Parenthood Timing and Socioeconomic Consequences in Young Adulthood: A Longitudinal Analysis of Women and Men
Mikko Myrskylä, London School of Economics and Political Science
Jessica Nisén, Max Planck Institute for Demographic Research
Ben Wilson, Stockholm University
Maarten Jacob Bijlsma, Max Planck Institute for Demographic Research
Pekka Martikainen, Centre for Health Equity Studies
The study is based on a 10-percent household sample in the census in Finland in 1975, with links to annual register-based socio-demographic attributes in 1987-2007. Women and men born in 1974-75 are followed from age 16 to 32 (N=11,061). In the statistical analysis we apply the g-formula, where we implement a counterfactual scenario where the entry into parenthood is postponed by five years. Time-varying confounding is controlled for by simultaneous mutual effects and time-invariant confounding by a set of background characteristics measured at age 15.
In the preliminary results, educational attainment and enrolment in both the female and male general population were affected by the postponement, but the effects appeared stronger among women. The cumulative effect on attainment by age 32 was 10.1 percent (95% CI 6.1-15.0%) among women and 5.7 percent (95% CI 1.0-10.9%) among men. In terms of employment and income only women were affected.
The postponement of parenthood changes the young adulthood life stage with regard to socioeconomic standing in particular among women. Given the current findings, the postponement of parenthood itself may contribute to the strengthening of the socioeconomic position of women relative to men.
Young adulthood is an intenseperiod of life characterized by multiple transitions occurring simultaneouslyand being associated with each other. Completing education, establishingoneself in the labor market, and entering a stable relationship and parenthoodconstitute crucial steps in the overall transition to adulthood. Due to its intensecharacter, the establishment of causality between changes in different lifespheres in young adulthood is difficult. Postponement of parenthood has beenone of the most profound changes in family formation throughout the Westernworld over the last half century. The relatedness of the age at entering parenthoodto outcomes in other life spheres in young adulthood has been explored innumerous studies among women. The few studies that have covered men tend toconclude that parenthood associates with socioeconomic characteristics in youngadulthood more strongly among women than men. It remains unclear what thecausal mechanisms behind these associations are, and what exactly causes theobserved gender differences in socioeconomic outcomes. The studies on thistopic that do assess causality are often compromised in external validity. Furthermore,the mediating roles of different socioeconomic characteristics on each otherremain largely unclear.
1. How does postponement of parenthood affect thesocioeconomic characteristics of women and men in young adulthood?
2. How does postponement of parenthood affect thesocioeconomic characteristics of parents-to-be in young adulthood?
3. Do socioeconomic characteristics mediate eachother?
4. Are there gender differences with respect tothe aforementioned questions?
The study is based on a 10-percent random sample of householdsfrom the Finnish census of population in 1975, with linkage to a large numberof register-based socio-demographic attributes in 1987-2007 measured on anannual basis. We follow women and men born in 1974-75 from age 16 up until age32 (N=11,061). To this data we apply the parametric g-formula: we firstestimate the joint distribution of the socio-economic covariates and fertilityin the data. Then, we use this joint distribution to estimate a counterfactualscenario where the entry into parenthood is postponed by five years and studythe resulting changes thereof in socioeconomic outcomes. This method accounts fortime-varying and intermediate confounding. Additionally, we control fortime-invariant confounding by a set of background characteristics measured at age15. We also introduce a sensitivity check with respect to gender differences bygiving men the age-specific fertility schedule of women and studying whetherthis affects the findings.
The preliminary results showed that in the counterfactual scenariowhere parenthood is postponed by five years the level of education wouldincrease among women in particular. By age 32 there was a ten percent increasein the share of women educated to higher tertiary level and a corresponding sixper cent increase among men in the general young adult population (Figure 1). Inline with these observations, the postponement of parenthood increasedenrolment in education throughout the twenties among women, with the effectamounting up to nearly five percent at its highest in the late twenties. Formen the respective effect was also significant, but apparently weaker.
In the counterfactual scenario, also employment of women wasstrongly affected: increases were witnessed from the early twenties, amountingup to over ten percent at most in the late twenties (Figure 2). Unemployment ofwomen decreased from the mid-twenties, with the largest relative effect in theearly thirties (15 percent). Mens employment or unemployment were not affectedby the postponement at the population level. Womens income in young adulthoodwas increased throughout the young adulthood from the late teens, with theeffect being most pronounced in the late twenties (four percent). Income of theyoung adult male population was not affected respectively. We also witnessed slightincreases in household income among women, but not among men.
The postponement of parenthood changes the young adulthood stageof life in particular among women. Overall, we find that in terms ofeducational attainment and enrolment thesocioeconomic standing of both women and men in young adulthood is affected bythe postponement of parenthood, but in terms of employment and income onlywomen are affected. Also with respect to education, the effects appear to bestronger among women than among men. Further results will look closer at theeffects among parents-to-be, on the mutually mediating factors of theseeffects, and will also further shed light on gender differences. Given thecurrent findings, we conclude that postponement of parenthood itself maycontribute to the strengthening of the socioeconomic position of women relativeto men, with potentially broad consequences on family life as well as other spheresin life.
Figure 1 The relative population-averaged effect of afive-year delay of parenthood on higher tertiary education attainment amongwomen and men at age 16-32
Figure 1 The relative population-averaged effect of afive-year delay of parenthood on employment status among women and men at age16-32
Presented in Session 1153: Fertility