The study uses the most recent Integrated Household Survey (2010 IHS) data to explain the
rural-urban differences in poverty in Malawi. In analysis, a welfare model is adopted from the
other studies on poverty, where the per capita consumption expenditure is used as the welfare
indicator. The paper further adopts the Machado-Mata decomposition technique to attribute the
rural-urban welfare gap into the ‘characteristics effect’ and the ‘returns effect’. In addition,
Dinardo, Fortin and Lemiueux (DFL) approach is employed to give a detailed decomposition of
the ‘characteristics effect’. The results show that significance of the rural-urban differences of
variables of the welfare model varies across the welfare distribution. This entails that the
differences are significant in some quantiles and insignificant in the others. Secondly, the
Machado-Mata decomposition technique found that both the differences in characteristics and
differences in returns to those characteristics significantly contribute to the urban-rural welfare
gap. Specifically it was found that the ‘returns effects’ were dominant across the whole
distribution. Thirdly, through the DFL technique, it shows the specific variables that contribute to
the ‘characteristic effects’.