Fertility expectations: a short cut or dead-end in predicting fertility

Abstract

The Norwegian Fertility Survey (1977) and a follow-up study (births recorded among those same women for the 1978-82 period) were used to examine if an easy method for estimating completed cohort fertility is also a viable one. If so this means that the responsibility of making fertility projections merely shifts from the forecaster to the women themselves. The discussion summarizes opportunities and results in analyzing fertility expectations. For this purpose fertility expectations are defined as beliefs or ideas about future births. Analyses of birth expectations can be divided into 2 main topics: do birth expectations to some extent remain stable; and what is the relation between fertility expectations and fertility behavior. The Norwegian Fertility Survey provides information about the womens childbearing expectations in 1977. The Central Population Register gives information about the number of realized births among these same women during the following 5 years (1978-82). Interviews were successfully completed with 4137 women. The sample was representative of Norwegian women aged 18-44 years (born 1933-59). The non-response rate was 18%. All the women interviewed were followed up in the register survey. Up to the end of 1982 12 of them had died. The Central Population Register gives information of all births in the same period excluding women who had emigrated or stayed abroad during all or a part of that period. Of the respondents in 1977 42% expected to have children in the future while the same percentage expected not to have children in the years to come. The question on expectations was directed to both childless women and women with children. 84% of the women younger than 25 expected at least 1 birth in the future; only 50% of the women in the last half of their 20s thought they still had births to come. The proportion expecting future children declines further to 19% among women 3034 years of age and for the older women almost none expect to give birth in the future. Almost all women who in 1977 expected not to have (more) children have been consistent in their behavior. 5 years after the interview only 3% of them had had children. As much as 51% of those who expected a birth in the future have not yet had any birth. Among the women expecting to have terminated their childbearing in 1977 5 out of 5 were older than 30 years. Among all respondents more births were wanted than were realized. As the results show overestimation not underestimation unwanted births cannot be the explanation of the difference between expectation and behavior. If the youngest women had had the same age-specific 1st birth pattern as women of age 18-20 at the end of the 1960s and the beginning of the 1970s a considerably higher percentage among them would have had children.