America's sandwich generation is a family map, not one caregiving story

More than half of Americans in their 40s now fit a familiar family label. Pew Research Center estimates that 54% are in the “sandwich generation”, positioned between a parent aged 65 or older and children of their own.

The number is striking. The definition behind it is more useful.

Pew counts an adult as sandwiched if they have a living parent aged 65 or older and either a child under 18, an adult child they supported financially during the previous 12 months, or both. It is a map of family relationships and money moving between generations. It is not, by itself, a measure of co-residence, hours of unpaid work, hands-on care or distress.

That distinction does not make the family pressure unreal. It stops one broad label from turning several different lives into the same stock story.

The new analysis draws on a nationally representative Pew survey of 8,750 U.S. adults conducted from 2 to 8 September 2025. Across all adults, 25% met the sandwich definition, up modestly from 23% when Pew last measured it in 2021.

Age changes the picture sharply. The share was 8% among adults under 30, 25% among those in their 30s, 54% among those in their 40s, 45% among those in their 50s and 10% among people aged 60 or older.

Those figures describe the overlap of generations, not the intensity of support. The underlying questionnaire asks whether someone has a living parent in specified age bands, whether they are a parent or guardian of a child under 18, and whether they gave any financial support to an adult child in the past year. It does not set a minimum amount for that financial help in the published question.

A person who paid one large bill for a 23-year-old and a person covering rent every month can therefore enter the same structural category. So can a parent of a teenager whose 68-year-old mother lives independently in another state. Neither case proves that three generations share a roof or that one adult performs daily care.

Pew's breakdown shows why a single image of harried midlife is too narrow.

Among sandwiched adults in their 30s, 90% had a minor child and had not financially supported an adult child. Among those in their 40s, 58% were in that minor-child-only group, 19% had supported an adult child but had no minor child, and 23% had done both while also having an older parent.

By the 50s, the pattern had flipped. Among sandwiched adults in that decade, 61% had financially helped an adult child but did not have a minor child, while 18% had a minor child only and 21% had both.

The label therefore follows a sequence. School runs and an older parent may define one period. Later, the child-side connection may be a transfer, groceries or temporary help with housing. An older parent's presence in the definition may mean an occasional check-in, regular errands or substantial personal care. The survey category does not decide which.

This also helps explain the change over time. The overall sandwich share moved only two percentage points between 2021 and 2025. The more notable shift was among adults in their 50s, where it rose from 36% to 45%. Pew says part of that increase reflects more people in their 50s having a living parent aged 65 or older.

A separate Pew report from the same survey wave asked people whether they considered themselves caregivers for a parent, spouse or partner aged 65 or older. Ten per cent of all U.S. adults said they were a caregiver for an older parent. Among adults who had a parent aged 65 or older, 24% described themselves that way.

That is a different question from the sandwich definition, and it produces a different denominator. Keeping the two measures separate is essential. “Has an older parent” should not be silently rewritten as “provides care to an older parent”.

For people who did identify as caregivers, the work was concrete. Among those caring for an older parent, 52% said they regularly helped with errands, housework or home repairs. Some 42% regularly helped manage health care, 39% helped manage finances and 16% provided personal care. In total, 68% regularly did at least one of the tasks Pew listed.

The reported effects were not uniformly bleak or uniformly rewarding. A majority of those regularly helping said the work had a positive effect on their relationship with their parent. At the same time, negative effects outweighed positive ones for emotional well-being, physical health, finances, work and social life. Women were more likely than men to report negative emotional and physical effects, though the survey cannot explain every reason for that gap.

The old cultural picture of the sandwich generation often starts with overload: one person, two dependants and no spare hour. That life is real for some households, but the current data does not assign it to everyone in the category.

Pew's 2021 survey found sandwiched adults were slightly more likely than other adults to say they were very satisfied with family life. They did not differ on reported satisfaction with their social life or personal finances. Those older findings are descriptive, not evidence that being sandwiched improves life. They do show why family connection and family pressure can exist at the same time.

Precision matters when families, employers and policymakers respond. A school-age parent with an independent 66-year-old father may need predictable hours. Someone managing an 82-year-old mother's appointments may need flexibility and respite. A parent helping an adult child with rent may need a financial conversation rather than a care service. The label points to the junction, but the useful question is what actually travels through it.

In the same 2025 survey, majorities of U.S. adults favoured several federal measures aimed specifically at people providing care for an older family member, including tax credits, short-term respite care, paid family leave and direct payments. Those results measure public preferences, not enacted benefits. They also concern caregivers, a narrower group than everyone who fits the sandwich definition.

The new 54% figure is best read as a demographic signal. Midlife family life increasingly spans more than one active generation, and the mix changes as children and parents age. To understand the real squeeze, ask the next questions: who needs help, what kind, how often, from whom, and whether the people involved live together. That is where a family map becomes an account of daily life.