Hybrid seminar: Measuring Social Mobility with Detailed Parental Information

This paper demonstrates the value of detailed parental information in social mobility research. Using Dutch administrative data, I link children's economic outcomes to a rich array of family background characteristics—including income, wealth, education, crime, health, and family structure—through a machine-learning approach. The results show stronger intergenerational dependence than traditional models, particularly for disadvantaged children. Gender differences in family background effects are minimal, despite allowing for complex, gender-specific patterns. Lastly, family background is significantly less predictive for adoptees than for own-birth children, highlighting the critical role of pre-birth factors.