Department of Agricultural, Food, and Resource Economics Food Security Group

IDWP 116. Poverty, AIDS, Orphanhood, Gender, and Child. Schooling in Sub-Saharan Africa: A Review of the Evidence. David Mather. November 2011

EXECUTIVE SUMMARY:
There is growing concern that the HIV/AIDS epidemic may reduce long-term human capital
development through reductions in child schooling in SSA, thus severely limiting the longterm
ability of orphans and their extended families to escape poverty. In response, some have
called for targeted schooling subsidies for orphans and other children made vulnerable by
HIV/AIDS, on the assumption that such children are under-enrolled. This paper provides an
overview of the data sources used by existing empirical studies that test for orphan schooling
deficits and the methodological challenges that they face. It then reviews the empirical
evidence on the effects of orphan status or adult mortality on child schooling, as well as the
prevalence of orphans in SSA and their living arrangements.

As portrayed by media and some donor reports, the term ‘AIDS orphan’ often elicits the
image of a child who has lost both parents to AIDS, and who either lives in a child-headed
household, a household headed by grandparents who cannot afford to send them to school, or
in an orphanage. While children in such circumstances are very likely to be vulnerable to
poverty and poor schooling outcomes, empirical evidence shows that in most SSA countries,
about 90% of orphans have a surviving parent, that a majority of these single-parent orphans
live with their surviving parent, and that child-headed households are extremely rare. While a
few countries (Kenya, Malawi, Namibia, Zambia, and Zimbabwe) continue to experience
growth in their overall orphan rates (as well as having the highest rates of double-orphans),
orphan rates have remained relatively stable in most SSA countries in the past decade (Beegle
et al. 2010). In addition, while on average one in six households with children in SSA are
caring for orphans, the evidence to date suggests considerable resilience among extended
families in absorbing orphaned children. However, there are signs that this is becoming more
difficult in some countries (those with rapidly increasing orphan rates), as an increasing
number of double-orphans and single-orphans not living with a surviving adult are living
with grandparents who tend to be relatively poor.

The findings of the studies reviewed in this paper demonstrate that there is considerable
heterogeneity in the effects of orphan status or adult mortality on child schooling in SSA. The
bulk of the evidence demonstrates that the extent to which orphans are under-enrolled relative
to other children is country-specific, and very often specific to certain kinds of orphans. For
example, several of the multi-country studies find that while orphan schooling deficits are
relatively large and statistically significant in some countries, in other countries orphan
schooling deficits are relatively small or not statistically different from zero. In addition,
children who have lost both parents are considerably more likely to have statistically
significant schooling deficits (and of larger magnitude) than single-parent orphans. Among
the countries where an orphan schooling deficit appears to exist, the existence and magnitude
of schooling deficits sometimes vary considerably by characteristics of the child (gender), the
deceased adult (gender or household position), and the household (wealth level). For
example, in some countries, schooling deficits are only found among female and not male
orphans, among maternal and not paternal orphans, or among orphans in relatively poor
households but not those from wealthier households. In addition, in most countries, the
gender schooling gap is not worse among orphans.

Second, the findings from three large multi-country studies demonstrate that household
wealth is a much better predictor of poor child schooling than orphan status in most SSA
countries. For example, these studies find that school attendance gaps between poorer and
wealthier non-orphan children are two to three times larger than single-parent orphan
schooling deficits in most countries, and in most cases as large as the double-orphan deficit.
Third, it is clear from several of the studies that the school enrollment of orphans relative to
non-orphans may change significantly over time within a given country, either for the better
in the case of abolition of primary school fees, or for the worse if orphan rates increase
dramatically over time. Fourth, the studies based on panel data also show that negative effects
of orphan status or adult death on child schooling may occur during the pre-death illness
period, after the death of the parent or adult, or both.

There are several policy implications from these results. First, because the extent to which
orphans are under-enrolled relative to other children is country-specific, social protection and
education policymakers concerned with primary school under-enrollment need to tailor
mitigation measures to the specific needs and situation of each country. Second, the results
also imply that it is inappropriate to categorize all children who are directly or indirectly
affected by HIV/AIDS-related morbidity and mortality as being especially vulnerable and in
need of targeted school subsidies. Use of orphan or OVC (orphans and other vulnerable
children) status alone as an indicator of poor schooling is often inappropriate, as the children
facing the biggest schooling deficits in many SSA countries are double-orphans and children
from the poorest 20-40% of households.

Third, in countries with relatively low to medium levels of primary school enrollment,
targeting children from poorer households with schooling subsidies should improve the
enrollment and schooling progress of children most likely to suffer from poor schooling, both
orphan and non-orphan alike (Ainsworth and Filmore 2006). Some countries have already
gone further than this by eliminating primary school fees for all children. Evidence from
Malawi and Uganda suggest that improvements in enrollments among the poor through
universal abolition of primary school fees can substantially raise the enrollment of orphans,
even to the point of eradicating orphan schooling deficits (ibid. 2006). In addition, the most
recent DHS data show that orphan attendance rates are nearly on par with non-orphans in
many SSA countries, and that double-orphan schooling deficits in many countries have fallen
dramatically (UNICEF 2010). Future research could usefully document these enrollment
improvements and investigate the role of abolition of primary school fees relative to subsidies
targeted to orphans in explaining the apparent improvements in primary enrollment among
orphans.

Fourth, even in countries that have abolished primary school fees, there may still be barriers
to enrollment such as continued household demand for child labor, additional educational
expenses for transport, school uniforms and books, and declining school quality if enrollment
outpaces new school construction and teacher hiring. These additional barriers to enrollment
may explain why adult mortality was found to result in child schooling losses in Zambia and
Mozambique, even during a period after those governments had abolished primary school
fees. In addition, targeted schooling subsidies alone may not reduce schooling deficits of
some orphans, in the event that their poor schooling progress is due to the emotional and
psychological trauma of losing one or both parents or a lack of interest by their adult
guardians in their schooling.

Fifth, the timing of the negative effects of adult mortality on child schooling implies a
potential dilemma with respect to targeted assistance to orphans. On the one hand, in many
countries, it would likely be inefficient for policymakers to target school subsidies to singleparent
orphans, as many of these children would likely attend school in the absence of a
subsidy. Yet, if policymakers were to target assistance instead to only children who had lost
both parents, such children would likely have already incurred schooling losses that might be
difficult to make up later. In a situation where school fee abolition is not possible for either
the poorest children or universally, policymakers intent on targeting schooling subsidies to
orphans may thus want to focus on two-parent orphans, as well as single-parent orphans in
poor households where the surviving parent is either non-resident or chronically ill.

Finally, it should be noted that because of the well-established positive correlation between
educational attainment and safer sexual behavior (World Bank 1999), Education for All is
itself an important policy that can help reduce the spread of HIV/AIDS.