Misallocation of human capital in developing countries

It is now well established and documented that capital and labor are very poorly distributed across and within sectors in developing economies. The impact of this misallocation is large enough to explain a non-negligible part of the gap between rich and poor countries. This analysis has, however, only pertained to worker counts and physical capital. What about human capital?

Dietrich Vollrath looks at the sectoral allocation of human capital in 14 developing economies, analyzing the marginal return by using wage data. He comes to the conclusion that the misallocation has an impact on GDP of at most 5%, more than in the US but clearly negligible to explain cross-country differences. I find this hard to believe from my casual observation. For example, in many developing economies, the brightest minds go into government administration because this is where they can extract the most rents (some call this corruption). They expand a bureaucratic machine at the expense of the productive sector. That cannot be good for GDP. Using wage data here can be misleading, as the marginal product of corrupt bureaucrats, at least in social terms, is certainly not reflected in their wage. Add to this important informal sectors where wages may be substantially mismeasured, even in household surveys, and I am not quite as confident in the data as Vollrath seems to be.