Vulnerability to asset-poverty in Sub-Saharan Africa

This paper presents a methodology to measure vulnerability to asset-poverty. Using repeated cross-section data, age cohort decomposition techniques focusing on second-order moments can be used to identify and estimate the variance of shocks on assets and, therefore, the probability of being poor in the future. Estimates from the Ghana Living Standard Surveys show that expected asset-poverty is a reliable proxy for expected consumption-poverty. Applying the methodology to nine Demographic Health Surveys countries, urban areas are found to unambiguously dominate rural areas over the unidimensional distribution of expected future asset-wealth, as they also generally do over the bi-dimensional distribution of present asset-wealth and expected future asset-wealth.