Discussion paper series // Forschungsinstitut zur Zukunft der Arbeit 5946
Behrman and Rosenzweig (2002) used data on a small sample of MZ (monozygotic, identical) twin parents and their children to show that father's schooling is more important than mother's schooling for children's schooling in the U.S. Recent studies based on much larger samples of twins from registry data in Scandinavian countries reach similar conclusions. Most of these studies, however, are unable to distinguish between MZ and DZ (dizygotic, fraternal) twins. Using data from the Swedish Twin Registry, we replicate the finding that father's schooling matters more than mother's schooling in a combined sample of MZ and DZ twin parents. In contrast, results based on MZ twin parents show that mother's schooling matters at least as much as father's schooling for children's schooling. We also estimate the effect of parents' schooling separately by child gender and find this effect to be entirely driven by the impact of mother's schooling on daughter's schooling. Our results show that (1) it is vital to have zygosity information to estimate causal intergenerational effects and (2) the conclusions reached by Behrman and Rosenzweig (2002) for the U.S. do not apply in Sweden.