Education has an important effect on wages but it not clear whether this is because education raises productivity or because education is simply a signal of ability. We implement a number of existing tests for discriminating between these two explanations and find that they do not support the signalling hypothesis. However, we have severe reservations about these results and we propose an alternative test based on changes in education incentives caused by changes in the minimum school leaving age in the 1970s. Using this idea we find that UK data appear to strongly support the human capital explanation.
No takes yet. Share an insight, caveat, or question.
Chevalier et al. (2004) studied this question.