June 2004

No. 1194: Intergenerational Effects in Sweden: What Can We Learn from Adoption Data?

We explore the adoption data approach to estimating causal effects of parental education and income on the same outcomes of their children. Thanks to a data set drawn from Swedish population registers with detailed information on biological background and history of adoptees, we can test basic assumptions that the adoption strategy relies on. We find that the adoption method survives these tests surprisingly well. Our empirical results suggest that one more year of either mother’s or father’s education raises children’s education by about 0.1 year. Our estimated income elasticities are around 0.1.