The statistics show African poverty is on a declining trend over the past fifteen years. Since 1993, African countries have succeeded in lowering the incidence, spread and severity of poverty in, not just a few, but the majority of countries. Resource-poor African countries consistently outperformed resource-rich ones. Alongside reductions in income poverty, this period witnessed substantial improvements in social outcomes such as health and education.
But, compared to other developing regions such as South and East Asia, Africa’s progress on poverty reduction has been consistently disappointing. In fact, when compared to a typical Asian country, growth in Africa’s GDP per capita generates only half the reduction in poverty.
Of course, the quality of data underlying poverty estimates in Africa makes the analysis more challenging. Poverty data from African countries is limited both as to quality and timeliness. There is broad consensus that the rate of poverty in Africa has declined over the past 15 years. However, the rate of poverty reduction remains a point of contention, as is the precise current level of poverty in Africa. Researchers arrive at different conclusions depending on their data sources and methodologies.
Two schools of thought emerge from this debate. The pessimistic view, espoused by the World Bank and many mainstream development economists, is that African poverty has been declining slowly. Their analysis is based on household surveys. Projections from this data cast doubt over the likelihood that Africa met its MDG target of halving its 1990 poverty level by September 2015. A more optimistic view emerges from combining GDP data from national accounts and distribution data from household surveys to compute mean incomes. Using this method, researchers such as Pinkovskiy and Sa-la-i-Martin conclude that poverty has been declining faster than commonly thought. Their projections suggest that most African countries have, in fact, achieved their MDG poverty target. It is sometimes argued that the two approaches produce different results because household surveys yield much lower mean incomes that grow more slowly than GDP per capita for most African nations. In sum, while we can make rough estimates as to the direction of progress from year to year, we cannot precisely calculate poverty rates and how they are changing.
For the purpose of this study, we have combined in-come -based poverty with asset-based poverty, using data from Demographic and Health Surveys for 37 African countries in multiple waves for each country, covering the life history of some 750,000 households. The findings suggest that the percentage of households deprived of basic assets, such as those living in thatched-roofed and mud-floor houses, and lacking access to clean water, electricity, radio/television or other assets, declined from about 42% in the 1990s to 25% in 2005. This rapid decline in asset poverty is consistent with Pinkovskiy and Sala -i-Martin’s (2012; 2013) account, except that the pace remained unchanged between 2005 and 2010. On the other hand, the long-term relationship between per capita GDP growth and asset poverty reflected the pattern in the World Bank data set, where asset poverty declines by around 0.92 percentage points for every 1 percentage point increase in per capita GDP.
With African countries committed to ending extreme poverty in the next decade, generating high quality poverty data that is consistent across countries is a high priority. This entails building up the capacity and resources of national statistical institutions across the continent.