Family background and the estimated return to schooling: Swedish evidence

Earnings regressions for married and cohabiting Swedish males in 1993 indicate that controlling for family background reduces the measured return to education by about 9 percent, net of measurement error bias. The Swedish evidence is generally consistent with the hypothesis that family background effects are primarily a result of an efficient marital sorting mechanism, which provides a signal about unobservable traits rather than being an indicator of nepotism.