This from a new NBER working paper by Ariell Reshef (U. Va.) and myself. We also tell some stories of the individual successes, some of which involve the local government.
The news is not that Africa is different from the rest of the world in this, but that it's the same.
This unstable uncertainty holds regardless of whether you include or exclude oil, minerals, and other export commodities. And so does the concentration of success -- the top-ranked non-commodity export is 23 times larger than the 10th ranked export.
The stereotype of African countries as unchanging mono-exporters based on some unchanging natural endowment just turns out to be...wrong.
Coping with such remarkably high and unstable uncertainty (the "unknown unknowns") as to what will be a hit seems like an a priori case for a lot of decentralized, highly motivated seekers and experimenters. We don't exclude ANY possible government involvement -- at the very least, governments need to be nimble to adjust regulations and infrastructure to support any new success that comes along from private entrepreneurs.
In sum, we think there is just as much a role for entrepreneurs in Africa, and just as little role for centralized and systematic government industrial policy, as in the rest of the world.