Home

Laboratory

Fertility and Well-Being

Project

Causal Inference Approaches to Fertility over the Life Course

Maarten Jacob Bijlsma, Jessica Nisén; in Collaboration with Ben Wilson (Stockholm University, Sweden)

Detailed description:

Demographers have shown considerable interest in understanding how socioeconomic processes collectively determine fertility. Most theories of fertility predict that a range of socioeconomic factors have an impact on the quantum and tempo of childbearing. In particular, a number of theories have postulated that socioeconomic processes are interrelated, and that they determine fertility collectively and simultaneously. However, it remains unclear (especially when examining the empirical literature) how different micro-level socioeconomic processes interact to collectively determine individual childbearing trajectories (tempo and quantum), and thus affect macro-level fertility.One reason for this lack of clarity is that traditional methods often struggle to disentangle path-dependent time-varying causal interrelationships between variables (e.g., socioeconomic factors).

We propose a new approach for studying the socioeconomic determinants of fertility, based within statistical theories of counterfactual causal inference. We focus in particular on mediation analysis with the parametric g-formula. The g-formula is a highly flexible statistical method specifically developed within counterfactual causal inference to account for the problematic aspects of causal interrelationships (selection bias, reverse causality, time-varying confounding, and intermediate confounding). It estimates effects specifically by generating counterfactual scenarios; in essence, a “what-if” scenario such as: "What would women’s fertility trajectory look like if they had postponed childbearing by one year?" This method estimates the effects by using individual-level data, allowing for the control of various confounders and avoiding the ecological fallacy, but then allowing for individual-level effects to be generalized across heterogeneous populations to provide national-level effect estimates. Because it has these qualities, the g-formula seems to be a very promising technique for various demographic applications.

We use data from the 1970 British Birth Cohort to study the (near)-completed fertility trajectories of women, and in particular the impact of various hypothetical scenarios on these trajectories. These scenarios are as follows: first, an increase in higher education attendance among women; second, a reduction in the preference for marriage (to simulate the societal trend of marriage decline and cohabitation increase); third, an increase in post-birth employment among women; and fourth, the effect of all women growing up as a single child (to gain insight into the intergenerational transfer of family size preferences on fertility). We find that all of these scenarios lead to a lower level of fertility; especially the effect of education increase is initially strong. The increased education scenario also shows some degree of catch-up: Women who initially postponed childbirth due to more time spent in education have somewhat higher fertility later on (though they had not yet completely caught up to non-intervention levels by age 38). We also perform a mediation analysis for the reduction in the marriage scenario and the only-child scenario, which shows that especially the effect of the only-child scenario on fertility operates not just directly through child birth, but also through changes in the mediating variables of employment, partnership, and education.

Another project that uses the g-formula is currently in progress. In this project, we use Finnish register data to study the effect of parenthood postponement, model male fertility, and compare it with female fertility.

Research keywords: Statistics and Mathematics; Life Course; Fertility Development; Family Behavior

Region keywords: United Kingdom

Publications

Nisén, J.; Bijlsma, M. J.; Martikainen, P.; Wilson, B.; Myrskylä, M.:
MPIDR Working Paper WP-2019-017. (2019).

Socialize