Regression-Based Simulation of Anti- Poverty Policies in Uganda

Abstract

Poverty has increased in Subsaharan Africa over the past two decades both in absolute terms and as a share of the world’s total poor. The number of persons estimated to be living on one dollar a day or less in Subsaharan Africa increased from 164 million in 1981 to 314 million in 2001 (World Bank, 2004). Over the same period, poverty in Subsaharan Africa as a share of world poverty rose from 11.3 percent to 28.5 percent. By 2015, one in every two poor persons in the world will live in Subsaharan Africa, compared to one in five in 1990 and one in ten in 1980.1 To measure poverty rates, African governments and multilateral aid organizations have devoted much effort and expense to the implementation of periodic household surveys. Poverty estimates derived from household surveys are used as performance criteria by the World Bank in aid negotiations with Heavily Indebted Poor Countries (HIPC).2 Beyond the measurement of poverty rates, the household survey data now available in many Sub-saharan African countries represents a rich source of data for analysis of the poverty impacts of domestic policies. In this paper, we utilize a regression model and data from the 2002/2003 household survey in Uganda to analyze the poverty alleviation potential of anti-poverty strategies.