Previous studies, based on event-history methods, reported that highly educated women in Northern, Western and Southern Europe are now more likely to have the second child than the low educated. Event-history models do not allow, however, for concluding whether they had the second birth earlier or were indeed more likely to have it. We address this oversight by estimating cure models, which allow to assess the effects of education on the timing of the event and the probability of its occurrence. We use Harmonized Histories and cover 16 European countries and the US. We find that highly educated women display higher probability to have the second child only in Belgium. In many other Western European countries and the US no educational gradient in second births persist, while in Eastern Europe it is negative. Finally, we find no evidence for a reversal in the educational gradient in second births over time.
Presented in Session 1. Fertility, Family Planning, Sexual Behavior, & Reproductive Health 1