Hannaliis Jaadla, Alice Reid, Eilidh Garrett  

Concerns about low and declining fertility are common in the media and feature in public discussions around much of Europe and South East Asia. The size of the future work force and the sustainability of pension systems in years to come both depend on the number of children born today. In England and Wales, the Total Fertility Rate (TFR) fell to 1.49 children per woman in 2022, and 2023 was the first year in nearly half a century – and only the second in the last 250 years – when there were fewer births than deaths.  

Image created by Ellie Shipman for a Cambridge Creative Encounters project.

There are also frequent media reports about rises in women’s average age of childbirth, which has now reached its highest level since 1938, when the recording of mother’s age on birth certificates began. A previous blog has described the long-term patterns in levels of fertility (TFR) since 1537, and discussed the role of marriage in keeping fertility rates moderate before the late 19th century. This blog tells the story of changes in fertility and the age pattern of childbearing in England and Wales from the first major fall of birth rates in the late 19th century until today.  

People from a poor area of Victorian Sheffield, date unknown. Reproduced by permission of Sheffield Libraries and Archives.

The relationship between levels of fertility and average age at childbirth 

Fertility levels in the last 10 years have declined to historically low levels, and at the same time average age at childbirth for women has increased to almost 31 years. This is a marked contrasts to the 1960s, when levels of fertility were high and average age at childbirth was low, close to 26 years.

However, this does not mean that when fertility was even higher in the past, the average age at childbirth was even lower. In fact, before the fertility transition, high levels of fertility were accompanied by high average ages at childbirth, and demographers in the 1960s thought that previous fertility decline would have been closely accompanied by a decline in the average age of childbirth. The reality was rather more complex, as shown in the graph below. 

The average number of children, or TFR, represents the level of fertility for an individual woman, and average age of childbearing represents the timing of births within her reproductive life. Interestingly, the changes in these two measures do not go hand in hand. When fertility was high and an average woman had between four and six children, even if she had her first birth in her mid-20s, her youngest child would have been born in her late 30s or even her early 40s, and this meant that the average age of childbirth was necessarily high. It is likely that the average age at which mothers had their children was in their 30s as far back as the 16th century, or even earlier. 

Demographers thought that fertility decline would be accompanied by a simultaneous decline in the average age of childbearing, because couples would form an ideal number of children and stop childbearing once this was reached. This would concentrate childbearing into younger ages and bring down the average age at which it happened. In fact, during the early years of the fertility transition (1880–1911) it was not just older women who reduced their fertility – women of all ages had fewer children, and the average age of childbearing barely changed.  

The lack of detailed data means that we don’t know much about the average age of childbirth in the later stage of the fertility transition (1911–1930), but it clearly did decline during this period, as demographers had predicted. The next period, however, exhibited far from predictable trends. 

Ernst Neuschul, Untitled (Two Mothers and Babies), 1943. Glynn Vivian Art Gallery.

In the UK the Baby Boom had two peaks; one starting during, but peaking immediately after, World War II: this was characterised by a temporary increase in the average age at childbearing, probably associated with catch-up childbearing by women who had postponed births during the Great Depression of the 1930s and in the early years of the war. The second peak was in the 1960s, and was accompanied by continued reduction in the average age at childbearing (at least partly due to a decline in the average age at marriage).  

Since the early 1970s, the average age of childbearing has increased virtually continuously, despite relatively small changes in the level of fertility. Once fertility is low, there is scope to change the timing of childbearing without changes in the level of fertility. The average age at which women have their children (calculated using the births of all their children) is now back to levels similar to the pre-industrial past, but the big difference is that the average age of first birth is now at a historic high.  

The ‘shape’ of fertility 

It is possible to get a much better sense of the timing of fertility using age-specific fertility curves, as shown in the graph below. These curves show the average number of children born to women in different age groups across their reproductive years, i.e. from around 15 to 50 years of age. The height of each curve tells us about the number of children being born, while the shape of the curve indicates the timing of their births. The graph shows curves for the pre-industrial period, every 10 years between 1881 and 1911, and then every 20 years between 1941 and 2021.  

The pre-industrial curve (1600–1740) is high and wide, indicating comparatively high numbers of births spread across the range of mothers’ ages. Women’s relatively low fertility in their teens and early twenties is because the average age at marriage for women in historical England was over 25. From 1881 until 1911, the height of the curves shifts downward with each decade, but little change occurs in the general shape of the curves. This suggests that as total levels of fertility began to decline, from 4.5 children per woman to 3, there was little change in the general timing of childbearing, which was still spread across the age range.  

By 1941, the shape of fertility had changed: women in their 20s had higher fertility relative to women in their 30s than had been the case previously. A reduction in the number of children to very small numbers (TFR in 1941 was only 1.7 children per woman) was associated with the concentration of childbearing into the early years of marriage.  

The 1961 curve is a perfect example of ‘the shape of fertility’ during the second phase of the Baby Boom. Although fertility had rebounded considerably, childbearing was even more concentrated among young women, with women in their 30s and 40s much less likely to bear children than preceding generations, but women in their early 20s more likely to do so. These patterns are linked to a combination of unprecedentedly young ages at marriage and the fact that the Baby Boom was not characterised by an increase in large families, but by a reduction in the numbers of people remaining childless or having a single child.  

Claudia Williams, Mother and Babies (1963).

By the 1980s levels of fertility had fallen once again to 1.8 children per woman, although they were now more likely to be born when their mothers were in their late 20s and early 30s. By the early 2000s, the rates had fallen even lower and the curves in the graph continue the gradual shift towards the right that was already visible in the 1980s, indicating a postponement of births to later ages. For the first time since the pre-industrial era, the 2021 fertility curve peaked amongst 30–34 year olds.  

Why did age at childbirth not fall when fertility started to decline? 

This exploration of changes in the levels and timing of fertility reveals an unanswered question: why were the early stages of fertility decline (1880s–1911) not accompanied by a concentration of childbirth into younger ages? The answer lies in the reliability of contraceptive methods. 

Since the mid-20th century, couples have had the option of using different effective contraceptives to plan and to limit the size of their family. This was not the case in the past. Evidence from both demographic measures and oral histories suggests that much of the historical fertility decline (1870–1930) was not due to the introduction and use of contraceptive methods such as the douche, sponge, caps, or condoms – which became increasingly available from the 1870s onwards – but was achieved instead through traditional, but relatively unreliable, methods such as withdrawal (which remained the predominant method until the 1930s), and abstinence. This meant it was difficult for couples to form a ‘target’ or ‘ideal’ number of children, although women in particular frequently expressed a desire to have fewer, or more widely spaced, children.  

Contraceptive cellulose sponge, ‘Clinocap’ brand. Wellcome Collection.

In one of the letters in Margaret Llewelyn Davies’ collection Maternity: Letters from working women (1915), one of the women writes about how she realised, after having her third child, that she really was not ready to have any more: 

“After [a miserable third pregnancy], I said to a friend one day, “If only I could feel that this was my last, I would be quite happy.”  “Well,” she said, “why don’t you make it your last?” and she gave me advice. As a result of this knowledge, I had no more babies for four and a half years…. My general health and nerves were much better. My health improved, and people said I looked years younger, and I found life a happy place… I had a fight with my conscience before using a preventative. But I have no qualms now. I feel I have better health to serve my husband and children, and more advantages to give them, while if another comes along, we will hail it with pleasure, as we did our last, instead of looking on it as a burden.”

This woman used one of the emerging barrier methods, but it was not completely effective as, after a gap of four and a half years, she did conceive again. Her letter suggests that she did not have a ‘target’ number of children, as assumed by the demographers of the 1960s, but developed an adaptive and flexible attitude, consistent with unreliable contraception, while trying to reduce the number of children she did bear. Age patterns of fertility suggest this attitude was characteristic of the early stages of the fertility decline, and that the ‘two-child family’ norm was only made achievable by modern contraceptive methods during the second half of the 20th century.  

How do we know?

The measurement of TFR has been discussed in a previous blog. Calculation of the average age of fertility and age-specific fertility rates are more data–demanding than many other demographic measures. We need to know the number of births by age of mother, and we also need to know the number of women in childbearing ages.

Measures for the pre-industrial period are based on marital histories created by linking baptisms in parish registers, together with estimates of proportions married. Measures for 1881–1911 have been estimated using census data of women living with their children and making adjustments for child mortality and children who might be living elsewhere – a technique called the ‘own children method’. From 1938 onwards, birth registration in England and Wales started to record age of the mother and we rely on published data. 

David Peat, Back Court (1968). University of St Andrews.

Further reading

• Llewelyn Davies, M., Maternity: Letters from working women (1915).
• Sigle-Rushton, W., ‘England and Wales: Stable fertility and pronounced social status differences’, Demographic Research 19 (2008), 455-502.
• Szreter, S., and Fisher, K., Sex before the sexual revolution: intimate life in England 1918-1963 (Cambridge University Press, 2010).

Tags: age at childbirth, birth rate, contraception, demography, fertility, marriage, total fertility rate

| This entry was posted on Thursday, November 28th, 2024 at 12:02 am and is filed under Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.