A Robust Multi-Dimensional Poverty Profile for Uganda

In this paper we compute a multi-dimensional poverty index (MPI) for Uganda following the approach proposed by Alkire and Forster (2007). Using household survey data we show how the incidence of multi-dimensional poverty has fallen in recent years and we use the decomposability features of the index to explain the drivers of reduction in multi-dimensional poverty. We also compare the results from Uganda with other countries for which the MPI has been computed and we note some caveats in such a comparison. The robustness of our estimates is tested in a stochastic dominance framework and using statistical inference. Notably, we extend the one-dimensional analysis of stochastic dominance to take into account household size in a second dimension, which is particularly important as some of the MPI indicators are sensitive to the number of household members. By exploiting a unique subsample of the integrated household survey programme in Uganda, which has not previously been analysed, we are also able to match the data-set used for the MPI with data used to compute the conventional estimates of monetary poverty. This enables a more robust assessment of the complementarities of the two types of poverty measures than has been previously possible.