Education, development and the MDGs in
Sub-Saharan Africa – The foundations of educational policies and some
estimations on the benefits of post-primary education in Ethiopia
Christine Fauvelle-Aymar
CFEE –
cfauvelleaymar@yahoo.fr
Preliminary version – 28 mars 2008
Do not quote without permission
Abstract
This
paper examines the relation between education, and particularly post-primary
education, and development. A survey of the economic literature presents the
main approaches that have been used to justify the orientation given to the educational
public policies followed by developing countries, and particularly by low
income countries. The paper discusses the reasons that explain that these
policies have given priority to primary education to the detriment of other
levels of education. The second part of the paper is devoted to an empirical
examination of the influence of education on human development in
Keywords: Education, human development, the Millennium
Development Goals (MDGs), developing countries, Sub-Saharan countries,
Introduction
It is widely held that
education contributes to alleviate poverty and is essential for economic,
social and human development. Among the different levels of education, a common
opinion is that primary education is supposed to play a determinant role. As a
result, in developing countries, and especially low income countries, emphasis
has been put on programs aimed at developing primary education. National
development plans, poverty reduction strategies but also foreign donors
programs have all given the priority to primary education. Basic education and
literacy programs have benefited from the largest share of aid to education,
whereas the costs per student in primary are very low compared to those in
secondary and especially in tertiary level of education. This strategy has been
detrimental to the secondary and higher levels of education and, as this paper
will show, this orientation of education policies indirectly undermine the
effectiveness of the development strategies itself. In other words, the promotion
of primary education at the cost of higher levels is a counter productive
strategy.
This paper offers a
double perspective on this problematic. First, it presents the theoretical and
empirical foundations of the current educational policies followed by
developing countries. Education policies in developing countries are, however,
diverse, depending on the level of development of the country but also,
strongly, on its historical past. In order to avoid erroneous generalization,
this paper concentrates on the situation of Sub-Saharan African countries,
which shares many common features susceptible to impact their education
policies. A survey of the economic literature on the
benefits of education shows that the potential benefits of post-primary education
are unrecognised or at least under estimated. Therefore, that explains, at
least partly, the lack of interest given to the development of post-primary
education. The second part of the
paper is devoted to an empirical examination of the influence of post-primary education
on human development in a low income Sub-Saharan country,
1. The orientation of education policies: theories and
practices
After the Independences,
education, and especially higher education, received a lot of attention from
the new African rulers. Higher education was viewed as especially important for
elite recruitment, the development of a national bureaucracy (particularly to
replace foreign experts) and more generally the “building of a modern nation”
(Banya and Elu 2001).
In the 80’s, the
implementation of the structural adjustment programs in many developing
countries led to considerable changes in the orientation of public policies
and, in particular, to a drastic reduction in education expenditures. The
public financing of higher education was particularly criticized. Among the
most frequently advanced arguments were the inequitable characteristics of
enrolment in higher education, its high costs per students in country where a
lot of children were not even enrolled in primary school, the brain drain
issue… As Samoff and Carrol (2004) notes, “pressures to reallocate resources
from higher to basic education –in policy statements, in reports and other
documents, and in loan conditionalities [from the World Bank]- was often
accompanied by harsh criticism of existing African higher education
institutions as costly, unproductive and irrelevant”. Whatever the validity of
these arguments, they have strongly contributed, in association to tights
resources constraints, to the shift in the orientation of education policies.
During the last three
decades and particularly, after 1990 and the World Conference on Education for
All, the attention of international agencies and national government has mainly
focused on primary education. This strategy has been reinforced, in 2000, by
the implementation of the time-bounded targets of the MDGs, which regarding education,
concern only the primary level of education, except for the gender parity issue
(MDG 3). As a result, the post-primary levels of education have been neglected,
or at least, have not been given due priority. For instance, while education is
central in the Poverty reduction strategy papers (PRSPs), higher education is
rarely mentioned, and when it is the case, the specific contribution that
higher education to Africa’s development needs is generally not recognized (Bloom
et al. 2006). Secondary education is certainly more frequently mentioned in the
various documents and reports accompanying the definition and implementation of
development strategies in African countries. Nevertheless, its intermediary
position between primary education, the top priority, and higher education,
still a prestigious sector, explains that this sector is probably the most
neglected one, at least in the African countries which are far away from having
reached universal primary education and thus are not under strong pressures to
develop secondary education.
Different reasons can be advanced to
explain that post-primary education is not given due priority in the
educational agendas of African countries. Here, I would like to emphasis the
role of the economic research on education as it has
had and still has important implications for public policies toward education. Thus, this section provides
a survey of the main economic analysis of the benefits of education. The aim is
not to offer an exhaustive presentation of the literature[1] but mainly to demonstrate
the influence of these theoretical and empirical researches on the orientation
of educational public policies.
1.1. The rate of return analysis of
education
The rate of return
analysis of education (RORE) is certainly the oldest perspective on the
benefits of education and it is the main research priority of the economics of
education since the early 60s. This analysis is based on the human capital
theory, developed by Becker (1964) and Mincer (1974), which has clearly stated
the benefits that individuals receive from their investments in human capital
by the way of education. The core of the human capital theory, which adopts a
microeconomic perspective, lies in the thesis that education increases the
productivity of the individual.
The return to education can be measured, on an
individual or an aggregated basis, by comparing the stream of benefits of an
investment in education to its stream of costs. The literature distinguishes
the private and the social rate of return to education according to the
benefits and costs used to compute the rate of return.
George Psacharopoulos
is one of the major authors in this RORE literature. At regular intervals
during the last 20 years, he has produced comprehensive reviews of RORE in
developed and developing countries (Psacharopoulos and Patrinos 2002). Among
the main findings of these empirical studies, two have had important
implications for higher education in developing countries:
-
rate of return to education in developing countries
(and especially
- the social rate of return to education are highest for primary education, followed by secondary education and then higher education.
The main explanation
for the second result relates to the low cost per student of primary education
(compared to other levels) and the substantial productivity differential
between primary school graduates and those who are illiterate. And, even if
earnings of tertiary graduates are a lot higher than those of primary school
graduates, the cost of higher education is significantly more important.
These empirical
analyses had major implications for educational policy. Their results have been
used to justify the necessity to invest in primary education in developing
countries compared to investments in other level of education. In other words, the RORE analyses
have largely contributed to the emphasis put by governments, international
organizations and aid agencies, on primary education sectors to the detriment
of secondary and higher levels of education. This was certainly reinforced by
the fact that the main author of these empirical analyses, Georges Psacharopoulos, was an economist at the World Bank, the head
of the research section on education (Palmer 2006).
The empirical results
and the estimation of rates of returns to education have been largely debated.
On one hand, the calculation methods have been criticized. For instance,
Bennell (1996, p. 195) argues that “the conventional rate of return on education
patterns almost certainly do not prevail in Sub-Saharan Africa (SSA) under
current labor market conditions”. Many features of the African labor markets
are not taken into account in the RORE analyses and their inclusion can have a
significant influence on the value of the RORE. For example, most RORE analyses
are based on low or minimal opportunity costs for investment in primary level
of education. These costs are the income foregone by children attending primary
school, an income that cannot be assumed to be almost nil in rural societies
where children provide a significant contribution to family production. The
inclusion of this foregone income significantly decreases the RORs to primary
education (Bennell 1996). On the other hand, a number of recent studies on
education in developing countries and in particular in
Another main limitation of the RORE analyses is that
they adopt a narrow conception of the public benefits of education. The
"social" rates of return do not include the true social benefits, or
externalities generated by education. They only take into account the earnings
of educated individuals and the tax revenue collected on these earnings. This
measure of benefits does not reflect the external benefits generated by
investment in human capital that affect society as a whole but are not captured
by the individual. The example of the treatment of research activities, a
direct output of the higher education sector, is worth to be mentioned. “According
to RoR formulations, the only benefits of research are reflected in the
marginal earnings of college graduates, baccalaureate or post-baccalaureate.
This means that all benefits of research, such as rust-resistant wheat and rice
strains, hybrid seed corn, Salk vaccine, and the myriad other university-based
discoveries and inventions have no economic or economic-proxy value to the
larger society, except as reflected in the taxes paid by those who make the
discoveries and inventions” (Leslie 1990, p. 279).
As RORE analyses do not estimate the true benefits of
education, they are not the most appropriate device to inform policy makers of
the potentiality of investing in the different levels of education. Nevertheless,
over the past 30 years, studies of rates of return to higher education have
impacted policies toward education more than any other results produced by
researchers (Leslie 1990). As a result, today there is a strong presumption
among many policy makers that secondary and higher education are less important
for development and that developing countries should better invest their low
resources in literacy programs and primary education.
In addition to these
RORE analyses, many studies have tried to directly estimate the impact of
education on productivity. Concerning developing countries, a lot of them have
concerned the agricultural sector since it is the main sector of activities in
many low income countries. The most cited paper is certainly the survey of
Lockheed et al. (1980) on the influence of farmers’ education on agricultural
productivity. The
authors’ conclusion that 4 years of schooling was a critical threshold to reach
in order to increase agricultural productivity has been afterwards used
extensively as an argument to demonstrate the potential impact of basic
education on development. Many publications issued by international
organizations and in particular by the World Bank have quoted this conclusion
and therefore have strongly contributed to validate the claim that 4 years of
education were a minimum threshold to be reached by developing countries[2] (King and Palmer 2006).
1.2. The endogenous growth theory and
the economic externalities of education
The former analyses adopt a microeconomic perspective
and consider only the impact of education on labor productivity, either at the
individual or social level. But if education impacts productivity, it
necessarily has consequences for the economic growth rate. Over the last four
decade, a large body of literature, the endogenous growth theory, has examined
the role of human capital in determining the level and rate of growth of GDP
per capita.
Compared to the
traditional models of economic growth (Solow 1957), the endogenous growth
models develop a broader perspective on the benefits of education. In addition
to the direct influence of education on labor productivity, these models take
into account the indirect effect on productivity, and therefore on growth, via
the externalities of education (Lucas 1988). For instance, educated workers can
raise the productivity of their less educated co-workers. They can contribute
to change the production technology, they can make it easier to import and
adapt foreign technologies… These spillovers and related external effects are
potentially very important and can add considerably to the macroeconomic
consequences of the initial investment in human capital. The endogenous growth
theory has also put the emphasis on the essential role played by research and innovation
(Romer 1990).
This theoretical literature
on economic growth has given rise to a huge number of empirical studies trying
to identify the statistically significant factors influencing economic growth.
Most of these studies have found that education is one of the most significant
determinants of cross-country differences in long-run growth rates. A higher
level of initial stock of human capital is correlated with a higher level of
economic growth. For instance, Barro (1991) first empirical analysis showed the
positive impact of primary and secondary enrolment rates in 1960 on the GDP
growth rate over the period 1960-
Some empirical studies
have also tried to distinguish the influence of the different levels of
education. But on this issue, the results are not really conclusive. Barro (1991,
1997) found that secondary education was more influential than primary level. The
study of Bloom et al. (2006) show that tertiary education can have an important
impact on the promotion of faster technological catch-up and thus can improve a
country’s rate of economic growth. On the other hand, Gemmell (1996) found that
primary human capital has the most impact in the less developed countries
whereas secondary human capital was more important for the intermediate group
of less developed countries. Evidences concerning the specific influence of higher
education are particularly limited. One of the main reasons is simply the lack
of data, a problem that is more severe in the case of higher education
statistics, especially in the context of developing countries.
Moreover, most empirical studies have used general indicators of education such
as enrolment rate or average years of schooling. If the enrolment rate in
primary or secondary education is certainly a good proxy of the stock of
primary or secondary human capital, this is less true for tertiary enrolment
ratio. The output of the higher education sector is much more diverse, as it
can produce such diverse human capital as social scientist, health workers,
engineers, etc. On this issue, Lin (2004) provides an interesting empirical study
which investigates the effects of the different tertiary academic fields on
economic growth in
The endogenous growth theory has contributed to
emphasis the importance of investing in education for economic growth and
therefore for development. It has greatly participated to the renewal of public
policies, and particularly policies aimed at improving human development, that
have followed the difficult time of the structural adjustment programs. Nevertheless,
the endogenous growth theory does not provide any robust conclusion concerning
the respective influence of the different levels of education. Therefore, if the
endogenous growth theory has provided many justifications for investments in
education, it has not led to a reorientation of these policies towards the
secondary and higher levels of education. Investments in education were
increased but the bias towards primary education was maintained.
1.3. Education and human development
The RORE analyses and
the endogenous growth literature focus on the economic or monetary benefits of
education. Education, directly or indirectly (through spill-over effects) enhances
economic output through its effect on factor productivity. But, education can
have also more direct impacts on alleviating poverty and on the different
dimensions of human development. It generates social or non-monetary benefits. These
social benefits of education have been largely neglected in the economic
literature. However, they are probably the most important way by which
education can contribute to improve the various dimensions of human
development.
First, education
decreases the risk for individual of being in situation of poverty. This is a
logical result since one of the main reasons that explain that individual invests
in education is the expectation of higher earnings. But, in addition to this
direct influence on income, education contributes also to break the vicious
circle of intergenerational transmission of poverty (UNESCO BREDA 2007). Many
empirical studies have shown that the parents’ level of education significantly
impact the educational attainment of their children.
Secondly, increasing
schooling is related to better health and increased life expectancy for the
educated persons (Owens 2004, McMahon 2000, Tilak 2006). The main reasons are
linked to the occupational choices, to better understanding of medical
information, to the adoption of better health related behaviors… Another
important impact of education on health occurs through intergenerational externalities.
The health benefits of education are transferred to children as there is a
strong relationship between the educational attainment of mother and the
children health status. Better-educated mothers are more aware of good health
and nutrition practices, are more inclined to immunize their children… (Owens
2004).
Since education
contributes to reduce infant mortality rate and more generally to improve the
health of the population, it has a positive effect on population growth.
However, at the same time, the education of girls reduces fertility rates by
delaying the age of first birth but also the number of children. This effect,
contrary to the first one, contributes to reduce the rate of growth of the
population. The final result depends on the dominant effect. It is only when
the fertility effect becomes dominant that one can notice a decrease of the
population growth and the start of a demographic transition. According the
McMahon (2000) it is only when women achieve at least 9 years of schooling that
the fertility becomes the dominant factor.
Since declining rates
of population growth are generally seen as likely to boost the growth of per capita income,
education has, through its impact on demography, an effect on growth. More
generally, by improving human development, education has several feedback
effects on productivity and growth[5].
These non-monetary
effects of education are, by definition, more difficult to assess than the
market effects. Moreover, the few number of studies that have examined the
impact of education on poverty, health and population growth, have concentrated
on analyzing the impact of basic and primary education (Tilak 2006) or have
considered only developed countries (McMahon 2000). Therefore, if the results
of these studies clearly justify the benefits associated to investing in
education, they do not offer any advice concerning the allocation of
investments between the different levels of education.
1.4. Education and the shortage of
skilled professionals
Post-primary education,
and especially higher education, is essential to train the various skilled
professionals needed to promote development. Most developing countries are
facing an acute shortage of a wide range of professionals, particularly in
areas essential to the achievement of the MDGs. The benefits related to the elimination
or, at least, the reduction of these shortages is difficult to evaluate but
these professionals are a direct input to any development strategy.
For instance, many
developing countries know a crisis of human resources for health. The situation
of the African countries is particularly worrying. Out of the 57 countries
experiencing critical shortages, 36 are located in
Another major
challenge facing developing countries is to train the very large number of
teachers required in both primary and secondary schools, taking into account
the high attrition resulting from a variety of factors, including HIV/AIDS. According
to UNESCO (2006a) estimations, in
Well-trained
professionals are also critically needed in the areas of policy
formulation, policy implementation, research and development, engineering,
technology and many other specialized areas of development process. The implementation and working of a
development-oriented state needs strong administrative capacities and an
efficient and accountable public service. This is critical for designing and
implementing sound economic and social policies, managing the public sector,
regulating private sector and providing key public services. Success in
implementing effective strategy for development requires, among other important
factors, the training of professional administrators and specialists in
economic and social policies. It is particularly important for African
countries to be able to compete in the global world and the knowledge economy.
Failing to raise the educational attainment of their population will not only
cost developing countries in terms of human development and foregone growth, it
will also widen the gap between them and developed countries.
In sum, human capital
is essential to strengthen the capacity of developing countries and to foster
their way towards the achievement of human and economic development. Developing
countries have to invest in human capital. Primary education cannot provide
these highly qualified professionals. The production of this human capital
requires the development of a post-primary level of education.
1.5. Economic research and the
orientation of education policies
Many branches of the
economic literature have analyzed the way education can contribute to
development. However, the necessity to invest in post-primary
levels of education does not emerge as a one of the main conclusions of this
literature. The problem is that this result does not come from an in-depth
examination of the benefits generated by post-primary levels of education.
Indeed, it mainly follows from the fact that most analyses have focused on
primary level of education. The only exceptions are the RORE analyses which
clearly but certainly wrongly emphasis the superiority of investments in
primary education. The other types of studies have generally neglected to
examine the particularity of the different levels of education. Therefore,
there is in-depth and clear investigation and evaluation of the
benefits of post-primary education in terms of economic
growth, human development and of its contribution to meeting the MDGs. As a
result, the potential benefits of investing in post-primary education have been
unrecognised or at least under estimated. That represents a serious lack in the
economic researches on education.
The most important consequence of this situation is
certainly that the economic
research on education has had and still has important implications for public
policies toward education. The absence of clear results concerning the benefits
of post-primary level of education has incidentally contributed to explain the
emphasis put on primary education in developing countries, and especially in
low income ones. This is not to deny that investments in primary education are
not important. But the lack of priority given to secondary and post-secondary
levels of education is detrimental for development, even if also no study has
ever tried to estimate the consequences of promoting primary education to the
detriment of secondary and higher levels of education.
2. Education and human development in Ethiopia
This section tries, in
the case of
Many indicators can be
used to monitor the situation regarding human development. However, the targets
of the MDGs provide a relatively complete perspective on the various dimension
of human development. This section presents some recent data concerning MDGs in
2.1. Education
The education related MDGs
Two MDGs refer to
education. The first one is MDG 2 whose aim is to achieve universal primary
education by 2015. The second goal is MDG 3 which states that developing
countries should also aim at eliminating gender disparity in all level of
education by 2015.
According to the
Ethiopia MDG report (MOFED 2004),
The PASDEP document is
the only one to provide details concerning primary completion rate. According
to it, completion rate is 57.4% for grade 5 and only 34% for grade 8 which
corresponds to a completed cycle of primary education. This result confirms
that a rise of enrolment does not always translate into a rise of completion
rate (UNECO 2007) and it is at least partly explained by the fact that increase
in enrolment in primary education has mainly occurred through a rise of the
pupils/teacher ratio. This ratio is now on average
Concerning the gender
parity situation in education,
The other levels of education
The MDGs concern only
the primary level of education (except for the gender parity issue).
Nevertheless, as this study is concerned with the impact of the other levels of
education, a look at their present situation is necessary.
If the situation
regarding participation in primary level of education is encouraging, the
situation in the two other levels of education is all the most worrying. In
2000, less than 20% of the correct-age population was enrolled in secondary
level of education, and even less than 10% in the rural parts of the country
(MOFED 2004). The situation improved over the last few years. Nevertheless, in
2005, the GER in the first cycle of secondary level (grade 9-10) is still only 29.3%,
a very low ratio compared to a GER in secondary level (full cycle) of 60% on
average in developing countries and 32% in Sub-Sahara African countries (UNESCO
2007). There is no clear data concerning the upper-secondary level of education
(grade 11-12) in
The participation at
the tertiary level of education is about 1.5% in 2005 (PASDEP 2006), a ratio
which ranks
The influence of education on the probability
of enrolment in primary school (MDG 2)
The DHS data offers
the opportunity to assess the influence of the parents’ level of education on
enrolment and completion at primary level of education.
I considered two
indicators:
- the enrolment
indicator which is a dummy variables equals to 1 if a child is enrolled in a
primary school (and equals to 0 else). The typical enrolment age in
- the completion indicator
which is a dummy equals to 1 if a child who is between the age of 15 and 18
have completed a primary course of education.
There are 3350
observations in the sample used to examine primary enrolment and the rate of
enrolment calculated on this sample is 54%. The rate of completion calculated
on the sample of 1633 children aged 15-18 is 42%. These data conforms
relatively to those presented above.
The empirical study
examines the impact of the head of household level of education (measured by
the number of years of education) on these two indicators. However, in addition
to this explanatory variable, other control variables are introduced in the
estimation, such as the level of wealth of the household[13], its localization (rural,
urban or in the capital[14]), and the child gender. This
empirical analysis, as those that will be presented later in the paper, does
not pretend to be an in-depth study of the determinants of enrolment and
completion in primary school. The idea is to take into account the most evident
important determinant in order to run an econometric analysis that can be
considered as fairly robust[15].
The table 1 presents
the estimation results. As one can see, the head of household’s years of
education significantly impact the probability that a child will be enrolled in
primary school and will complete this level of education.
|
Table
1: Results of the logit estimations |
||
|
|
Probability of enrolment |
Probability of
completion |
|
Years of education |
0.085*** |
0.182*** |
|
|
(4.67) |
(6.37) |
|
2nd wealth quintile |
0.759*** |
0.861*** |
|
|
(6.91) |
(3.68) |
|
3rd wealth quintile |
1.154*** |
0.719** |
|
|
(10.00) |
(3.00) |
|
4th wealth quintile |
1.541*** |
1.367*** |
|
|
(12.88) |
(6.16) |
|
5th wealth quintile |
2.270*** |
1.978*** |
|
|
(12.86) |
(8.30) |
|
Male |
0.072 |
0.480*** |
|
|
(0.92) |
(3.56) |
|
Rural |
-0.828*** |
-1.519*** |
|
|
(4.14) |
(-7.32) |
|
Capital |
1.294* |
|
|
|
(2.12) |
|
|
Constant |
-0.259 |
-0.957*** |
|
|
(-1.20) |
(-3.48) |
|
Wald Chi2 |
545.567 |
385.461 |
|
Prob>Chi2 |
0.0000 |
0.0000 |
|
Pseudo R2 |
0.1817 |
0.3142 |
|
Number of Cases |
3350 |
1633 |
As can be expected,
the probability of enrolment increases with the wealth of the household[16], the localization in
urban areas and particularly in the capital. The influence of these variables
is similar concerning the probability that children have completed a primary
course of education, except for the “capital” dummy which is not significant in
the completion estimation[17]. The main difference
between the two estimations concerns the “male” dummy variable. One can notice
that the coefficient associated to the “male” variable is not significant in
the enrolment equation whereas it is significant and positive in the completion
equation. These results indicate that there is no significant difference in
terms of primary enrolment according to the gender of the child, which is
conform to the general situation of the country presented before[18]. They also indicate that
there is a gender difference concerning the probability of completion. The completion variable used in the
estimation measures the probability of completion of child aged 15 to 18.
Therefore, they are the outcome of primary education a few years ago (when
these children were enrolled in primary school) and, as noted before, the GPI
was very low at that time in
The figures below
present the relation between the estimated value of these dependent variables
and the number of years of education of the household head (taking into account
the other explanatory variables). The figure 1 concerns the enrolment variable
and the figure 2 the completion variable.
Note
: This figure presents the relationship between the estimated probability of
enrolment (the predicted value of the estimation presented in table 1) and the
level of education of the household head (taking into account the other
explanatory variables). There is no head of household with 14, 16 and 18 years
of education in the sample.
Note : This figure presents
the relationship between the estimated probability of completion (the predicted
value of the estimation presented in table 1) and the level of education of the
household head. There is no head of household with 14, 16 and 18 years of
education in the sample.
One can clearly notice
that the probability of enrolment and the probability of completion increases
with the education of the parents. The relationship is particularly significant
in the rural areas. While the probability of being enrolled in primary
education is about 41% in the rural areas when the head of household has no education,
it increases to 62% after 4 years of education (which is considered as an
important threshold), to about 72% at the end of the primary level of education
(7 years) and to 80% with a completed secondary cycle of education (13 years).
The influence of
parents’ education is less pronounced in urban areas at least concerning
enrolment in primary school. However, concerning the probability of completion,
the relation holds in both types of environment. The probably of completion is
about 20% in rural areas when the head of household have no education (and
about 76% in urban areas). It increases to 48% when he (she) has completed 4
years of education, to 62% after a full cycle of primary education and to
almost 90% when he(she) has a secondary level of education (the respective
value are 88%, 91% and 98% in urban areas).
The pursuit of
education after a secondary level seems to have an additional effect, at least
for the completion rate and in urban areas. The probability of completion in
urban areas reaches almost 100% when the head of household has 18 years of
education. However, as one can notice on
the figure 1 and 2, the data do not provide much information on the impact of
post-secondary education. There is a very few numbers of households having more
than a post-secondary level of education, in
This empirical
analysis clearly shows that:
- the impact of parents’
education on child level of education is essential.
- the influence of
parents’ education does not reach its maximum after 4 years of education and
even not after a full course of primary education.
In other words, the
achievement of universal primary education (MDG 2) cannot be reach without
investing in post-primary levels of education.
2.2 Poverty and hunger
The MDG 1
The objective of MDG 1
is to eradicate poverty and hunger. More precisely, the targets are to halve
between 1990 and 2015 the proportion of people whose income is less than one
dollar a day and the proportion of people who suffer from hunger.
According the Ethiopia
MDG report (MOFED 2004, p. viii), the threshold of one dollar a day is not well
adapted to the Ethiopian context, since about 89% of the population fall below
that income level. Using a more pertinent definition (a poverty line based on a
basket of food items consumed by the poor), the report estimated that, in
2004-2005, about 40% of the population was poor (head count index). Despite the
economic growth rate, there was almost any improvement over the last 10 years
due in particular to a rise in income inequality. This present situation
suggests that
The situation
regarding the goal of reducing hunger and its recent evolution is more
encouraging. According the Ethiopia MDG report, the proportion of stunted
children[19]
has declined in the recent years from a percentage of 66.6% in 1995 to about
56.8% in 2000. However, the very insecure and volatile situation of
The influence of education on poverty and
hunger
The
DHS data offers the possibility to examine the influence of education on
poverty and hunger. The indicators used to measure poverty and hunger are the
following:
-
the
probability for a household of being in the poorest wealth quintile (using the
DHS wealth index)
-
the
probability for a child aged 0- 5 of being underweighted.
The measure of poverty
that can be defined using DHS data is not based on income data since there is
no direct question about income in the DHS questionnaire. However, the survey
offers an indicator of the households’ relative economic status, the Wealth
index. This index is generated with a principal component analysis of data
concerning households’ ownership of selected assets, such as televisions and
bicycles, materials used for housing construction, types of water access and
sanitation facilities and other indicators of the economic status[20]. A factor score is
generated for each household. These scores are standardized and used to define
the break points of the wealth scale. Each household is then assigned to a
wealth quintile, from the poorest to the richest.
The measure of hunger
retained in this analysis is the most-used one. It corresponds to the
probability for a children aged 0 - 5 years of having a weight for age which is
less than minus two standard deviations from the median of the international
reference population ages 0 - 59 months.
The table below
presents the results of the econometric analysis. A certain number of control
variables have been introduced to take into account the other potential
determinants of poverty and hunger.
|
Table 2: Results of the logit estimations |
||
|
|
Probability of poverty |
Probability of hunger |
|
Years of education[21] |
-0.247*** |
-0.090*** |
|
|
(-14.40) |
(-5.21) |
|
2nd wealth quintile |
|
0.295* |