How does variance in fertility change over the demographic transition?

Abstract

Most work on the human fertility transition has focused on declines in mean fertility. However, understanding changes in the variance of reproductive outcomes can be equally important for evolutionary questions about the heritability of fertility, individual determinants of fertility and changing patterns of reproductive skew. Here, we document how variance in completed fertility among women (45–49 years) differs across 200 surveys in 72 low- to middle-income countries where fertility transitions are currently in progress at various stages. Nearly all (91%) of samples exhibit variance consistent with a Poisson process of fertility, which places systematic, and often severe, theoretical upper bounds on the proportion of variance that can be attributed to individual differences. In contrast to the pattern of total variance, these upper bounds increase from high- to mid-fertility samples, then decline again as samples move from mid to low fertility. Notably, the lowest fertility samples often deviate from a Poisson process. This suggests that as populations move to low fertility their reproduction shifts from a rate-based process to a focus on an ideal number of children. We discuss the implications of these findings for predicting completed fertility from individual-level variables.

Wealth, fertility and adaptive behaviour in industrial populations

“…It is equally obvious, however, that the effects of wealth are modest, and that both the mean and variance in the low-fertility highincome populations covered by our literature review are very low (see electronic supplementary material and also [11,91,92]). This low variation is suggestive of a two-child norm [89,93], something that is well established in studies of people's preferences [88].…”

Status competition, inequality, and fertility: implications for the demographic transition

“…[3,67,68]). Not only is it often true that statistical models that incorporate aspects of multiple types of causation perform best in explaining actual fertility decisions (e.g.…”

Worldwide fertility declines do not rely on stopping at ideal parities

“…However, under a count process (e.g. a Poisson process) the realized completed fertility for a specific woman will not perfectly reflect a lifetime rate because of stochastic variation introduced by the count process (Hruschka and Burger 2016). Thus, the rates estimated for a Poisson process reflect a summary measure of a stochastic process, and not some "master schedule" of planned interval lengths (Bongaarts and Potter 2013).…”

“…Changing timing of births can arise for a number of reasons, including birth spacing focused on the interval between one birth and the next, a later start of first reproduction, stopping for reasons other than current parity, and postponing for reasons unrelated to the age of the last Towner et al 2016;Mattison et al 2018).…”