Age of Parent or Timing of Birth? Addressing the Challenge of Identifying and Interpreting Parental-Age Effects in Studies of Child Outcomes

Do children of older parents do better because their parents are older, or because they are born into "better times"? I argue that this question is challenging to answer empirically because it poses an age-period-cohort-type identification problem. Once parental birth year is treated as a relevant confounder—either explicitly or implicitly, as in sibling-comparison and family fixed-effects designs—parental age at birth and child birth year become perfectly collinear. I address this problem by adapting Fosse and Winship’s bounding approach and applying it to educational attainment in the Generations and Gender Survey for Belgium, Germany, the Netherlands, and Sweden. Under weak, historically grounded assumptions, the admissible parental-age profiles rise from teenage motherhood into the late twenties and early-to-mid thirties, whereas child-birth-year effects are modest and cannot account for much of this pattern. I also show why prior family fixed-effects studies consistently report attenuated parental-age estimates once child birth year is introduced: this attenuation can arise mechanically from the model’s identification structure and need not be given a substantive interpretation. Overall, the analysis adds nuance to claims that delayed-parenthood advantages are largely explained by period improvement.