other expenditure crucial to education quality (teaching aids,
medication, refresher training, maintenance, etc.). Remuneration that is adjusted
to the per capita income will make it possible to free up additional resources for
school and health inputs that have a direct impact on services provided.

Meanwhile, these teachers and health workers who receive most of the allocated
resources as wages are often absent from their workplaces. Surveys, consisting of
unannounced visits were conducted in primary schools and health structures in
Bangladesh, Ecuador, India, Indonesia, Peru and Uganda (Chaudhury, Hammer,
Kremer, Muralidharan, Rogers, 2006). This entailed verifying onsite whether
teachers and health workers were present at work. The results indicated that, on
average, 19% of teachers and 35% of health workers were absent (tableau 1). The
authors point out that most of those present onsite were not working; in India, one
quarter of primary school teachers were absent and only about 50% of teachers
were teaching at the time of the visit. Another disturbing outcome is that
absenteeism is higher in the poor regions of the sample.

Source: Chaudhury, Hammer, Kremer, Muralidharan, Rogers, 2006.

28
Conclusion

This paper has demonstrated that there are many theoretical and empirical
arguments to prove that individuals, families and societies reap substantial gains
from investing in education and health. In fact, a lot of progress has been made
under the development SAPs in these two areas since the 1990s. However, the road
towards attainment of the MDGs is still long, especially for most sub-Saharan
African countries. Greater mobilization of resources will certainly be necessary.
However, the accumulated experience of many developing countries leaves no
doubt about the need to continue giving prominence to the design and long-term
application of sound policies in education and health. Public policymakers have
many mechanisms for improving the quantity and quality of educational and health
services with a given volume of financial, human and material means; such
mechanisms may be financing, intra-sectoral resource allocation, unit costs,
governance system, etc. It is through these same policies that they will succeed in
ensuring equitable access to quality education and health for all.

29
 Bibliography

Aghion, P. et Cohen, E. 2004. « Éducation et croissance, » La Documentation

française, Paris.
Ainsworth, M. K. Beegle and Nyamete, A. 1996. ‘The Impact of Women’s Schooling
on Fertility and Contraceptive Use: A Study of Fourteen Sub-Saharan African
Countries’, World Bank Economic Review, 10 (1), January, pp. 85-122.
Alba-Ramirez, Alfonso, and Maria-Jesus San Segundo. 1995. ‘The Returns to
Education in Spain’, Economics of Education Review, 14 (2), June, pp. 155-66.
Banque mondiale. 1990. Rapport sur le développement dans le monde, Washington
D.C.
Barro, R. “Economic growth in a cross section of countries”, Quarterly Journal of
Economics, n°151, pp. 407-443, 1991.
Barro, R. J. 1996. “ Health and Economic Growth. mimeo, Harvard University.
Barro, R. J. 1996. “Health and Economic Growth”. mimeo, Harvard University.
Barro, R. J. et J. Lee. 1994. “Sources of Economic Growth”, Carnegie- Rochester
Conference Series on Public Policy, 40, 1—46.
Barro, R. J. et X. Sala-í-Martin. 1995. “Economic Growth”, McGraw- Hill, New
York.
Barro, R. J., and Lee, J-H. 1994. “Sources of Economic Growth”, Carnegie-
Rochester Conference Series on Public Policy, 40 (0), June, pp. 1-46.
Barro, R. J., and Sala-i-Martin, X. 1995. “Economic Growth”, New York: McGraw-
Hill.
Barro, Robert J., and Xavier Sala-i-Martin. 1995. Economic Growth, New York:
McGraw-Hill.
Behrman, J R., and M. R. Rosenzweig. 2002. ‘Does Increasing Women’s Schooling
Raise the Schooling of the Next Generation?’, American Economic Review, 92
(1), March, pp. 323-34.
Behrman, J. and A. Deolalikar. 1988. “Health and nutrition,” in Hollis Chenery and
T.N. Srinivasan eds. Handbook of Development Economics, Volume 1,
Amsterdam: North-Holland.
Benhabib, J, and M. Spiegel. 1994. ‘Role of Human Capital in Economic
Development: Evidence from Aggregate Cross-Country Data’, Journal of
Monetary Economics, 34, pp. 143-73.

30
Benhabib, J. et M. Spiegel. 1994. “The role of human capital in economic
development : Evidence from aggregate cross-country data”, Journal of
Monetary Economics, n°34, pp. 143-179.
Bhargava, A., D. Jamison, L. Lau et C. Murray 2001. “Modelling the Effects of
Health on Economic Growth”, Journal Economics, 20(3), 423—440.
Bhargava, A., D. Jamison, L. Lau et C. Murray. 2001. : “Modelling the Effects of
Health on Economic Growth”, Journal Economics, 20(3), 423—440.
Black, S, Devereux.P and Salvanes, K. 2005. “Why the apple doesn’t fall far:
Understanding intergenerational transmission of human capital,” American
Economic Review 95(1), 437-449.
Bloom, D. E. et P. Malaney 1998. “Macroeconomic consequences of the Russian
Mortality Crisis”, World Development, 26, 2073—2085.
Bloom, D. E. et P. Malaney. 1998. “Macroeconomic consequences of the Russian
Mortality Crisis”, World Development, 26, 2073—2085
Bloom, D. E., J. Sachs et Williamson. 1998. “Geography, Demography, and
Economic Growth in Africa”, Brookings Papers on Economic Activity, 2, 207—
273.
Bosca, J., et A. de la Fuente et R. Domenech. 1996. Human capital and growth :
theory ahead of measurement, Mimeo.
Caldwell, J. C. 1979. ‘Education as a Factor in Mortality Decline: An Examination of
Nigerian Data’, Population Studies, 31 (1), pp. 5-27.
Case, A, Fertig. A, and Paxson.C. 2005. “The lasting impact of childhood health
and circumstance.” Journal of Health Economics 24, 365-389.
Case, A, D. Lubotsky and C Paxson. 2002. “Economic status and health in
childhood: The origins of the gradient,” American Economic Review 92(5), 1308-
1334.
Confemen. 2005 “Rapport mondial de suivi sur l’EPT”.
Chaudhury, Hammer, Kremer, Muralidharan, Rogers, 2006, “Missing in Action:
Teacher and Health Worker Absence in developing Countries”, Journal of
Economic Perspectives, Vol 20, Number 1, Winter, 91-116.
Currie, Janet, and Enrico Moretti. 2002. ‘Mother’s Education and the
Intergenerational Transmission of Human Capital: Evidence from College
Openings and Longitudinal Data’, Working Paper No. 9360, NBER, November.
De la Croix, D. et M. Doepke. 2003. “Private versus Public Education when Fertility
Matters”, Forthcoming in Journal of Development Economics.

31
Desai, Sonalde and Soumya Alva. 1998. “Maternal education and child health: Is
there a strong causal relationship?” Demography 35(1), 71-81.
Desai, Sonalde and Soumya, Alva. 1998. ‘Maternal Education and Child Health: Is
There a Strong Causal Relationship?’, Demography, 35 (1), February, pp. 71-81.
Dessus, S. 1998. ‘Human Capital and Growth: The Recovered Role of Educational
Systems’, working paper, World Bank.
________. 1998. Analyses empiriques des déterminants de la croissance à long
terme, Thèse de doctorat en sciences économiques, Université de Paris I.
Education: A Further Update’, Policy Research Working Paper No. 2881, World
Bank, September.
Gwatkin, D. R. 2005. “How much would poor people gains from faster progress
toward the Millennium Development Goals for Health?, Lancet, 365: 813-17.
Glewwe, P. 1997. “How does Schooling of Mothers Improve Child Health? Evidence
from Morocco”, document de travail n° 128, World Bank LSMS, Washington,
D.C.
Glewwe, P. 1999. “Why does mother’s schooling raise child health in developing
countries? Evidence from Morocco,” Journal of Human Resources 34(1), 124-
159.
Groot, W. & H. Maassen van den Brink. 2003. Investeren en Terugverdienen,
Rapport voor het Sectorbestuur Onderwijsarbeidsmarkt SBO, Den Haag.
Grossman, M. 1972. ‘On the concept of health capital and the demand for health’,
Journal of Political Economy 80, p. 223-255.
Hanushek, E.et D. Kimko. 2000. “Schooling, labor-force quality and the growth of
Nations”, American Economic Review, n°90, pp. 1184-1208.
Leigh, J. 1998. ‘Parents’ schooling and the correlation between education and
frailty’, Economics of Education Review 17, p. 349-358.
Lucas, R. E. 1988. “On the Mechanics of Economic Development”, Journal of
Monetary Economics, 22, 3—42.
Lucas, R. E. 1988. “On the Mechanics of Economic Development”, Journal of
Monetary Economics, 22, 3—42.
Mankiw.N G, Romer, D et Weil,D.N. 1992. “A contribution to the empirics of
economic growth”, Quarterly Journal of Economics, n°107, pp. 407-437.
Martorell, R. et J.-P. Habicht. 1986. “Growth in Early Childhood in Developing
Countries”, in Human Growth: A Comprehensive Treaty, Vol. 3:

32
 Methodology, Ecological, Genetic and Nutritional Effects on Growth, Plenum
Pub., New York.
Maurin, E., Thesmar, D. et Thoening, M. 2003. “Mondialisation des échanges et
emploi: le rôle des exportations”, Economie et statistique, n° 363-365, pp. 33-
44.
McNay, K., Arokiasamy, P. and Cassen, R.H. 2003. ‘Why Are Uneducated Women in
India Using Contraception? Insights from a Multilevel Model into the Changing
Relationship between Education and Contraception’, paper presented at the 7th
Oxford International Conference on Education and Development, Oxford, 9-11
September.
Nelson, R. et Phelps, E. 1996. “Investments in Humans, technological diffusion and
economic growth”, American Economic Review, n°56, pp. 69-75.
Oliver, R. 1999. ‘Fertility and Women’s Schooling in Ghana’, in Paul Glewwe, ed.,
TheEconomics of School Quality Investments in Developing Countries, St.
Martin’s: Macmillan, pp. 327-44.
Psacharopoulos, G. and Patrinos, H. A. 2002. ‘Returns to Investment in
Romer, D. 1990. “Endogeneous technical change”, Journal of Political Economy,
n°98, pp. S71- S102,.
Ronsmans C., Graham W. J., 2006, “ Maternal Survival 1. maternal mortality: who,
when, where, and Why, Lancet, Sept 28, 368: 1189-200.
Schultz, T. 1975. “The value of the ability to deal with disequilibria”, Journal of
Economic Literature, n°13, pp. 827-846..
Schultz, T. Paul. 1997. ‘Demand for Children in Low Income Countries’, ch. 8 in
Mark R. Rosenzweig and Oded Stark, eds. Handbook of Population and Family
Economics, Amsterdam: North Holland, pp. 349-430.
Siphambe, H. K. 2000. ‘Rates of Return to Education in Botswana’, Economics of
Education Review, 19 (3), June, pp. 291-300.
Solow, R. 1956. “Technical change and the aggregate production function”, Review
of Economics and Statistics, n°39, pp. 312-320.
Thomas, D. J. Strauss, and M-H. Henriques. 1990. ‘Child Survival, Height for Age
and Household Characteristics in Brazil’, Journal of Development Economics, 33
(2), pp. 197- 234.
Thomas, Duncan. 1999. ‘Fertility, Education and Resources in South Africa’, in C.
Bledsoe et al., eds., Critical Perspectives on Schooling and Fertility in the
Developing World, Washington: National Academy Press.

33
Todaro, Michael and Stephen Smith. 2005. Economic Development, ninth edition,
Pearson-Addison Wesley: Boston.
Unesco. 2000. « Education pour tous, Bilan à l’an 2000 : Document statistique ».
Van Doorslaer, E. , O’Donnell, Rannan-Eliya, A. Somanathan, et al., 2005. Paying
out-of-pocket for health care in Asia: Catastrophic and Poverty Impact “,
EQUITYAP Paper # 2, Erasmus University, Rotterdam and IPS, Colombo.

Van Doorslaer, E., O’Donnell, 2006. “Measurement and explanation of inequality in

health and health care in low-income settings, UNU-WIDER Conference on
Advancing Health Equity, 29-30, Sept, Helsinki.
Uzawa, H. 1961. “Neutral Inventions and the Stability of Growth Equilibrium”,
Review of Economic Studies, 28, 117—124.
World Bank. 1993. World Development Report 1993: Investing in Health, New York:
Oxford University Press.

34
 Annexes
Graph 1A: regional variation in maternal mortality ratio and number of maternal
deaths

Source : Ronsmans, Graham, 2006.

35
Tableau 1A : Population, poverty, and under-5 mortality in 42 developing countries

Sources : Gwatkin, 2005.

Table 2A: Sub-Saharan African countries that devote less than 15 dollars per person
per year.

Country A B C = B/A * 35

Public spending on Public spending on Public spending on

health per capita health as a % of health as a % of total
and at the official total public public expenditure,
exchange rate [US expenditure, [US necessary for
$], 2004 $], 2004 attainment of a health
expenditure of $35 per
capita per year
Senegal 15 9.6 22.40
Mauritania 14 11.7 29.24
Congo 13 4.3 11.57
Sudan 13 10.8 29.08
Cameroon 12 8.1 23.62
Zambia 12 10.0 29.17
Benin 11 8.3 26.40
Burkina Faso 11 12.6 40.09
Mali 10 9.2 32.20
Gambia 9 13.7 53.28
Côte d’Ivoire 8 4.6 20.12

36
 Kenya 8 7.3 31.94
Mozambique 8 9.1 39.81
Comores 7 6.9 34.50
Chad 7 9.5 47.50
Nigeria 7 3.5 17.50
Ghana 6 4.5 26.25
Tanzania 6 9.4 54.83
Uganda 6 9.4 54.83
Malawi 5 8.9 62.30
Central African 5 15.4 107.80
Republic
Togo 5 9.3 65.10
Niger 5 12.3 86.10
Madagascar 5 10.7 74.90
Rwanda 4 7.2 63.00
Erythrée 4 4.2 36.75
Sierra Leone 4 8.2 71.75
Guinea-Bissau 3 4.4 51.33
Guinea 3 4.8 56.00
Liberia 3 14.3 166.83
Ethiopia 1 4.9 171.50
Burundi 1 2.0 70.00
Democratic Republic 1 4.0 140.00
of Congo
Source: WHO (http://www.who.int/nha/country/en; consulté le 5 juillet 2006)

Table 3A: Some reproductive health indicators.

37
Source: ADB, 2006.

38