Over the last two decades, Uganda has registered impressive economic growth accompanied
by a significant reduction in income poverty, from 56 per cent in 1992 to 25 per cent
in 2010. Consequently, the country has met the first Millennium Development Goals
(MDGs) target of cutting the poverty level in half by 2015. However, with less than two
years until the deadline set by the international community to attain the MDGs, the
recent Uganda MDG progress report gives evidence that this accomplishment has not
been accompanied by improvements in other important areas. It shows that Uganda’s
performance on child-related MDGs indicators such as nutrition, health and education,
remains ‘off-track’ (MFPED, 2010). In this way, the Uganda MDGs progress report
for 2010 affirms that progress in one dimension of human well-being is not necessarily
associated with improvements in others, and that to be comprehensive, any assessment of
human deprivation needs to be done in a multidimensional framework. Therefore, this
study aims to examine the incidence, intensity and determinants of multimensional child
poverty in Uganda.
In this study, a child is defined as any individual below 18 years of age in accordance
with Uganda’s Constitution. Children are not a homogeneous group, children in different
stages of childhood suffer different forms of deprivations and vulnerabilities and the impact
of these vary across different age groups. Consequently, a three-step analysis of child
poverty is performed, constituting the three empirical chapters of this study. The first
empirical chapter applies the Alkire and Foster (2007, 2011b) counting and dual cut-off
approach to the 2006 Uganda Demographic and Health Survey (UDHS) to generate a
robust profile of multidimensional child poverty among pre-school children. The second
iii
empirical chapter extends the first analysis by analysing the changes in multidimensional
child poverty among schoolchildren in Uganda between 2002 and 2010 using Uganda
National Household Survey (UNHS) data sets. The third empirical chapters provides
an econometric analysis of the determinants of multidimensional child poverty. It uses
logistic regressions to examine the role played by children and household characteristics
in increasing or decreasing the probability of a child being considered multidimensionally
poor.
In general, the findings show that multidimensional child poverty is highest among
children living in rural areas, and in the Northern and Eastern regions. The analysis of
changes in child poverty suggests a strong reduction in multidimensional child poverty
between 2002 and 2010, driven relatively more by household-related indicators such as
access to safe drinking water, improved sanitation facilities and household consumption
expenditure, than by child-specific indicators such as schooling, literacy or child labour.
However, the reduction in multidimensional child poverty has not been uniform across
different population subgroups. The pattern of reduction across rural/urban areas and
regions has been less pro-poor than that of income poverty. In addition, the poorer
subgroups have shown slower progress, thus widening inter-group multidimensional child
poverty. The results also show a considerable poverty mismatch between the monetary
and multidimensional approaches. Finally, the findings demonstrate the importance
of education, employment in the non-agriculture sector, small family size and health
infrastructure in reducing multidimensional child poverty in Uganda.