Background: Increased postponement of fertility, especially among higher-educated women, means it is important to know whether women recuperate births at older ages, but evidence for the UK is lacking. The extent to which the timing and quantum of mothers’ fertility underlie the strong educational gradient in completed family size is also unclear.

Objective: We investigate the relative contributions of childlessness, timing, and quantum to educational differences in completed fertility within cohorts born between 1940 and 1969.

Methods: We analyse retrospective fertility histories from 44,351 women, born 1940-1969, interviewed in the British General Household Survey (1979-2009) and the UK Household Longitudinal Study (2009-2010). After describing educational differences in the timing of first birth and parity distributions, we quantify the relative contributions of childlessness, delayed entry into motherhood, and fertility rates conditional upon age at entry into motherhood, to educational differences in completed family size.

Results: Within each cohort, the educational gradient in completed family size is explained, in demographic accounting terms, almost entirely by educational differences in the proportions remaining childless and the age distribution of mothers at entry into motherhood. Conditional upon age at entry into motherhood, subsequent fertility rates are similar across educational groups and across cohorts.

Conclusions: Unlike for some other European countries, the postponement of motherhood to later ages in Britain has not resulted in a significant increase in childbearing among more-educated women who enter motherhood at later ages. The stability of aggregate measures of completed fertility in Britain is not the result of a straightforward process of postponement followed by recuperation.