Atlantic Cities
Charts

America's Housing Stock Mismatch

America's Housing Stock Mismatch
REUTERS/Kevin Lamarque

When the census first started measuring the size of American households, back in 1790, the average home had nearly six people in it. By 1960, that number had fallen to just over three. According to early 2010 estimates—although the data also shows many of us doubling up to make rent in the recession—we’re now down to about 2.6.

Behind much of this downsizing in the last few decades has been the rise of the single-person household. The phenomenon, which reaches ominous proportions when projected decades into the future, has been celebrated in trend stories, studied by social scientists, and parsed by family-values advocates who see in so many solitary people the latest sign of society unraveling.

But there’s one group that still hasn't caught on: homebuilders. American households have been shrinking for years at exactly the same time as our houses have been expanding. By a lot. Data from the Center for Neighborhood Technology compares trends in household size to census data on home construction, showing that the average size of new homes swelled from 1,400 square feet in 1960 to 2,100 four decades later.

Arthur C. Nelson, who directs the Metropolitan Research Center at the University of Utah, predicts that the average U.S. household won’t continue to shrink in the next 20 years. Gen X and Gen Yers are now pairing up and having children, and soon many will have baby-boom parents living with them, too. As a result, we’ll be at 2.5-3 people per household for the next generation, he says.

Still, trends over the last several decades have left us with a dramatic mismatch in the coming years between demographics and housing stock.

“And it’s not pretty,” Nelson warns.

The city of the future will need to look much different than it does today if it’s going to house people where they want to live, how they want to live, and with—or without—all the other people who will need to pile in, too.

Nelson’s own research
supports what a number of industry surveys have shown. About 38 percent of Americans want to buy or rent an attached unit, meaning a townhouse, a condo or an apartment. Thirty-seven percent of people say they want to live on a small lot (either in a small house, or a large one with scant yard space). And 25 percent of people want to live on a large lot of more than 7,000 square feet.

“The surveys are telling us something,” Nelson says. “They’re all consistent about that distribution of demand. But what does the supply look like?”

It doesn't match up. Currently, only a quarter of Americans say they want to live on large lots, but 43 percent of them do. Only 28 percent actually live in attached units while a plurality of people want to, and only 29 percent of people live on small lots while millions more would like the chance. America currently has about 30 million more homes on large lots than the market demands. “That’s where the mismatch is,” Nelson says.

He suggests that the McMansion was the right housing stock for a very specific moment in time. Such large houses on large lots became popular in the mid-1990s, just as mature baby-boom families were at the height of their income and household size, before retirement and when all the children were still home. That moment, from the standpoint of demographics, has passed. And other trends, among them rising gas prices and rising minority populations, are now pushing in the opposite direction.

We can still use the McMansions, though, Nelson says. He proposes converting them into the next-generation affordable housing for multi-generation or multi-family households.

After all, McMansions typically have more bathrooms than bedrooms, allowing for residential space that could be easily divided into private units, with a common kitchen and living room. Many are already outfitted to install a second kitchen, or to convert a ground-floor bedroom. There’s easily parking for five or six cars. And there would still be space enough to go around, Nelson suggests. In 1950, the average 3.5-person home was smaller than 1,000 square feet. Someone living in a 6,000 square-foot McMansion tomorrow with 12 extended family members or family friends would fare even better.

“When you add up the spaces and how they’re distributed, the typical McMansion can be occupied by three-to-five households with their own splendid privacy, their own large space,” Nelson says.

But what about all the people who do want to live alone? In the absence of building yet more McMansions, Nelson argues that the real housing boom in this country should come from people retrofitting their existing homes to include accessory dwelling units: garage apartments, basement suites or even back-yard studios.

Zoning regulations in most cities are hostile to this idea, but it’s this kind of in-my-backyard housing stock that will satisfy the coming singletons without expanding the urban footprint. And it’s a good way to increase density where a lot of single people say they want to live: near transit.

Karen Chapple, an associate professor at the University of California at Berkeley, recently completed a feasibility study of building secondary units within half a mile of five metro stations on the Bay Area Rapid Transit system. These neighborhoods could accommodate thousands of such units – in fact, Chapple built one in her own back yard to suggest what they might look like – but current zoning laws would allow only about one in five of them to be built.

Chapple’s work speaks to one of the other great imbalances in existing housing stock: the gap between people who want to live by transit, and those who do. About a third of American households say they want the ability to walk to rail transit, but currently less than 10 percent do. If we wanted to fulfill that demand, Nelson says, every new residential development between now and 2050 would have to be built within half a mile of transit (or transit would have to be built out to within half a mile of housing).

Equally challenging are trends in homeownership, which has fallen from 69 percent in the mid-1990s to 66 percent today. By 2020, it should be at about 62 percent. And so not only do we need to be building exclusively near transit – and in some cases in other peoples’ back yards – but every new residence between now and 2020 also needs to be built for a renter. “That’s the killer,” Nelson says.

And it’s not likely to happen. But in the future, it’s not just the homes that will look different. In a sense, the people will, too.

“The modern renter is not the renter you imaged from 10 years ago,” Nelson says. “The renter you imagined 10 years ago was a marginal-income, marginal-education couple, a small family that can’t afford to buy a home, singles just starting out, or the recently divorced when you’ve got to park yourself someplace.”

In the future, renters will look increasingly like homeowners, as McMansions look increasingly like attractive group housing, and as your back yard looks more and more like a place you might want to plant a rental unit. If all this sounds drastic, well, that may be the cost of realigning housing stock with where the population is headed.

Emily Badger is a former staff writer at The Atlantic Cities based in Washington, D.C. She now writes for The Washington Post. All posts »

Join the Discussion