The Older, the Better? The Causal Effects of Parental Ages on Children’s Educational Performance

Michael Grätz , Université de Lausanne, Stockholm University
Felix C. Tropf , University of Oxford
Torkild Hovde Lyngstad , University of Oslo

During the 20th century, there has been an increase in the ages at which parents have children. The consequences of this demographic shift are of great interest to policymakers, the public, and various scientific disciplines. In particular, parents and their offspring seem to face a trade-off between negative biological and positive socioeconomic consequences of delays in parental ages at birth. On the one hand, epidemiological research has shown that children of older parents are at a higher risk of a low birth weight and to develop symptoms of mental health conditions such as autism and schizophrenia. On the other hand, numerous sociological studies have documented a positive association between higher parental ages and children’s educational outcomes. A crucial question is whether such associations are causal. Only if parental ages causally influence children's educational outcomes, a shift in the distribution of ages will lead to improved educational attainment at the population level. Our study contributes to this research by using a novel approach to estimate the causal effects of maternal and paternal ages on children’s educational performance. We employ instrumental variable estimation to estimate the effects of maternal and paternal ages on their offspring’s educational performance. We apply Mendelian randomization (MR), leveraging the random assortment of genes influencing the age at first birth of men and women as instrumental variables for maternal and paternal ages. We avoid problems of genetic confounding in MR IV analysis by only using genes of parents for constructing the polygenic scores of age at first birth, which parents have not transmitted to their children.

See extended abstract

 Presented in Session 62. New Methodological Approaches in Fertility Research