Multidimensional Spatial Poverty Comparisons in Cameroon

The study investigates poverty comparisons across the various strata and urban/ rural areas in Cameroon. A composite poverty indicator is constructed using multiple correspondence analysis by taking into account 33 non-monetary indicators that have been identified as describing a real poverty situation. The composite poverty indicator is combined with per capita consumption to estimate poverty measures showing that income poverty affects 39.6% of households, whereas 80.6% of households are poor in the non-monetary dimension. The incidence of multidimensional poverty is estimated to be at 81.3%. Decomposition of the Chakravarty indexes fails to establish robust regional poverty orderings and comparisons. By resorting to the stochastic dominance approach we find that bi-dimensional poverty for urban areas is robustly lower than that for rural areas. Between regions, there is clear evidence that bi-dimensional poverty in Yaounde/ Douala is less than other regions and that the Rural Savannah is the poorest region of the country for a wide range of poverty lines and a broad class of poverty measures. The discriminatory measures of variables reveal that water, sanitation, housing materials, levelof education and roads are the major indicators of non-monetary poverty in Cameroon. Policy should therefore, in addition to promoting income-generating activities, focus on these variables and target the rural areas as well as the northern regions to better alleviate poverty in Cameroon.