LAUSR

Abstract Estimation of a panel model during the period 1960–2013 shows that there is no significant negative correlation between adult mortality and urbanization in sub-Saharan Africa. For the rest of the world the correlation between adult mortality and urbanization is significantly negative. A high prevalence of HIV likely explains the absence of a significant negative correlation between adult mortality and urbanization in the sub-Saharan African region. During the 1960s and 1970s, i.e. prior to the HIV epidemic, adult mortality and urbanization are significantly negatively correlated in sub-Saharan Africa. Panel model estimates show that the urban health premium is significantly decreasing in HIV prevalence. So much so, that for countries with very high HIV prevalence rates, i.e. above 10 percent, there is a significant urban health penalty. Despite a number of features of sub-Saharan Africa that significantly increase the urban health premium – such as a large share of the population at risk of contracting Malaria and high average temperatures – the high HIV prevalence rates during the 1990s and 2000s completely overshadow these features so that, on average, there is no significant negative correlation between urbanization and adult mortality in the sub-Saharan African region.