2 Trends in first childbirth
The fertility rates since the late 1960s reflect the considerable change in family establishment among post-WWII generations. Women born after WWII have in many significant areas had totally different opportunities than women born earlier. Free access to abortion and better contraceptives have made it a lot easier for women to choose when they want children and how many. Also, more education and participation in the labour force have given women greater economic independence. Other societal changes, such as increasing women's liberation and broader acceptance of cohabitation contribute to these changed opportunity structures.
There are significant changes in age at first childbirth. The proportions of women that have had their first child at different age levels are shown in Figure 2.
Proportion first childbirth by age and birth cohort
The figure shows an upward trend in age at entry into motherhood. For women born in 1950 the majority had given birth to their first child before they turned 25 and less than one fifth gave birth to their first child between ages 25 and 30. For women born in 1968, the proportion that had given birth at age 30 is almost at the same level as the proportion with children at age 25 among women born in 1950. At the same time, the proportion at age 25 had declined more than the proportion at age 30, and the proportion that gave birth for the first time between ages 25 and 30 had increased to more than one third. This illustrates a strong postponement of first childbirth among younger women in Norway. This postponement has been followed by an increasing number of births to women in their thirties. The proportion giving birth for the first time between the ages of 30 and 35 has increased from 5 per cent among the 1950 cohort to 10 per cent among those born ten years later. There is also an increasing but still small proportion with their first birth between 35 and 40 years.
The proportion of women giving birth for the first time as a teenager peaked among those born around 1950. For the 1950 cohort this proportion was 23 per cent and the corresponding proportion for the 1978 cohort was 8 per cent. The proportion of teenage mothers has remained at a stable low level during the 1990s [Note 1].
Another way of looking at these trends is through median and quartile age at first childbirth. This is shown in Figure 3.
Median and quartile age at first childbirth by birth cohort
Fifty per cent of the women born in 1950 had become mothers when they turned 22.8 years (median age). Among those born in 1970 the median age has increased to 26.7 years. Quartile age (age where one fourth of the cohort have become mothers) has also increased, but not at the same speed as median age. Quartile age among women born in 1950 was 20.2 years and 22.6 in the 1970 cohort. Among women born in 1973 it has increased further to 23.3 years. The difference between quartile and median age has increased with almost two years from the 1950 to the 1970 cohort. There is a plateau over the cohorts born in the late 1960s in the curve for quartile age and a similar plateau for somewhat earlier birth cohorts in the curve for the median age. This happened in the late 1980s, at the same time as the total fertility rate started to increase. A possible explanation that hasn't been examined, is the start of an expansion in family reforms, such as the increase in child allowance, extended duration of paid leave at childbirth and the growth in number of kindergartens.
Postponement of first childbirth is more pronounced for some groups than others. When talking about women's way of living, a distinction between different generations is frequently made. Traditional patterns of fertility are typical for women in the cohorts from the 1930s to around mid 1950, while the new patterns characterise younger women. However, not all women follow new trends. Some women follow traditional patterns more closely than others which shows that women's lifestyles vary more today than they did a generation ago .
Education is a significant factor in analyses of women's fertility . Postponement of first childbirth may be seen in connection with the sharp growth in the proportion of women with a higher education among younger women. We shall look closer into educational differences in the timing of the first childbirth among women of different generations. The proportion of women completing college or university has doubled since 1980 and in 1996 more than 20 per cent of all women over the age of 16 had a higher education. Among women aged 30-39, 30 per cent had completed college or university in 1996, while the proportion was 18 per cent in 1980 . Median age at first childbirth by educational level is shown in Figure 4.
Median age at first childbirth by birth cohort and educational level
Median age is increasing by education level. In all birth cohorts women with a lower education level become mothers earlier than women with a higher education. There has been a postponement of first childbirth in all education groups, but it started first among university-educated women with higher degrees. For them the postponement started among those born in 1945 and was more pronounced for women born until the late 1950s. For women with no education beyond compulsory level median age started to increase first among women born in the mid-1950s.
The postponement of first childbirth has occurred at different paces and education has had an increased effect on the timing of first childbirth. Education induces differences in the timing of first childbirth, and these differences are increasing. Among women born in 1967 with no education beyond compulsory level median age was 21.9 years, while it was 30.7 years among university-educated women with higher degrees. Compared to women born a few decades earlier the difference in median age among those with lower education and those with the most education has increased. For instance median ages were respectively 20.6 years and 28.4 years among those born in 1950 in the two education groups.
New fertility trends in Norway
© 2000 Max-Planck-Gesellschaft ISSN 1435-9871