The Mid-Twentieth Century Baby Boom and the Role of Social Influence. An Agent-Based Modelling Approach

Eli Nomes, Katholieke Universiteit Leuven
Andre Grow, University of Leuven (KU Leuven)
Jan Van Bavel, University of Leuven

Around the middle of the 20th century, most Western countries experienced an unexpected Baby Boom. It was characterized by high levels of inter- and intra-country variability in fertility, as some regions even experienced fertility decline. In this paper, we suggest that social influence processes, propelling a shift towards two-child families, might have played an important role. Interactions in social networks can lead new types of childbearing behaviour to diffuse widely. The emergence and diffusion of a two-child norm resulted in homogenization of fertility across regions. Overall, this led to a reduction of childlessness and an increase of fertility, as more people aspired to have at least two children. Yet, in those regions where larger family sizes were still common, the two-child norm contributed to a fertility decline. To explore the role of social influence, we use agent-based computational modelling. We explicate the underlying behavioural assumptions in a formal model and assess their implications with computational simulation experiments. We use Belgium as a case study, using census data to generate realistic starting conditions and to empirically validate the outcomes that our model generates. Our results show that the proposed mechanism could explain an important part of the variability of fertility trends.

Keywords: Historical Demography, Computational social science methods, Fertility

See paper.