Back to Materials
The Marriage Advantage: Why Children Thrive with Married Biological Parents

The Marriage Advantage: Why Children Thrive with Married Biological Parents

15^th December 2025

In brief: Children raised by their own married mother and father do better, on average, than children in any other kind of family - more likely to keep both parents, to do well at school, to stay free of mental disorder, to be safe at home and to escape poverty. https://futureofchildren.princeton.edu/sites/g/files/toruqf2411/files/media/marriage_and_child_wellbeing_revisited_25_2_full_journal.pdf In Britain the first difference is stability. Married parents with dependent children split up at 1.9 per cent a year, cohabiting parents at 6.5 per cent - 3.4 times the rate. https://marriagefoundation.org.uk/research/married-poor-more-stable-than-unmarried-rich/ The gap is there in every income band. A June 2026 re-analysis of the Millennium Cohort found between half and two thirds of the stability gap between parents who marry and parents who never do still there after allowing for 27 differences between the couples, income and education among them. https://marriagefoundation.org.uk/wp-content/uploads/2026/06/PhD-The-Timing-of-Marriage_digital-1.pdf None of this diminishes the heroic work of many single parents, but it does mean that marriage gives children the most stable foundation there is, and that a country serious about children's wellbeing would say so and act on it. The wider case for marriage is on our Why Marriage Matters page, at https://www.c4m.org.uk/marriage-support/why-marriage-matters, and the national figures are on Marriage Facts and Figures, at https://www.c4m.org.uk/marriage-support/marriage-facts-and-figures.

STABILITY COMES FIRST

Children thrive on stability, and so much of the rest follows from it.

Married parents with dependent children split up at 1.9 per cent a year and cohabiting parents at 6.5 per cent, on the Marriage Foundation's figures. https://marriagefoundation.org.uk/research/married-poor-more-stable-than-unmarried-rich/ Unmarried cohabiting parents accounted for around 19 per cent of couples with dependent children in 2010, on the Office for National Statistics (ONS) figures the Marriage Foundation used, but roughly half of all family breakdown, according to its briefing note of December 2013, at https://marriagefoundation.org.uk/wp-content/uploads/2019/09/MF-briefing-note-proportion-of-family-breakdown-from-cohabitation-v2.pdf.

The gap isn't money in disguise. The Marriage Foundation's May 2022 study, built to test exactly that, found the gap present in all five income quintiles. https://marriagefoundation.org.uk/research/married-poor-more-stable-than-unmarried-rich/ Among the very poorest parents, 8.9 per cent of cohabiting couples separated each year against 2.8 per cent of married couples. The poorest married parents were slightly more stable than the richest cohabiting parents. https://marriagefoundation.org.uk/research/married-poor-more-stable-than-unmarried-rich/ The Centre for Social Justice studied 3,286 first-time parent couples followed by the Millennium Cohort Study over 14 years. Once other social and economic differences between the couples were allowed for, the chance of splitting up was 26 per cent for the poorest parents who married, against 46 per cent for the richest who never married. https://www.centreforsocialjustice.org.uk/wp-content/uploads/2026/07/CSJ-Poverty_and_Family_Stability.pdf

Over a childhood the gap compounds. Marriage Foundation analysis of the Millennium Cohort Study - the national study following around 19,000 children born at the millennium - finds that 46 per cent of UK first-born children aren't living with both natural parents by age 14, and that the break-up of unmarried parents, not divorce, accounts for over two thirds of them. Among the parents still together at that 14th birthday, 84 per cent are married and only 16 per cent are cohabiting. https://marriagefoundation.org.uk/research/source-of-family-breakdown/ https://cls.ucl.ac.uk/cls-studies/millennium-cohort-study/ A June 2026 re-analysis of first-time parents in the same cohort, published by the Marriage Foundation with the Centre for Social Justice, used the full sample rather than the three quarters the Institute for Fiscal Studies had analysed. Once 27 differences between the couples were allowed for, income, education and ethnicity among them, the chance of separating by the child's 14th birthday was 45 per cent for parents who never married, against 26 per cent for those who married before conception. Those differences left between half and two thirds of the original gap unexplained. https://marriagefoundation.org.uk/wp-content/uploads/2026/06/PhD-The-Timing-of-Marriage_digital-1.pdf

The early years tell the same story. The Institute for Fiscal Studies found in 2010 that 26 per cent of cohabiting couples had separated by their child's third birthday against 7 per cent of married couples, and that by the fifth birthday 27 per cent of the couples who had been cohabiting at the birth were no longer living together, against 9 per cent of the married. https://ifs.org.uk/sites/default/files/output_url_files/bn107.pdf The Centre for Longitudinal Studies found that 88 per cent of married parents were still living together when their child was five, against 67 per cent of parents who had been cohabiting at the birth. https://cls.ucl.ac.uk/wp-content/uploads/2017/05/01_briefing_web.pdf The Government's own Family Stability Indicator, in its latest release, found 82 per cent of babies under one living with both birth parents in 2013-14, against 63 per cent of 15-year-olds. https://assets.publishing.service.gov.uk/media/5a80875e40f0b62305b8bad8/social-justice-family-stability-indicator-2013-2014.pdf

SCHOOL AND ADULT LIFE

Family structure shows up in the exam tables. Marriage Foundation analysis of GCSE outcomes found that among children born to married parents, 26 per cent of boys and 19 per cent of girls failed to reach a grade 4 in English and Maths. Among children born to parents who never married, the failure rates were 39 per cent for boys and 34 per cent for girls. https://marriagefoundation.org.uk/wp-content/uploads/2023/05/MF-report-Families-and-Education.pdf Among children born to single parents, 54 per cent of boys and 37 per cent of girls failed. With every other factor allowed for, girls whose parents never married were still significantly more likely not to pass both subjects, 27 per cent against 20. https://marriagefoundation.org.uk/wp-content/uploads/2023/05/MF-report-Families-and-Education.pdf

Edward Davies, the Centre for Social Justice's director of policy, drew on its analysis of the Family Resources Survey in The Spectator in August 2025. "What does correlate with exam results in every ethnicity is marriage rates," he wrote. https://www.spectator.co.uk/article/the-statistic-that-explains-why-white-working-class-children-do-so-badly/ "The more stable your home, the better you do." Just two in ten children in the poorest white quintile live with married parents, against almost nine in ten in the wealthiest. https://www.spectator.co.uk/article/the-statistic-that-explains-why-white-working-class-children-do-so-badly/

The advantage lasts into adult life. The Institute for Family Studies, analysing Britain's 1958 and 1970 birth cohorts, found that people born to married parents, whatever their family and social background, were 23 per cent more likely to have been to university, 10 per cent more likely to have married themselves and 16 per cent less likely ever to have received government benefits. https://ifstudies.org/blog/the-long-term-benefits-of-marriage-evidence-from-the-uk And the harm of a break-up hasn't faded as divorce became common. Across Britain's 1946, 1958 and 1970 birth cohorts, children whose parents divorced or separated were 30 to 40 per cent more likely to leave school with below-median attainment. https://static.cambridge.org/content/id/urn:cambridge.org:id:article:S0047279499005693/resource/name/S004727949900569Xa.pdf Comparing the 1958 and 1970 cohorts, Wendy Sigle-Rushton, John Hobcraft and Kathleen Kiernan found "little evidence of any change in the size of the relationship as divorce became more commonplace". https://pubmed.ncbi.nlm.nih.gov/16235607/ In America, among those born in the 1990s, 46.6 per cent of children whose mothers were married when they were born had a four-year degree by 25, against 17.3 per cent of those whose mothers weren't, according to a September 2026 working paper for the National Bureau of Economic Research, at https://www.nber.org/system/files/working_papers/w35765/w35765.pdf.

MENTAL HEALTH, BEHAVIOUR AND FATHERS

England's national survey of children's mental health found that in 2017, 6.2 per cent of 5 to 10 year olds whose parent was married had a mental disorder, against 12.2 per cent where the parent was cohabiting and 17.0 and 18.2 per cent where the parent was a single or previously married lone parent. https://files.digital.nhs.uk/85/7FFC57/MHCYP%202017%20Predictors.pdf Among 11 to 16 year olds the gap was wider still, 11.6 per cent against 27.4 per cent with a single lone parent. At both ages, children of a lone parent were still more likely to have a disorder after the survey allowed for how well the family got on, the parent's own mental health and other differences. https://files.digital.nhs.uk/85/7FFC57/MHCYP%202017%20Predictors.pdf The follow-up surveys since, to 2023, haven't published the figures by family type, so 2017 is still the latest year. https://files.digital.nhs.uk/2D/D88231/mhcyp_2023_tab.xlsx The Government should start publishing them again.

The Millennium Cohort shows the same order at age three. Kathleen Kiernan and Fiona Mensah found high levels of behaviour problems in 4.6 per cent of children living with their married parents, 9.7 per cent with cohabiting parents and 14.6 per cent with a lone parent. https://eprints.whiterose.ac.uk/id/eprint/10118/1/kiernanand_coversheet.pdf Later analyses of the same cohort found that children in stable married families had fewer of the acting-out problems researchers call externalising behaviour than children in other family arrangements, independent of their families' other characteristics. https://link.springer.com/article/10.1007/s10826-021-02159-z At 14, children of cohabiting biological parents still had significantly more of these problems than children of married biological parents, even after family closeness, money, parents' health and behaviour, children's own health and family stress were all allowed for. https://pmc.ncbi.nlm.nih.gov/articles/PMC9779363/ The Centre for Social Justice's 2020 report Family Structure Still Matters found an outcome gap between children of married and cohabiting parents even after income and education were controlled for, and says family structure has a greater bearing on children's externalising behaviours - the problems linked to school attainment, criminal justice involvement and emotional development - than education or poverty. https://www.centreforsocialjustice.org.uk/newsroom/new-report-family-structure-still-matters

A 2016 study for the Ministry of Justice found the same order at age 11. https://dera.ioe.ac.uk/id/eprint/25543/1/child-outcomes-after-parental-separation.pdf Children of continuously married parents "tended to have the best outcomes at age 11, followed by children of parents who were cohabiting at the time of birth and remained together", and children of separated parents the worst. After a separation, contact with the parent who had left tended to be higher, on a number of measures, where the parents had been married. https://dera.ioe.ac.uk/id/eprint/25543/1/child-outcomes-after-parental-separation.pdf Lidia Panico and colleagues, also using the Millennium Cohort, found that whether parents were married, cohabiting or single, how stable that was, and the two together "are important components of family structure trajectories which correlate to children's early physical health" at age five, with family stress appearing to lie behind the link. https://doi.org/10.1016/j.socscimed.2019.05.006

The pattern isn't only British. In Sweden, a population-based study of almost a million children, published in The Lancet in 2003, found that those living with a single parent were at least twice as likely as those living with two to suffer psychiatric disease in childhood and to attempt suicide, after allowing for the family's socioeconomic position and the parents' own addiction or mental illness. https://pubmed.ncbi.nlm.nih.gov/12559862/ A 2019 meta-analysis of 54 studies and 506,299 people found that those whose parents had divorced had 1.29 times the odds of depression and 1.48 times the odds of suicidal thinking. https://pubmed.ncbi.nlm.nih.gov/31622869/

Fathers are half of this story. A literature review for the Government Equalities Office in January 2021 found that "Fathers' involvement in childcare improves children's emotional well-being, cognitive development and academic achievement". https://www.gov.uk/government/publications/childcare-shared-care-and-well-being-outcomes-for-families/shared-care-fathers-involvement-in-care-and-family-well-being-outcomes-a-literature-review In a study of up to 8,409 children in the Avon Longitudinal Study of Parents and Children, a father's absence from home in early childhood was strongly linked to depression at 24 and to higher depressive symptoms through adolescence and early adulthood. https://pmc.ncbi.nlm.nih.gov/articles/PMC10666570/ A married father is the father most likely to stay in his children's daily lives, because married parents are the parents most likely to stay together. Yet more than one in five dependent children in the UK, 3.2 million in 2025, live in a lone-parent family. https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/families/datasets/familiesandhouseholdsfamiliesandhouseholds Lone mothers head 84.6 per cent of lone-parent families. https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/families/bulletins/familiesandhouseholds/2025

Of 200 children sentenced to custody in the second half of 2008, 76 per cent were known to have had an absent father, according to the Prison Reform Trust, at https://prisonreformtrust.org.uk/wp-content/uploads/old_files/Documents/PunishingDisadvantage.pdf. In America, young men who were growing up without their biological father in the home in 1997 were about twice as likely to have spent time in jail by around 30, 21 per cent against 10 per cent, and the gap held after allowing for family income, race, the mother's education, age and test scores, according to the Institute for Family Studies in June 2022. https://ifstudies.org/blog/life-without-father-less-college-less-work-and-more-prison-for-young-men-growing-up-without-their-biological-father

The law is moving the other way. In October 2025 the Government announced it would repeal the presumption in the Children Act 1989 that a parent's involvement in a child's life furthers the child's welfare. https://www.gov.uk/government/news/government-action-to-protect-children-from-abusive-parents Clause 17 of the Courts and Tribunals Bill makes the repeal, and the Bill is listed for its report stage in the House of Commons on 13 October 2026. https://bills.parliament.uk/bills/4083 The Government says the repeal means children "will be protected from harm". https://www.gov.uk/government/news/government-action-to-protect-children-from-abusive-parents But the presumption already gives way wherever there's evidence that a parent's involvement, in whatever form, would put a child "at risk of suffering harm". So in all other cases, the repeal takes away the courts' starting point that a child is better off with both parents involved. https://www.legislation.gov.uk/ukpga/1989/41/section/1

THE SAFEST HOME

The safest home for a child is with married biological parents. https://www.acf.hhs.gov/sites/default/files/documents/opre/nis4_report_congress_full_pdf_jan2010.pdf In the United States, the Fourth National Incidence Study of Child Abuse and Neglect found that children living with their two married biological parents had the lowest rate of maltreatment of any family arrangement. Children whose single parent lived with an unmarried partner suffered abuse at more than 10 times the rate of children with married biological parents - 45.8 per 1,000 against 4.3. https://www.acf.hhs.gov/sites/default/files/documents/opre/nis4_report_congress_full_pdf_jan2010.pdf In Canada, Martin Daly and Margo Wilson found that pre-school children living with one natural parent and a stepparent were 40 times as likely to be recorded as child-abuse cases as children living with both natural parents. https://www.martindaly.ca/uploads/2/3/7/0/23707972/daly___wilson_1985_child_abuse.pdf They later reported that Canadian stepfathers beat children under five to death at more than 120 times the rate of genetic fathers. https://www.martindaly.ca/uploads/2/3/7/0/23707972/cinderella_effect_2008.pdf

POVERTY BY FAMILY TYPE

In 2024/25, 27 per cent of children in single-parent families were in deep material poverty - going without at least four of 13 essentials for lack of money - three times the rate for children in couple families, and 41 per cent were in relative poverty after housing costs. Half of all children in deep material poverty were in single-parent families, according to the Department for Work and Pensions' child poverty baseline report, at https://www.gov.uk/government/publications/our-children-our-future-tackling-child-poverty-monitoring-and-evaluation-baseline-report/annex-a-vulnerable-and-high-risk-groups.

Meanwhile marriage itself is becoming a possession of the better-off. The Marriage Foundation's March 2025 analysis found 71 per cent of new parents in the highest earning group were married in 2022, against 35 per cent in the lowest earning classified group - a 36 percentage point gap, double the gap present in 1988. https://marriagefoundation.org.uk/wp-content/uploads/2025/03/MF-paper-Marriage-gap-March-2025.pdf The stability marriage brings is concentrating where the money already is. Sara McLanahan found the same pattern in America and called it "diverging destinies", with children born to the most educated mothers gaining resources in their parents' time and money, and those born to the least educated losing them. https://pubmed.ncbi.nlm.nih.gov/15622946/

America shows the way out. Young adults who finished high school, worked full time and married before having children, in that order, were almost never poor in their mid-thirties - 3 per cent, against 52 per cent of those who missed all three steps. Of those who grew up without both biological parents but followed the sequence, 95 per cent weren't poor, according to the Institute for Family Studies and the American Enterprise Institute in May 2022, at https://ifstudies.org/ifs-admin/resources/reports/successsequencedisadvantagedya-final.pdf. Family structure matters beyond the front door too. Raj Chetty's Opportunity Atlas, built from the tax records of more than 20 million American children, found a strong link between the share of single-parent households in a neighbourhood and children's chances of rising. Children of single parents themselves rise further where fewer of the families around them are single-parent households. https://www.nber.org/system/files/working_papers/w25147/w25147.pdf

The bill for the country is at least £64 billion a year. That's the Centre for Social Justice's estimate of what family breakdown costs the state, an earlier figure of £51 billion uprated for inflation, in its Lost Boys report of March 2025, at https://www.centreforsocialjustice.org.uk/wp-content/uploads/2025/03/CSJ-The_Lost_Boys.pdf.

SELECTION OR CAUSATION?

The strongest objection to everything above is that marriage doesn't make families stable, but stable people marry. The Institute for Fiscal Studies concluded in 2013 that the differences between married and cohabiting parents, and between their children, "mainly or entirely reflect the fact that different types of people choose to get married". https://ifs.org.uk/sites/default/files/output_url_files/r87.pdf Its 2010 briefing note had found the gap in separation by the third birthday still large and significant after the mother's background was allowed for, an odds ratio of 4.3. https://ifs.org.uk/sites/default/files/output_url_files/bn107.pdf Once the couple's education, income, housing and relationship quality were held constant as well, it fell to 2.1 percentage points, an odds ratio of 1.5, and by the fifth birthday it was no longer significant. Yet the Institute itself warned that by controlling for such things "we might inadvertently 'control away' some of the indirect effects of marriage". https://ifs.org.uk/sites/default/files/output_url_files/bn107.pdf

Its companion study of the same children found that those born to cohabiting parents had more social and emotional difficulties at three than children born to married parents. https://ifs.org.uk/sites/default/files/output_url_files/comm114.pdf Children whose cohabiting parents had split did worst. The gap narrowed but held as the parents' education, income and housing were allowed for, and went only when their relationship quality was held constant. Its authors warned that "if marriage itself actually causes higher income, better relationship quality or better parenting practices, then controlling for these indirect effects will lead to an understatement of the possible benefits of marriage". https://ifs.org.uk/sites/default/files/output_url_files/comm114.pdf The 2013 report's own figures couldn't explain away the gap in teenagers' risky and antisocial behaviour. https://ifs.org.uk/sites/default/files/output_url_files/r87.pdf And the Institute's study of the parents left out a quarter of the sample. The June 2026 re-analysis put it back, and found between half and two thirds of the gap between parents who marry and parents who never do left unexplained. https://marriagefoundation.org.uk/wp-content/uploads/2026/06/PhD-The-Timing-of-Marriage_digital-1.pdf

The same answer meets the snapshot surveys whose married-cohabiting gaps fade once poverty, a mother's depression or family functioning are held constant. They class families as they stand on the day of the survey, so cohabiting couples who have already split are counted as lone parents. The 2017 survey notes that "The stability of the cohabitation was not assessed". https://files.digital.nhs.uk/85/7FFC57/MHCYP%202017%20Predictors.pdf They also hold constant the family's income, the mother's health and how well the family gets on, which a break-up can itself damage.

The studies best able to tell cause from coincidence point the same way. Among 1,613 pairs of young American twins and siblings, Erin Horn and colleagues found that marriage-like relationships, married or cohabiting, "apparently are (despite age, gender, ethnicity, genes, and rearing environment) causally protective against internalizing and externalizing problems", that is, emotional and behaviour problems. https://pmc.ncbi.nlm.nih.gov/articles/PMC3645280/ Identical twins who were cohabiting were 67 per cent more likely than their married co-twins to have broken the law in the previous year. The link with physical health, they found, was completely explained by who marries. They call their design "perhaps the strongest nonexperimental evidence" that the effects they found are causal. https://pmc.ncbi.nlm.nih.gov/articles/PMC3645280/

The same test has been run on children. Paul Amato's 2001 meta-analysis of 67 studies found children of divorce scoring lower on academic achievement, conduct, psychological adjustment, self-concept and social relations. https://pubmed.ncbi.nlm.nih.gov/11584788/ He and Christopher Anthony then compared the same American children before and after their parents' divorce, which removes everything about a family that doesn't change over time. The effects were small, but they judged the evidence that divorce caused them reasonably strong. https://pmc.ncbi.nlm.nih.gov/articles/PMC3956656/ Using American tax and Census records, Andrew Johnston, Maggie Jones and Nolan Pope followed 4.5 million children born between 1988 and 1993 to married parents who later divorced, and compared brothers and sisters who lived through the same divorce at different ages. Divorce early in childhood cut adult earnings by 9 to 13 per cent and raised the risk of a teenage birth by roughly 60 per cent, and of imprisonment and early death by about 40 and 45 per cent. https://www.nber.org/system/files/working_papers/w33776/w33776.pdf Comparing brothers and sisters in the British Household Panel Survey, John Ermisch and colleagues found that life in a single-parent family was usually linked to less education, a greater chance of being out of work and study, earlier parenthood, more distress and more smoking in early adulthood. Most of the harm came with a break-up when the child was five or under, and family structure mattered more than whether the parents had jobs. https://doi.org/10.1111/j.1467-985X.2004.00292.x

In Norway's registers, children whose parents' marriage broke up were still 6 to 13 percentage points less likely to get through the main stages of secondary school, even after differences between their mothers that never show up in the records were allowed for. The authors found these large and significant differences "even in a country like Norway where there is far less economic inequality". https://pmc.ncbi.nlm.nih.gov/articles/PMC2831341/ And following 6,245 Millennium Cohort children, Emla Fitzsimons and Aase Villadsen compared each child before and after a father left. A departure later in childhood raised emotional problems in boys and girls, and behaviour problems in boys. https://pubmed.ncbi.nlm.nih.gov/30660390/ In Denmark, a 2025 study of 6,000 children, using methods built to control for differences between families that can't be observed, found "a substantial increase in emotional and behavioral problems" after parents split, which its authors read as a causal effect. https://academic.oup.com/sf/article/104/1/202/7979179 A second Danish study found that children whose parents separate score 5 to 7 per cent lower on reading tests, and that school does nothing to close the gap. https://academic.oup.com/esr/article/41/5/737/8160879 Comparing the same child before and after, or brothers and sisters with each other, also answers the claim that children were already struggling before their parents parted. Following 15,293 Millennium Cohort families, Flouri, Narayanan and Midouhas tested which comes first and concluded that "father absence seems to be mainly the cause rather than the outcome of child problem behaviour in young UK families". https://pmc.ncbi.nlm.nih.gov/articles/PMC5098165/

The field's reviews agree on the direction. Introducing the Future of Children's 2015 issue on marriage and child wellbeing, Sara McLanahan and Isabel Sawhill wrote that "most scholars now agree that children raised by two biological parents in a stable marriage do better than children in other family forms across a wide range of outcomes". https://futureofchildren.princeton.edu/sites/g/files/toruqf2411/files/media/marriage_and_child_wellbeing_revisited_25_2_full_journal.pdf Sara McLanahan, Laura Tach and Daniel Schneider's 2013 review of the most rigorous studies found that they "continue to find negative effects of father absence", strongest and most consistent for finishing school, children's social and emotional adjustment and adult mental health. https://pmc.ncbi.nlm.nih.gov/articles/PMC3904543/ The objection that stability, not marriage, is what matters has its answer in Wendy Manning's own review of cohabitation. Children born to cohabiting parents, she writes, see their parents break up more often, so that being born into a cohabiting family "sets the stage for later instability". https://pmc.ncbi.nlm.nih.gov/articles/PMC4768758/ Susan Brown's review of a decade's research found that children in two-biological-parent married families "appear to enjoy the best outcomes, on average", and that married stepfamilies and cohabiting biological parents both fall short. She concludes that "both marital status and biological parentage are integral to children's well-being" and that "stability alone is not sufficient to maximize children's outcomes". https://pmc.ncbi.nlm.nih.gov/articles/PMC3091824/ In the same Future of Children issue, David Ribar argued that studies controlling for the routes by which marriage helps children, among them income, fathers' involvement, parenting quality and family stability, "typically find that direct positive associations remain between children's wellbeing and marriage, strongly suggesting that marriage is more than the sum of these particular parts". Its advantages for children, he argued, "are likely to be hard to replicate through policy interventions other than those that bolster marriage itself". https://futureofchildren.princeton.edu/sites/g/files/toruqf2411/files/media/marriage_and_child_wellbeing_revisited_25_2_full_journal.pdf A Swedish working paper found that weddings set off by a 1989 pension change made no difference to children's grades at 16, but it looked only at children whose parents were still living together in 1990, and its authors couldn't rule out that marriage matters "in countries where cohabitation is less prevalent than it is in Sweden" or for outcomes other than education. https://doi.org/10.2139/ssrn.1049041

So selection explains part of the gap but not all of it. The rest is the harm of break-up, which marriage does most to prevent. Marriage is a commitment device - legal, social, financial, normative - that changes how couples behave.

THE SAME PATTERN EVERYWHERE

The international evidence runs one way. Sara McLanahan and Gary Sandefur's Growing Up With a Single Parent, based on four American national surveys, found children whose parents live apart twice as likely to drop out of high school, twice as likely to become single parents themselves and one and a half times as likely to be idle in young adulthood. https://www.hup.harvard.edu/books/9780674364080

The Fragile Families study, which followed nearly 5,000 births in large American cities, about 3,600 of them to unmarried parents, found over 60 per cent of the unmarried couples' unions dissolving within five years of the child's birth. https://futureofchildren.princeton.edu/sites/g/files/toruqf2411/files/media/fragile_families_20_02_fulljournal.pdf In a 2016 analysis for the Institute for Family Studies, Jaap Dronkers found that across Europe "unmarried cohabiting relationships were three to five times more likely to dissolve than both types of marriages", with or without living together first. In no country was the cohabiting risk less than twice the married one. https://ifstudies.org/blog/cohabitation-marriage-and-union-instability-in-europe/

The 2017 World Family Map, from the Institute for Family Studies and the Social Trends Institute, drew on data from 100 countries and found marriage "associated with more family stability for children across much of the globe". https://ifstudies.org/ifs-admin/resources/reports/wfm-2017-fullreport.pdf In the United States and 16 European countries, children born to cohabiting couples were over twice as likely as those born to married couples to see a change in their parents' union by age 12. https://ifstudies.org/ifs-admin/resources/reports/wfm-2017-fullreport.pdf In the UK it was 60 per cent against 30 among children of mothers with moderate education. The UK was also the only European country where fewer than three in four children lived with both natural parents. https://ifstudies.org/ifs-admin/resources/reports/wfm-2017-fullreport.pdf Laurie DeRose's 2023 review of that evidence for an Oxford handbook of family policy found "few exceptions to the generalization that children born to marital unions have the best chance of stability across cultures, legal systems, welfare regimes, and levels of cohabitation". https://doi.org/10.1093/oxfordhb/9780197518151.013.21 The Institute for Fiscal Studies' Deaton Review of inequalities found in 2024 that children in the UK "are more likely to experience unstable family lives compared with their counterparts in western and northern European countries". https://doi.org/10.1093/ooec/odad031

COMPASSION, AND THE STRUCTURAL TRUTH

All of this must be read with compassion. Single parents - overwhelmingly mothers - often do extraordinary work in hard circumstances, and many of their children thrive. Some marriages are harmful, and where there's abuse, separation may be the best outcome for the children. These are averages and risks, not destinies, and they judge nobody.

But compassion includes telling the truth about structures. Successive governments' benefit rules that penalise couples, housing costs, economic insecurity and cultural drift have done more to shape the decline of marriage among lower-income families than any failure of personal virtue. Removing those barriers is social justice, not sentimentality.

WHAT WE BELIEVE

Marriage - the voluntary, exclusive, lifelong union of one man and one woman - gives children the most stable foundation there is, and every child deserves the best possible start. That's the conclusion the UK's most authoritative datasets, the peer-reviewed literature and the international evidence all point to, and it doesn't depend on explaining selection away. The stability gap holds in every income band, and much of it survives even when other differences between the couples are allowed for as fully as possible. https://marriagefoundation.org.uk/research/married-poor-more-stable-than-unmarried-rich/ https://marriagefoundation.org.uk/wp-content/uploads/2026/06/PhD-The-Timing-of-Marriage_digital-1.pdf

We believe marriage is a commitment device, not a certificate for couples already destined to last. The public, legal, lifelong promise changes the relationship it seals. It tells husband and wife, their families and their community that this is built to endure, and the behaviour follows the promise.

We believe the marriage gap between rich and poor is one of the quiet injustices of our age. The children of the affluent get married parents, and the children of the poor increasingly don't - not because poorer parents love their children less, but because the supports that make marriage reachable have been pulled away from those who need them most. So we call on the Government to act on the evidence: end the couples penalty in the benefits system, support marriage preparation, raise the Marriage Allowance to a meaningful level, and back a culture that holds up lifelong, faithful marriage between a man and a woman as the gold standard for family life. Marriage preparation earns that support, since a meta-analysis of premarital programmes found the average participant better off afterwards than 79 per cent of people who took none, in their relationship skills and the quality of their relationship. https://doi.org/10.1111/j.1741-3729.2003.00105.x

And a word of realism about divorce, since fear of it hangs over every discussion of marriage. Of the couples who married in 2015, just 16.3 per cent had divorced by their tenth anniversary. More than eight in ten, 83.7 per cent, had not, and 24.7 per cent of the couples who married in 1995 had divorced by the same point. https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/divorce/bulletins/divorcesinenglandandwales/2024and2025 The claim that half of all marriages fail is a myth. Research by Linda Waite and colleagues found that almost two thirds of unhappy American spouses who stuck with their marriage were happily married five years later, and among those who had rated their marriages very unhappy, 78 per cent were. https://www.healthymarriageinfo.org/wp-content/uploads/2018/05/UnhappyMarriages.pdf That finding describes ordinary unhappy patches, not abuse, where safety comes first.

More babies are again being born within marriage or civil partnership. In 2025, 53.2 per cent of births in England and Wales were within marriage or civil partnership, up three years running from 48.6 per cent in 2022, on our calculation from the ONS birth registration tables, at https://www.ons.gov.uk/file?uri=/peoplepopulationandcommunity/birthsdeathsandmarriages/livebirths/datasets/birthsinenglandandwalesbirthregistrations/2025/2025birthregistrations.xlsx.

C4M runs marriage courses for churches and other organisations across the United Kingdom, free of charge - Astonish the World for married couples, and Building on Rock for couples preparing to marry. Details are at https://www.c4m.org.uk/marriage-support/marriage-courses.

Please send this page to anyone who thinks it makes no difference to a child whether their parents marry.

Join us. Receive the Coalition for Marriage briefings at https://www.c4m.org.uk/subscribe, and support the campaign at https://www.c4m.org.uk/donate.

RELATED PAGES ON C4M.ORG.UK

Why Marriage Matters - https://www.c4m.org.uk/marriage-support/why-marriage-matters Marriage Facts and Figures - https://www.c4m.org.uk/marriage-support/marriage-facts-and-figures Marriage courses - https://www.c4m.org.uk/marriage-support/marriage-courses

KEY SOURCES

Every link on this page was accessed on 8 October 2026.

Centre for Longitudinal Studies, UCL, Millennium Cohort Study. https://cls.ucl.ac.uk/cls-studies/millennium-cohort-study/ Marriage Foundation, Married Poor More Stable than Unmarried Rich (2022). https://marriagefoundation.org.uk/research/married-poor-more-stable-than-unmarried-rich/ Marriage Foundation, Source of Family Breakdown (2023). https://marriagefoundation.org.uk/research/source-of-family-breakdown/ Marriage Foundation, The Timing of Marriage and Union Dissolution (2026). https://marriagefoundation.org.uk/research/the-timing-of-marriage-and-union-dissolution/ Institute for Fiscal Studies, Briefing Note BN107 (2010). https://ifs.org.uk/sites/default/files/output_url_files/bn107.pdf Institute for Fiscal Studies, Commentary C114 (2010). https://ifs.org.uk/sites/default/files/output_url_files/comm114.pdf Institute for Fiscal Studies, Report R87 (2013). https://ifs.org.uk/sites/default/files/output_url_files/r87.pdf Kiernan, Crossman and Phimister, Families and inequalities, IFS Deaton Review (2024). https://doi.org/10.1093/ooec/odad031 NHS Digital, Mental Health of Children and Young People in England 2017, Predictors of mental disorders (2019). https://files.digital.nhs.uk/85/7FFC57/MHCYP%202017%20Predictors.pdf Fourth National Incidence Study of Child Abuse and Neglect, Report to Congress (2010). https://www.acf.hhs.gov/sites/default/files/documents/opre/nis4_report_congress_full_pdf_jan2010.pdf Horn and colleagues, Journal of Family Psychology 27(1) (2013). https://pmc.ncbi.nlm.nih.gov/articles/PMC3645280/ Johnston, Jones and Pope, NBER Working Paper 33776 (2025). https://www.nber.org/system/files/working_papers/w33776/w33776.pdf McLanahan, Tach and Schneider, The Causal Effects of Father Absence, Annual Review of Sociology (2013). https://pmc.ncbi.nlm.nih.gov/articles/PMC3904543/ Brown, Marriage and Child Well-Being, Journal of Marriage and Family (2010). https://pmc.ncbi.nlm.nih.gov/articles/PMC3091824/ The Future of Children, Marriage and Child Wellbeing Revisited, 25(2) (2015). https://futureofchildren.princeton.edu/sites/g/files/toruqf2411/files/media/marriage_and_child_wellbeing_revisited_25_2_full_journal.pdf McLanahan and Sandefur, Growing Up With a Single Parent, Harvard University Press (1994). https://www.hup.harvard.edu/books/9780674364080