how americans live together

how americans live together, read off public data, is mostly a story about cost. renting is expensive enough that a large share of households are stretched by it, rents are sharply higher than they were before the pandemic, and sharing a place, with roommates, with family, by "doubling up", is a substantial part of how people manage. that is the defensible reading of the census and harvard data. the part people over-read is the precise size of the roommate trend, because the most-quoted number is not actually a roommate number. this piece walks the figures that hold up, and labels the one that does not.

the housing math explains why so many people end up sharing. it says nothing about whether the sharing goes well — and it does not pick the person.

how americans live together, in one line

how stretched are renters right now

very, and the size of it is the cleanest number here. according to harvard's joint center for housing studies, in its state of the nation's housing 2025 report, about 22.6 million renter households, roughly half of all renters, were cost-burdened in 2023, meaning housing and utilities ate 30% or more of household income. of those, 12.1 million were severely burdened, spending more than half their income on housing. the national low income housing coalition's coverage frames the same report as a record number of cost-burdened renters for a third straight year. that is the backdrop everything else sits on: for about half of renters, the rent is already the thing that does not fit.

worth separating the two tiers, because they describe different intensities of the same squeeze. the 22.6 million is the whole cost-burdened group; the 12.1 million severely burdened sit inside it, which leaves about 10.5 million moderately burdened (30–50% of income on housing) as the difference between the two published figures. the chart below splits the total that way so the severe slice is visible rather than buried in the headline.

why does the rent feel so much higher than it used to

because it is, and the size of the jump is the second clean number here. median asking rents were about $1,830 in early 2025, roughly 32% above their 2019 level, per harvard's joint center for housing studies. the headline figure is that +32%; the 2019 dollar amount is not separately published in the same line, so we get it by working the percentage backward, $1,830 divided by 1.32 is about $1,386, and we label that 2019 bar as implied rather than reported. either way the point survives the rounding: in about half a decade the typical asking rent rose by roughly a third, faster than most incomes did, which is the mechanism turning "i could live alone" into "i need a roommate" for a lot of people. the rent chart and the cost-burden chart are the same story from two angles: prices climbed, and the share of renters who cannot comfortably absorb them is now about half.

is sharing a place actually common

yes, and "doubling up" is the broader way the data describes it. sharing and doubling up, more than one adult who is not a spouse or partner of the householder living under one roof, are a substantial part of how americans house themselves, in census and nlihc analyses. that includes roommates, but it also includes adult children living with parents, siblings, and other arrangements that have nothing to do with a roommate match. nlihc treats doubling up as an affordability indicator precisely because some of it is involuntary, people sharing because living alone is not affordable, not because they wanted a housemate. so the honest version of the claim is broad: shared living is a major, ordinary part of the housing picture. it is not the same as saying "roommate arrangements specifically are surging by some exact rate."

what about the 38.4% number

that figure is real, and it is the one most likely to be misused, so here is the careful version. the census reported that 38.4% of adults ages 25 to 34 lived in a shared household in 2019. but a "shared household" in that data includes adults living in a parent's home, in fact 17.8% of that age group lived in their parents' household that year, which is most of what moves the shared-household share. so 38.4% counts roommates and adult kids back home and other non-householder adults all together. it is not a roommate statistic, and using it as roommate-growth evidence overstates a roommate trend by folding in people who live with their parents. we keep it here only labeled for what it is, a shared-household share, because that is the only honest way to use it. the roommate-growth thesis leans on the rent-burden and doubling-up data above, not on this one number.

do older adults share too

yes, and it is worth saying because the cultural image of a roommate is a twenty-two-year-old. bowling green state university's national center for family and marriage research tracks older adults living with an adult roommate or sibling, and the count has grown over recent decades in its public family profiles. that is a demographic trend in public data: a meaningful and growing share of older adults live with housemates. it is not a claim about why any particular older adult shares, and it is not a matching statistic, it is a count of how people are living. but it widens the picture in a useful way: shared living is not a phase young renters pass through, it spans ages, which is consistent with the cost story rather than a generational quirk.

what this means if you are about to share

the data does not tell you whether to get a roommate, your budget and your lease mostly already did that. what it tells you is that a lot of people are sharing for the same reason, cost, and that sharing is the norm rather than the exception. it says nothing, on its own, about whether sharing will go well, and it specifically does not say that sharing protects you from the stress of an expensive rent; that is a claim the evidence does not support, and we will not make it. the gap between "sharing is common" and "sharing went well" is where the rest of this hub lives: what predicts a good roommate is aligned habits and communication, not personality; how roommates actually fight is about reading intent, not winning arguments; and how to split rent fairly is the logistics that decide whether a shared place stays shared. the housing math explains the share. it does not pick the person.