Alice Reid

In 2011 David Cameron asked “Do we have the determination to confront the slow-motion moral collapse that has taken place in parts of our country these past few generations?” He went on to present a list of examples of the moral collapse he was talking about: “Irresponsibility. Selfishness. Behaving as if your choices have no consequences. Children without fathers.” This focus on a rise in lone parenthood (particularly lone motherhood) as an indicator of the erosion of moral fibre has been a popular refrain over recent decades, particularly among the political right wing, and has often been accompanied by calls to bring back Victorian values. An article in the Telegraph in 2017 focused on the rise in lone parenthood since Britain joined the EU in 1972, and suggested that Brexit was an opportunity to reverse this social decline. 

Miles MacGregor, Clothed with the Sun (2011).
Bell Lane, Bristol. An homage to single mothers in a country that has the highest rate of single mothers in Europe.

There is no doubt that the percentage of families which are headed by a lone parent has increased since the mid-20th century, and this has often been equated with the breakdown of the nuclear family system. However, it is not clear that the nuclear family is actually in decline. Most children are still living in two parent homes, and the percentage of lone parent families in the 19th century was not very different to the percentage today – although as explained below, such families were very differently formed. 

Changing makeup of families with children 

The graph below shows the percentages of family units with dependent children (ie at least one parent plus one or more children under 16 or in full-time education) headed by lone mothers or lone fathers. The second half of the 20th century clearly saw a very large increase in lone parent families, rising from around five percent of families with dependent children in 1951, to between 20 and 25 percent in the early 21st century, with the bulk of this rise accounted for by lone mother families.  

Looking further back in time, however, it is clear that the situation in the mid-20th century was not, as is often assumed, a continuation of some historic constant. In fact, the situation was very different in the second half of the 19th century, when levels of overall lone parenthood were very similar to those today. The balance between lone mothers and lone fathers, however, was rather different. In the 19th century there were far more lone fathers, and this gives a clue about important differences in the drivers behind lone parent households.  

Today the majority of lone parent households are formed by divorce or separation, but in the 19th century divorce was virtually unheard of, and although separation (or abandonment) did happen, it was rare. Instead, lone parent families were formed mainly by the death of one or other parent. There were more lone mother households partly because male mortality is generally higher, and husbands tended to be a bit older than wives, but predominantly because widowed men were more likely to re-marry than widowed women, or to send their children to live with other relatives. In more recent decades there are more lone mothers, because unlike dissolution of partnership through mortality, if a couple separates there is a choice about where the children live, and it is more common for them to live with the mother.  

It is worth considering whether a change from mortality to divorce or separation as a cause of partnership dissolution is a bad thing or not. While you could argue that separation is not good for children, it is also worth remembering that separation tends to only split up unhappy couples, whereas mortality is non-selective – it is as likely to form lone parent families out of happy couples as unhappy.  

Antony Donaldson, Family in a Large Room. UCL Culture.

Of course, lone parent families are not only caused by the dissolution of existing partnerships; some are the result of births to unpartnered women. As we showed in a previous blog, there has been a very large increase in the percentage of births to unmarried women since the 1960s, but the bulk of this increase is accounted for by births to co-resident couples. In other words, the decrease in births within marriage has been largely compensated for by an increase in births to cohabiting couples. There has been less change in the percentage of births to women where the father is not recorded on the birth certificate, or is recorded as living at a different address.

Nevertheless, cohabiting couples are more likely than married couples to split up, and this could contribute to an observed increase since the early 1980s in the proportion of lone parent families which are headed by a single woman. 

Re-constituted families

Changes in family forms are not just about lone parent families. Increases in divorce have also brought increases in re-marriage or re-partnering, leading to more re-constituted families. But reconstituted families were also common in the past. Although it’s difficult to get a handle on the precise numbers of families with dependent children, as many as 30 percent of people marrying in the late 16th and 17th centuries were widowers or widows. As already mentioned, widowed men were much more likely to have remarried than widowed women, so there were more stepmothers than stepfathers, and this could be why so many fairy stories contain wicked stepmothers (the fact that there were fewer female writers might also play a role).  

Firs Sergeyevich Zhuravlev, The Stepmother (1874).

What happened in the 20th century?

In the longer–term context, then, it is the mid-20th century which looks unusual, and it is worth considering why. Part of the reason is decline in mortality which, together with the decline in fertility from around five to fewer than two births per woman, formed the ‘first demographic transition’ (see our previous blogs on long, and even longer-term trends in children per woman, and how fertility decline was achieved).

Declines in mortality gathered pace from around 1870, causing an increase in life expectancy at birth from 42 in 1841 to 72 by the 1960s. As explained in an earlier blog, this does not mean that in 1841 most, or even half, of the population died by the age of 42, but death during early and mid-adulthood was much more common in the mid-19th century than in the mid-20th. By the later date, reduced mortality had removed most the lone parent families produced by widowhood. 

Another part of the reason that the mid-20th century looks unusual is the fact that divorce was still difficult. Although divorce for ‘common people’ was possible in England and Wales from 1857, it was a difficult and costly process, and it was more difficult for women to petition for divorce than for men. Women had to prove adultery and desertion or cruelty on the part of their husband, whereas men could cite adultery alone.

Melmiro de Almeida, Arrufos (1887).

This double standard was removed in 1923, and additional grounds for divorce were added in 1937, but the number of divorces remained low until the Divorce Reform Act of 1969, which simplified the divorce process and removed the need to prove matrimonial offences.  

A rapid rise in divorces was enabled by the 1969 legal reform (which came into force in 1971), but its growing popularity over the later decades of the 20th century has also been linked to a change in ideas about relationships and family building in relation to other ways of becoming fulfilled adults. This change is known as the ‘second demographic transition’ and is characterised by declines in marriage and rises in cohabitation, as well as increases in divorce and re-partnering. Since the 1990s divorce rates have plateaued, but as marriage has been increasingly replaced by cohabitation, divorce provides an increasingly incomplete picture of partnership dissolution.  

The baby boom of the 1950s and 1960s might also have contributed to the low proportions of lone parent families. This was, to a large extent, driven by a marriage boom which combined catch-up marriages which might have been delayed by the depression of the 1930s and the war, and new cohorts who were getting married at ages which were unusually young for the British population. This means that a higher proportion of families in the 1950s and 1960s would have been relatively recently formed, and had less time to have split up to create lone parent families.  

Reading history sideways

There have been remarkable declines in the popularity of marriage over the past 50 years, with increases in divorce, cohabitation, and births outside marriage. But although this has resulted in a substantial increase in lone parent and reconstituted family households since the 1960s, over the longer term it has not.

The use of the 1960s as a reference point overstates a view of the nuclear family which involved a couple marrying young and staying married for decades. As explained in a previous blog, far from representing a typical view of history, those marrying during the 1950s and 1960s were unusual because they married younger and had more children than previous or subsequent generations. Mortality among adults had declined substantially, so lone parenthood due to widowhood was at a minimum. It was not for nothing that this period has been called the Golden Age of the Family.  

Photo by Boston Public Library on Unsplash

Those using increases in lone parenthood since the 1960s as evidence of a decline in the nuclear family are, as explained in the previous blog, guilty of ‘reading history sideways’ – assuming that the situation in 1960s was typical of history up to that date, instead of being a highly unusual situation. Although the nuclear family has been a long-standing feature of British society (as explained in another previous blog) this did not mean that every family was always made up of two parents with their children.  

Nevertheless, despite these longer-term continuities, there have definitely been changes which have profound implications for how people experience family life, such as co-parenting from different households. And lone parents and their children continue to face socioeconomic and other disadvantages, today, as in the past.  

How do we know?

The figures for 1851 to 1911 are derived from individual-level census records, with individuals classified into family units. Dependent children are counted as children under age 16, and those aged 16 and over who are in full–time education or described as ‘scholar’ in the census. Because this process is based on relationships to the head of household rather than between all people within a household, some family units within households may be missed. This is particularly likely for lone family units living with the lone parent’s parents, as it can be difficult to know which son or daughter is the parent of a grandchild.  

Although the 1951 figure is also derived from published census data, slight differences with other years may be due to the facts that the 1951 figure relates to the whole of Great Britain rather than just England and Wales, and does not include dependent children over the age of 16 as part of family units (this will reduce the number of family units for couples as well as lone parents). The figures for 1971 to 1995 are derived from the General Household Survey, and those from 1996 onwards use the Labour Force Survey. Differences in the questions asked and the way categories have been grouped may introduce small differences in the estimates. In all cases couple families include married, civil partnered, and cohabiting couples in both heterosexual and same-sex unions, although these may not all be identifiable in the earlier data.  

Further reading 

• Berrington, A., ‘The changing demography of lone parenthood in the UK.’ Working Paper 48, ESRC Centre for Population Change Working Papers (2014). https://www.cpc.ac.uk/docs/2014_WP48_The_changing_demography_of_lone_parenthood_Berrington.pdf 
• Haskey, J., ‘One-parent families and their dependent children in Great Britain’. Population Trends 91 (1998), 5–14. 
• Lesthaeghe, R., ‘The second demographic transition: A concise overview of its development.’ Proceedings of the National Academy of Sciences 111:51 (2014), 18112–18115. https://doi.org/10.1073/pnas.1420441111 
• Schürer, K., Garrett, E. M., Jaadla, H., and Reid, A., ‘Household and family structure in England and Wales (1851–1911): Continuities and change.’ Continuity and Change 33:3 (2018), 365–411. https://doi.org/10.1017/S0268416018000243 
• Thane, P., ‘Unmarried Motherhood in Twentieth‐Century England.’ Women’s History Review 20:1 (2011), 11–29. https://doi.org/10.1080/09612025.2011.536383 

Tags: demographic transition, demography, divorce, family, family history, historical demography, lone parenthood, marriage, social history, women's history

| This entry was posted on Thursday, July 24th, 2025 at 12:03 am and is filed under Alice Reid. 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.