Abstract
Background: Many countries are stabilising at below-replacement fertility. While all of the reasons are unclear, postponement of childbearing is considered the main driver.
Aim: To examine determinants of postponement and differences in age at first birth.
Methods: A longitudinal study of a cohort born in New Zealand between 1972-1973. Outcomes were (a) postponement of childbearing past age 31, and (b) the occurrence of a first birth for those who have not had a child at the start of each of three periods: prior to age 21, from age 21-25, and from age 26-31. The impact of potential determinants was assessed using robust Poisson regression to calculate Prevalence Ratios and 95% Confidence Intervals. A weighted average across these periods was also calculated.
Results: Overall, 60% of men and 46% of women had not had a child by age 31. Factors that influenced postponement and age at first birth impacted differently by gender and within each age period. Proximate factors: Postponement by age 31 differed by age at first intercourse. This was driven by the increased likelihood of first birth prior to age 21 for women aged 14 or younger at first intercourse, and by the much lower likelihood of first birth at any age for men and women aged 18 or older at first intercourse, compared with age 15-17. The likelihood of first birth from 21-25 and 26-31 was nearly double for men and women who reported a prior miscarriage. Overall the likelihood of first birth was lower for men who reported an abortion, but not for women.
Family background factors: Most family background factors impacted most strongly on first birth prior to age 21, driving overall differences in postponement by 31. The impact declined from 21-25, and further from 26-31. An exception was having a mother who was younger than 20 at her first birth, which had a similar impact at all ages for women. For men, postponement was greater for those from families with high socioeconomic status than medium or low, or for those whose mothers had a university education compared to high-school or less. First birth prior to age 21 was more than five times more likely for men from the most disadvantaged homes. For women, postponement was greater among those who did not identify as Māori, from those from families of higher socioeconomic status, or whose mothers were 20 or older at their first birth. The strong differences in first birth prior to age 21, with ongoing differences at later ages for women who identified as Māori, remained important after accounting for individual factors like education and income. Individual factors: First birth was more likely for women of low socioeconomic status at all ages, but not for men. The likelihood of first birth from 21-25 and 26-31 decreased with higher levels of education, and more so for those enrolled in education. This appeared to be a function of low income for men, but not women. Marriage and cohabiting strongly predicted first birth at all ages, and marriage particularly so for men from 26-31. Adjustment for family background made little difference to the findings.
Conclusions: Higher education has played a large role in driving postponement for men and women, as has low income for men. The stronger findings for marriage and cohabitation at older ages suggest a preference for childbearing within stable unions, supporting largely traditional values about family formation, with an increase in the normative age for this. Induced abortion allows women to stay on their initial fertility trajectories, and to conform to this norm. Social norms for age at first birth appear to vary by ethnicity, and through intergenerational transmission of age at first birth. Diversity in age at first birth was created by differences in these social factors in adulthood, and more markedly by differences in births prior to age 21, driven by the experience of social disadvantage in childhood. This is resulting in a reproductive polarisation, with those experiencing greater disadvantage in society having children at a younger ages and those who are more advantaged waiting until older. Family background does not account for later differences in the likelihood of first birth by individual factors such as marriage or education.