Harry Greenwell responds to an NGO advertisement I posted last week, which claims life expectancy in Africa is 47 years:
An average life expectancy of 47 years is so low that I was a bit stunned, and I began to wonder about the degree of country-by-country variation within sub-Saharan Africa (SSA). No doubt there is plenty of debate about quality and definitions of life expectancy measures but for some quick-and-dirty research, I've simply relied on the UN 2005-2010 life expectancy figures (published 2006) and the CIA Factbook figures (published 2009), both reproduced on Wikipedia.
Three things struck me. First, there are some substantial discrepancies between the UN and CIA figures, suggesting either some dramatic changes in a short period or serious data quality and measurement issues (or, most likely, both). For example, Botswana's life expectancy is either 62 years (CIA) or 51 years (UN). Similarly, Equatorial Guinea is either 62 years (CIA) or 52 years (UN). The UN figures aren't uniformly lower, however, since it reports Sudan's LE as 59 years whereas the CIA says it is 51 years. And on others, like Nigeria, they are nearly identical (46.9 years).
Second, there is a fair bit of variation between countries. For example, on the CIA measures, average life expectancy is around 61-62 years in a few SSA countries like wealthy but AIDS-afflicted Botswana, wealthy but corruption and conflict-afflicted Equitorial Guinea and also in Mauritania. By contrast, a cluster of countries including Swaziland, Mozambique, Sierra Leone, Zambia and Zimbabwe do indeed have very low life expectancies ranging from 40-45 years.
Third, it seems that the figure of 47 years cannot be consistent with the UN or CIA figures. Nigeria (154 million people, 47 years), sub-Saharan Africa's most populous country, sits near this purported average and depending on which data source you prefer, some other large countries (DRC: 66 million, 46.5 years, [UN]) are there or thereabouts. However, several other populous nations fare somewhat better (Ethiopia: 79 million, 55 years; South Africa: 49 million, 49 years; Sudan: 39 million, 51 years; Kenya: 39 million, 58 years [CIA]), which should drag the average above 47 years.
Overall, these quibbles should not detract from the ad's fundamental points — life expectancy in sub-Saharan Africa is appallingly low (the world average is 67 years) and AIDS is presumably a big contributor. The ad can certainly be forgiven for not referring to data quality issues and the variation between SSA countries, and can perhaps also be forgiven for not including a source for their life expectancy figure, although I don't think that would be too much too ask.