The Family Formation Reset

In 1960, the median American woman married at 20, had her first child at 21, and was living independently of her parents by her early twenties. In 2024, the median first marriage occurs at age 28.6 for women and 30.2 for men (CDC NCHS). The median age at first birth has risen to 27.5. Eighteen percent of Americans aged 25–34 now live with a parent — a level last seen in the 1940s before post-war prosperity launched the household-formation boom that defined the late twentieth century. The U.S. total fertility rate sat at 1.62 children per woman in 2023, the lowest level ever recorded and well below the 2.1 replacement threshold (CDC).

The temptation, when looking at this data, is to read it as a story of family decline. The more accurate framing is that the sequence and structure of family formation has been reset. Americans are not abandoning partnership, parenthood, or household formation — they are deferring those milestones, often by a decade, and they are unbundling them from one another. Marriage is no longer a precondition for cohabitation, cohabitation is no longer a precondition for childbearing, childbearing is no longer a precondition for adult identity, and moving out of the parental home is no longer a marker of having achieved adulthood. Each of these traditional packages has been pulled apart, and the consequences for housing, labor markets, consumption, civic participation, and intergenerational wealth flows are still working their way through the system.

The Long Deferral

The single most important fact about American family formation is that nearly every milestone has shifted later. The median age at first marriage rose from 20 (women) and 23 (men) in 1960 to 28.6 and 30.2 in 2024. The median age at first birth rose from 21.4 in 1970 to 27.5 in 2023. The share of women aged 30–34 who have never been married rose from 6% in 1970 to 33% in 2024 (Census ASEC). The share of births to women aged 35 and over rose from 5% in 1970 to 19% in 2023 (CDC).

These deferrals are not uniform across the population. They are sharpest among college-educated women, where median first birth now occurs at 31.3, and least pronounced among women without a bachelor's degree, where it remains near 24. The bifurcation has produced what demographers call the "diverging destinies" pattern: a college-educated upper-middle class that delays family formation, partners with similarly educated spouses, has fewer children at older ages with more economic and time investment per child, and remains more stably married; and a non-college majority that experiences earlier childbearing, less stable partnerships, higher rates of single parenthood, and substantially less accumulated household wealth at every stage of the life course.

The deferral is real but the endpoint is not zero. Pew's 2024 analysis finds that among Americans aged 40–44, 84% have had at least one biological child — down only modestly from 89% a generation earlier. The share of women who reach the end of their reproductive years childless has risen, but most of the change in fertility statistics reflects later and lower-parity childbearing rather than mass childlessness. The story is compression and deferral, not abandonment.

The Fertility Floor

U.S. total fertility has fallen below 2.1 every year since 2007 and has continued to drift downward. The 2023 reading of 1.62 is the lowest in the historical series and is consistent with the pattern across nearly every wealthy democracy: South Korea (0.72), Italy (1.20), Japan (1.20), Spain (1.16), Germany (1.36), the United Kingdom (1.44), Canada (1.33). The U.S. is no longer the outlier in the developed world; it is now firmly in the middle of the low-fertility cluster, though above the very low fertility seen in East Asia and southern Europe.

The most important empirical work on the U.S. decline is by Melissa Kearney and Phillip Levine (Brookings, 2023), which examined whether the fall could be attributed to standard explanations — student debt, housing costs, child care affordability, women's labor force participation, contraception access, or recession effects. They find that none of these factors, individually or in combination, can account for more than a fraction of the decline. The post-2007 fertility drop appears to reflect "shifting priorities across cohorts" — a durable change in how younger generations weight childbearing against other life options, not a temporary response to economic conditions.

Surveys of intended fertility add a critical detail: stated ideal family size in the United States has remained stable at roughly 2.5 children per woman since the early 1990s. The gap between intended and actual fertility — now larger than at any point in the post-war era — suggests that many Americans who want children are not having them, or are having fewer than they wanted. The mechanisms behind that gap include partnering difficulties (the marriage market itself has thinned, especially for non-college men), the rising age at first attempt (which compresses the window for having additional children), and the opportunity costs of childbearing in dual-career households where both partners' careers depend on continuous engagement.

Marriage as a Capstone, Not a Cornerstone

The cultural meaning of marriage has shifted from a foundational institution that launched adult life to a "capstone" achievement that follows financial stability, residential settling, and often parenthood itself. The Institute for Family Studies and the National Center for Family & Marriage Research have documented this shift in detail: the modal sequence today is education → independent residence → cohabitation → economic stabilization → marriage, and increasingly, → first child after marriage. This is the inverse of the 1960 sequence in which marriage typically came first and almost everything else followed.

Cohabitation has absorbed much of the function that early marriage once served. Roughly 70% of first marriages now begin with a period of cohabitation, up from 11% in 1965. Cohabiting couples now account for 17% of all U.S. parents, up from under 2% in 1968 (Census). For the under-35 cohort, cohabiting parents outnumber single parents and approach the share of married parents. The institutional architecture of American family law — divorce, custody, inheritance, health insurance, immigration — has only partially adapted to a world in which long-term partnership and childbearing are routinely disconnected from marriage.

Marriage itself has bifurcated. Among college-educated Americans, marriage rates have stabilized and divorce rates have fallen substantially since the 1980s. Among Americans without a bachelor's degree, marriage rates have continued to decline and rates of relationship instability remain high. The college- educated marriage is now more stable, more egalitarian, and more likely to involve children raised in two-parent households than at any point since reliable measurement began. The non-college marriage market has thinned dramatically, and the share of non-college adults who reach age 40 having never married has more than doubled since 1980.

The Household Recomposition

Underneath the marriage and fertility shifts, the basic structure of the American household has changed. Single-person households now account for 29% of all U.S. households, up from 13% in 1960. The share of households consisting of a married couple with their own children fell from 40% in 1970 to 18% in 2024 (Census). The share of households with three or more generations under one roof — the multigenerational household — rose from 7% of the population in 1971 to 18% in 2021, with the absolute number more than tripling to 59 million Americans (Pew).

The rise in multigenerational living is the most underappreciated structural change in the household data. It is occurring across nearly every demographic group but is concentrated among Asian- American (29% of population in multigen households), Hispanic (26%), and Black (24%) families, compared to 13% among white Americans. The drivers are economic (housing affordability and the cost of elder care), cultural (immigrant family norms that persist into second and third generations), and demographic (the aging of the Boomer generation creating elder-care needs that many adult children absorb in shared households).

Independently, the share of young adults aged 25–34 living with a parent reached 18% in 2024 — the highest level since the 1940 Census, before the post-war housing and family-formation boom. The driver is primarily affordability rather than dependence: most boomerang adult children are employed, contribute financially, and are saving for eventual independent housing that is no longer affordable on a single income at entry-level wages in major metro areas. The pattern is increasingly normal, not exceptional, and the cultural stigma that once attached to adult children living at home has substantially diminished among the cohorts now experiencing it.

The Economic Drivers

Even though the Kearney-Levine work concludes that no single economic factor accounts for the fertility decline, the affordability environment for family formation has unambiguously deteriorated. The Joint Center for Housing Studies finds that the share of renters paying more than 30% of income on housing reached a record 50% in 2023; among households earning under $30,000, the cost-burdened share is 83%. The median home price- to-income ratio rose from 3.0 in 1990 to 5.6 in 2024, with much of the gap occurring among entry-level homes that historically served young families.

Child care costs have risen even faster. Child Care Aware's 2024 report finds the average annual cost of infant care in a licensed center is now $13,128 nationally, exceeding in-state public university tuition in 36 states. For families with two children in care, costs routinely exceed $25,000 per year — equivalent to a second mortgage payment. The combination of housing, child care, and student debt service has substantially reduced the disposable income of households in the prime family-formation years, even as headline employment and income figures have improved.

The opportunity cost of childbearing for women is also materially higher than a generation ago. Female labor force participation in the prime working ages stands at 77%, with college-educated women approaching 85%. The earnings trajectory of educated women has converged substantially with men's; the wage penalty for taking time out of the labor force, especially in professional careers with seniority systems, is larger than in occupations where women historically concentrated. For families optimizing total household income across both partners, the financial logic of postponing or limiting childbearing is more compelling than at any prior point.

The Cohort Effect

Perhaps the most important finding for forecasting is that successive cohorts now look meaningfully different from their predecessors at the same age. Millennials at age 30 had lower marriage rates, lower homeownership rates, and lower fertility than Gen X had at 30, and Gen X had each of these at lower levels than Boomers. The pattern continues with Gen Z, where survey data on intentions, partnership rates, and timing expectations all suggest further deferral — though Gen Z's full life-course outcomes remain to be observed.

The cohort effect matters because it implies that catch-up is only partial. Women who postpone first birth from age 25 to age 35 do not, on average, end with the same number of children they would have had at the earlier start. The biological window narrows; the partnering window narrows; the energy and economic bandwidth for additional children narrow. The deferred family-formation pattern compounds into permanently lower completed fertility, smaller completed families, and a materially different distribution of household types in the adult population than the one that produced the post-war American family system.

Implications for Housing, Labor, and Consumption

The downstream consequences of the family-formation reset are substantial and largely under-anticipated by institutions organized around the older household model.

Housing demand is shifting. The starter-home market — small detached houses in close-in suburbs, historically purchased by young married couples expecting one or two children — has weakened structurally. Demand has shifted toward urban rentals (single adults and cohabiting couples), large multigenerational homes (extended families pooling resources), and small two-bedroom homes (single parents, child-free couples, late-life downsizers). The mismatch between the existing housing stock — built largely between 1950 and 1990 for the post-war family — and the household composition that will dominate the next thirty years is one of the largest unaddressed structural challenges in the housing market.

Labor market participation patterns are changing. The traditional career arc — full-time continuous employment from early twenties to retirement, with women reducing hours during early child-rearing years — is fracturing. The deferral of child-rearing pushes the high-intensity caregiving years into the late thirties, when career investments are at peak return. The result is increased demand for parental leave, flexible scheduling, remote work, and re-entry programs after caregiving breaks; employer practices designed around the older family sequence are increasingly mismatched with the actual life courses of their workforce.

Consumption patterns are shifting. Categories historically driven by young-family formation — first homes, family vehicles, child-related goods, life insurance, suburban retail — face structurally lower demand from the under-35 cohort than they did a generation ago. Categories that benefit from deferred or non-traditional family structures — pet ownership (now at record highs), travel, premium experiences, urban services, and adult-oriented housing — are absorbing the spending that earlier cohorts directed toward family formation. The pet industry alone has grown from $50 billion in 2010 to over $150 billion in 2024, partly absorbing the caregiving and nurturing expenditure that earlier generations directed toward children.

Public-finance pressures will intensify. A society with sustained sub-replacement fertility, rising life expectancy, and a growing dependency ratio at older ages faces structural pressure on Social Security, Medicare, and the entire pay-as- you-go structure of intergenerational transfers. The U.S. demographic position is more favorable than Japan's or Italy's — partly because of immigration, partly because U.S. fertility sits closer to 1.6 than to 1.2 — but the trajectory matters, and policy responses (immigration levels, retirement age, family policy, child allowance design) will become increasingly consequential over the next two decades.

The Pronatalist Politics

The political response to the family-formation reset has begun to organize itself across the ideological spectrum. The right- leaning version emphasizes traditional family structure, child tax credits, religious community, marriage promotion, and critiques of the cultural conditions (career intensity, secularism, screen-mediated socialization) seen as deferring family formation. The left-leaning version emphasizes paid family leave, universal child care, housing affordability, student-debt relief, and the structural economic conditions that constrain choices. Both diagnoses contain real elements, and both political traditions have begun to converge on policy questions (child allowances, paid leave, early-childhood supports) that earlier generations of policymakers viewed as the province of the other side. (See "The Pronatalism Wars" report for the full politics of this debate.)

The empirical evidence on pronatalist policy is mixed. Generous child allowances and paid leave appear to raise completed fertility modestly — the strongest evidence comes from Hungary, France, and the Nordic countries — but rarely bring it back to replacement. The deeper drivers of low fertility appear to be cultural and structural in ways that are not easily reversed by transfer payments alone. The family-formation reset is therefore likely to be a durable condition that policy can attenuate at the margins but cannot fully reverse.

What the Reset Means

The family-formation reset is not a temporary deviation from a twentieth-century norm. It is the new structure of American household life, and it will define the demographic, economic, and cultural environment of the next several decades. The country has not become anti-family — most Americans still partner, most still want children, and the meaning attached to family relationships in survey research remains as high as it has ever been. What has changed is the timing, the sequence, and the institutional packaging of family life.

The institutions that will adapt successfully — employers, housing developers, retailers, schools, churches, civic organizations, governments — are those that recognize that the modal household of 2030 will not look like the modal household of 1980 and will design their products, policies, and assumptions accordingly. The institutions that continue to operate on the older template will find their relevance narrowing year by year as the families they were built to serve become a smaller and smaller share of the population they need to reach.

The reset is most fundamentally an unbundling. What was once a tightly packaged sequence — leave home, marry, have children, own a home, build a career — has become a loose set of options that Americans now combine in many different orders, skip selectively, and execute on widely varying timelines. The diversity of household forms in the United States in 2025 is greater than at any point in the country's history, and the institutions that serve American life are still, in most cases, organized around a household structure that no longer describes most of their customers, members, employees, or constituents.