Education vs TFP: Empirical Evidence from The Sub-Saharan Countries

This paper investigates the \education-total factor productivity trade-o " in explaining per worker income di erences between Sub-Saharan (unlucky) and G7 (lucky) economies. Following Hall and Jones (1999) and Caselli (2005), on a country basis, I am able to study separately the dynamic of the average years of schooling (i.e. education level), the per worker capital, the per worker income, and the total factor productivity (TFP). I con rm that physical capital and education levels partially explain income di erences between unlucky and lucky economies. In a time-series setup I create, on a country-by-country basis, ad hoc TFP shock times series. The main result of this paper is that the impact of TFP shocks on per worker income is larger in unlucky economies than in lucky ones. The result holds both for negative and positive shocks. I show that average TFP volatility in the "unlucky world" is 8 times higher than the "G7 world" average TFP volatility. I argue that the order of magnitude of the impact heavily depends on the level of the TFP volatility. It turns out also that the e ect of a TFP shock on a relative low per worker income growth rate is higher. I conclude by arguing that the presence of low levels of per worker capital and of human productivity pushes the unlucky economies into a poverty trap.