Does the Human-Capital/Educational-Sorting Debate Matter for Development Policy?

If education increases human capital, subsidizing education can generate economic growth and combat poverty. Estimates of its return suggest that education is a good social investment. In sorting models, the return reflects in part the information about productivity revealed by the worker's education. Thus the social and private returns diverge. It might appear that if we believe the sorting model, we should be less swayed by evidence that estimated returns to education exceed the social discount rate, and therefore less likely to support education-based development policies. This conclusion is shown to be incorrect.