Abstract
Twins-based estimates of the return to schooling have featured prominently in the economics of education literature. Their unbiasedness hinges critically on the assumption that within-pair variation in schooling is explained by factors unrelated to wage earning ability. This paper develops a framework for testing this assumption and shows, in a large sample of monozygotic twins, that the twins-based estimated return to schooling falls if adolescent IQ test scores are included in the wage equation. Using birth weight as an alternative proxy for ability yields qualitatively similar results. Our results thus cast doubt on the validity of twins-based estimates. © 2013 The Authors.