This paper uses data on individual earnings in manufacturing industry for five African countries in the early 1990s to test whether firms located in the capital city pay higher earnings than do firms located elsewhere, and whether such benefits accrue to all or only certain types of workers. Earnings equations are estimated that take into account worker characteristics (education and tenure) and relevant firm characteristics (notably size and whether owned by a foreigner). Any location effect identified is therefore additional to appropriate control variables. There are two main findings. First, we find evidence of a 'capital city premium' that varies between 12% and 28% in the five countries. This location premium does not always exceed plausible consumer price differentials, between the capital and other areas, and therefore does not demonstrate that real (purchasing power) manufacturing wages are higher in the capital city. This suggests that spatial inequality in real earnings is unlikely to be significant for manufacturing employees. Secondly, while we find that skilled workers earn a higher wage premium in the capital city than less skilled workers, this is not because of location effects on earnings per se, but rather because of other firm characteristics of firms located in the capital city such as size and foreign ownership. This suggests that spatial inequality in itself does not directly contribute to skilled/less-skilled wage differentials in manufacturing.