Theoretical and practical issues The impact of employment trends

4 Trends in first births

The drop in first births has been particularly strong among women under 30. As we see in Figure 4, first-birth rates at these ages have followed each other remarkably closely. At each age, the rates declined until the middle 1980s, rose in parallel up to a highest value around 1990, and then declined strongly during the 1990s. The first-birth rates of women under 25 have actually been cut in half between 1990 and 1997. For somewhat older childless women, the birth rates have also moved in parallel, but they have a different pattern (see the right panel of [Figure 4]). The increase started earlier and continued longer than at younger ages, but in later years first-birth rates have declined even at these ages. During 1997, the first-birth rate for 36-year-old women was about the same as the rate at age 25. This is an entirely new pattern. During 1990, women at age 25 had 75 percent higher rates than women at age 36. The changes in the age structure of first-birth rates are highlighted in [Figure 5].)

It is no news that young people delay entry into parenthood. A rise in the age at first birth started at the end of the 1960s already, and it continued until the mid-1980s. The postponement then stopped as first-birth rates increased at the younger ages, but it got going again as fertility dropped in the 1990s.

Postponed first births are reflected in an increasing prevalence of childlessness at steadily higher ages [Figure 6]. Among women born at the beginning of the 1950s, about 40 percent were childless at the end of the year in which they turned 25. This fraction increased successively and reached 60 percent among women born only ten years later. The fertility increase during the late 1980s cause a break in the postponement of motherhood which produced a plateau in the percent childless at all ages, followed by a new rise in childlessness as fertility dropped in the 1990s. At the end of 1997, over 70 percent of women were childless at age 25. The differences between one cohort and the next in the percent that are childless decrease as we proceed to higher ages. This means that so far, the delay in childbearing has not caused an increase in permanent childlessness. Swedish women have compensated for their low first-birth rates at younger ages by getting more first births at higher ages. One may well ask whether such compensation will be possible today as well, for childlessness has become so very prevalent at middle childbearing ages.

Using classical demographic (indirect) standardization techniques, we may produce an index for the general level of first-birth rates. [Figure 7] displays the time trend, standardized for age (for ages 17 through 38). We have computed the index by dividing the annual age-standardized figure by its value for 1990. The level fell by 35 percent between 1990 and 1997, which is much more than the rise in the second half of the 1980s.

This standardization amounts to a very simple hazard regression with only the two factors `age' and `(calendar) period'. The curve in [Figure 7] shows the average fertility change across all ages. Interesting further features appear if we run an interaction between the two. As is seen in [Figure 8], the increases in first-birth rates essentially went in lockstep between 1986 and 1990 (the annual curves in the diagram are almost horizontal); over this period the rates increased by almost fifteen per cent at all ages. After 1990, the changes have been very different at different ages. Fertility has decreased successively for the youngest of the childless women, and was only about half in 1997 of what it was in 1990. As we move up through ages above 25, the decrease is progressively smaller, and at ages between 33 and 36 there mostly has been an increase instead. The fertility decrease among childless women after 1990 has been concentrated to the younger age groups. This surely reflects the greater problems that young people have had in the labor market and the fact that they can postpone childbearing more easily than older women.


Theoretical and practical issues The impact of employment trends

logo70.gif (2450 bytes)

Entry into motherhood in Sweden:
the influence of economic factors on the rise and fall in fertility, 1986-1997

Britta Hoem
© 2000 Max-Planck-Gesellschaft ISSN 1435-9871