America Could Have Too Many Homes by 2035—Just Not Where Buyers Need Them

As early as 2035, housing supply may outpace demand, according to a new analysis—potentially cooling prices in fast-building markets even as shortages persist elsewhere.

The warning comes from the Mortgage Bankers Association’s “Implications of a Persistent Slowing in Housing Demand” report, which estimates the nation could add between 10.6 million and 14.6 million net housing units in the next decade—just as household growth slows, driven by an aging population, lower fertility, and falling immigration projections.

It’s an unexpected, if not ironic, twist in a housing market that has spent the better part of two decades facing a structural deficit.

Since the Great Recession, new construction has lagged woefully behind household formation, leaving the country short an estimated 4.03 million homes.

But the MBA report argues that the demand conditions that fueled that shortage are coming to an end. In markets that built aggressively during the COVID-19 pandemic-era boom—especially across the South and West—new homes and apartments could keep arriving as the number of households able or willing to absorb them falters.

“That simple arithmetic has profound implications for how we think about housing supply adequacy—and calls into question whether the supply shortage that defined the post-2010 housing narrative will remain the right framework for the decade ahead,” the report suggests.

But that warning comes with a major asterisk, according to Joel Berner, senior economist at Realtor.com®. 

Housing demand is far from fixed, he says. Builders can cut production when sales slow, households can form when housing gets cheaper, and a home built in a fast-growing Sun Belt market does little to relieve scarcity in a high-cost metro hundreds of miles away.

The ‘arithmetic’ of housing supply

At the core of the report are two competing projections: how many additional homes the country will need over the next decade, and how many homes builders could deliver.

MBA estimates that demand for additional housing will total about 11.34 million units between 2025 and 2035. That tally includes projected household growth, homes needed to replace units removed from the housing stock, additional vacancies needed in a growing market, and second-home demand.

On the supply side, the group models three construction paths based on historical completion levels. Under its low scenario, net housing supply would grow by about 10.57 million homes from 2026 through 2035. Under its middle scenario, supply would grow by 12.65 million homes; under its high scenario, it would rise by 14.56 million.

That leaves projected demand sitting between the report’s low- and middle-supply cases—a gap MBA says could allow supply growth to exceed demand by 2035 if construction remains elevated.

“These are reasonable assumptions to make, but they don't give much opportunity for builders to adjust based on the demand that they're facing,” says Berner. 

But he cautions that the projections give builders limited room to respond to a changing market.

“If demand for homes really fell off and ended up short of supply, builders wouldn't keep producing at the same rate,” he explains.

There are already signs that builders are becoming more cautious. Just as the inventory of newly built homes for sale reached 10.3 months—solidly buyer’s market territory—new starts dropped.

Total housing starts fell 15.4% in May from April, to a seasonally adjusted annual rate of 1.17 million units, driven largely by a sharp decline in multifamily construction. Single-family starts slipped 1.9% from April and 6.7% from a year earlier, to an annualized pace of 882,000.

Builders can slow down, but household formation can speed up

There’s further evidence that household formation may be just as dynamic.

Central to MBA’s warning is the projection that the U.S. will add 8.6 million households between 2025 and 2035, down from 10.1 million in the prior decade.

But Berner suggests that household formation is at least in part shaped by whether people can afford to move out, rent a place of their own, buy a first home, get married, have children, or stop sharing housing with relatives and roommates.

“Part of how we quantify the supply gap nationally is to count households that would have formed if affordability conditions were better,” he says. “This piece just takes falling household formation as a given, while we see it as a variable result of conditions in the marketplace.”

There’s already evidence that a large share of potential household demand remains delayed rather than absent. A record 25.2 million adults under 35 lived with their parents in 2025, according to Realtor.com research—nearly 1 in 3 young adults. 

On top of that, 7 in 10 adults aged 25 to 34 who lived at home were employed, suggesting that delayed independence cannot be explained by joblessness alone.

That’s not to suggest that there isn’t a borderline demographic slowdown.

Fertility fell to an all-time low of 1.64 births per woman in 2024, well below the roughly 2.1 births per woman needed for population replacement absent immigration, according to the latest Centers for Disease Control and Prevention data.

But high housing costs may also be playing a role in these trends.

One recent study found that rising housing costs accounted for 51% of the overall decline in U.S. fertility between the 2000s and 2010s. Had housing costs remained flat after 1990, the researchers estimated, the country would have had 13 million more births by 2020 and a fertility rate in the 2010s 77% closer to replacement level.

That’s why Berner argues that more housing can create some of the demand MBA’s model treats as missing. 

“We argue that building more will create more demand by creating the conditions—lower home prices—for new households to form,” he says.

The risk: Building in places that can build—not the places that need housing most

But any real estate agent will tell you: National trends can obscure the question that matters most—what is happening in the local market?

Builders tend to add homes where land is available, permitting moves faster, and development costs are lower. But those aren’t always the metros where housing costs have climbed furthest beyond what local incomes can support.

“For home prices to fall significantly and uniformly, there would have to be delivery of new housing units to the markets that need them most,” Berner says. “In hot-market pockets of the Northeast and Midwest, prices will continue to grow because inventory remains scarce.”

MBA’s report draws a similar divide. 

It expects price growth to slow—and in some cases decline—in states where single-family construction has expanded significantly, including Arizona, Texas, and Florida. In New York, New Jersey, and Massachusetts, however, it expects relatively stronger price growth to persist as restrictive zoning, high permitting costs, and elevated labor expenses continue to limit new construction.

That pattern is reflected in the most recent Affordability and Homebuilding Report Cards from Realtor.com. 

New York, New Jersey, and Massachusetts were among the worst-performing states, held back by high prices, constrained land, restrictive zoning, and construction costs that have moved far beyond what many middle-income buyers can afford. Arizona, Florida, and Texas, meanwhile, all ranked in the top 20.

That split is why a national supply surplus, even if it materializes, would not automatically mean broad affordability relief.

A renter in Austin, TX, may see more apartments, more concessions, or slower rent growth as new units come online. A buyer in Boston, meanwhile, may still find that the homes they can afford remain scarce, with little new competition arriving to pull prices down.

“Builders will have to overcome the cost and regulatory burdens of building in these markets to get anywhere near a scenario where supply exceeds demand,” Berner says. “This scarcity will keep prices rising in parts of the country that are not adequately supplied.”