Are Millennials Finally Ready to Binge on Housing?
(Bloomberg Opinion) -- According to one Census Bureau survey, 2020 was the first year on record (going back to 1947) in which the number of households in the U.S. declined. According to another Census Bureau survey, 2020 saw the second-biggest increase on record (in this case going back to 1955) in the number of U.S. households.
This disagreement can be chalked up partly to the Covid-19 pandemic, which sent response rates to government surveys plummeting over the spring and summer, and also to the timing of the two surveys — the first was conducted in March and April, the second over the course of the year. But it turns out that estimating how many households there are in the U.S. is always hard. “For such a basic and fundamental measure, the number of households in the U.S. is largely a mystery,” three researchers from Harvard University’s Joint Center for Housing Studies lamented in 2015.
It’s a really useful number to know, though, given that household formation drives demand for housing, and by extension demand for lots of other things. In general it follows demographics, but over the past decade-plus the progression of the giant millennial generation through its 20s and 30s — usually prime years for forming households and buying first homes — hasn’t been accompanied by the expected amounts of household formation or home buying.
The consumer behavior of millennials has been the phenomenon that launched a million takes, with early arguments that they had fundamentally different priorities and values from their elders eventually giving way to an acknowledgement that no, they were mainly just broke. As three Federal Reserve Board economists put it in 2018:
We found that many of the demographic attributes associated with millennials—such as higher rates of racial diversity, higher educational attainment, and lower rates of marriage—are consistent with secular trends in the population and are therefore not the aberrations of a single generation. We showed that millennials do have lower real incomes than members of earlier generations when they were at similar ages, and millennials also appear to have accumulated fewer assets.
The distributional financial accounts that the Fed started publishing on a quarterly basis in 2019 allow us to compare the household net worth of the millennials, defined here as those born from 1981 through 1997, with that of generation X (1965-1980).
This understates the per-person gap, given that as of 2019 there were 17% more millennials than generation Xers. Still, the millennials had been catching up in recent years as the impact of Great Recession waned, and while the gap had widened again since early 2019 that was mainly because the gen Xers of circa 15 years ago were enjoying the upside of a housing bubble that would soon have its downside (which was in turn followed by a lot more upside; at the end of last year gen X controlled an estimated 27% of U.S. household wealth).
That brings us back to household formation, and estimating its growth or lack thereof. The two different Census Bureau surveys mentioned above are both offshoots of the Current Population Survey, the monthly interrogation of about U.S. 60,000 households from which the unemployment rate, labor-force participation rate and the like are derived. The Annual Social and Economic Supplement to the CPS is a more-detailed survey of about 100,000 households usually conducted mostly in March, while the Housing Vacancy Survey is based on the monthly CPS but also includes the 10,000 or so housing units selected for the CPS each month that turn out to have nobody home.
The Census Bureau’s 3.5-million-household American Community Survey also delivers annual estimates of the number of households, but it has only been in existence since 2005 and isn’t conducted in decennial census years such as 2020. Then of course there’s the census itself, which in theory surveys everybody and thus delivers what should be the most reliable household numbers, although they’ll be coming out extra late because of the pandemic and aren’t entirely compatible with the higher-frequency estimates anyway.
Here’s household formation as measured by the two annual surveys with 2020 numbers and long records, averaged out over three years because the results can be pretty volatile year-to-year, as compared with the Census Bureau’s count of housing starts, also averaged over three years.
The three series tend to follow similar if far from identical courses, most recently with the big drop-off in household formation and housing starts during the housing bust and modest recovery in early 2010s. Over the past five years, though, the ASECS survey has been showing a tailing off of household formation while the Housing Vacancy survey and housing starts have been showing an acceleration. We probably shouldn’t take the 2020 numbers from either survey too seriously, but the divergence was apparent before then.
The upward trend does seems a bit more likely to be real, given that the other annual estimates of household formation, from the American Community Survey, also show a rising trend line through 2019. There’s also the unmistakable evidence of a real estate boom in 2020, with new-home sales rising 20% and existing-home sales 5.6%, both reaching levels not seen since 2006.
But while some of 2020’s home buyers were new-household-forming millennials, most weren’t. The average age of buyers rose to an all-time high of 55 in 2020, according to a National Association of Realtors survey, the 31% first-time buyers’ share was the lowest since 1987 and the 22% of first-time buyers moving directly from a family member’s home (forming a new household, that is) also represented a decline from previous years. According to a recent Apartment List analysis of CPS data, millennial home-ownership rates have risen sharply over the past five years — which you’d expect as more millennials age into their 30s — but continue to lag those of previous generations. At age 30, 42% of millennials own homes; for gen X that was 48% and for baby boomers 51%.
This could signify a lot of pent-up demand. “There would be some 5.7 million additional households today if Americans formed households at the same rate they did in 2006,” Zillow Group Inc. economist Jeff Tucker wrote in December, “a testament to widespread difficulties in securing affordable, accessible housing over the past decade-plus but also a potential indicator of enduring housing demand to come.” Then again, those “widespread difficulties in securing affordable, accessible housing” could persist and possibly worsen as tight supply and rising demand drive up prices. As the millennials-focused Business Insider pithily summed up in a headline this week, “Millennial homeownership is causing the US to run out of houses.”
Near the beginning of this column I wrote that “household formation drives demand for housing,” but it’s clear that over the past decade-plus the supply of affordable and appropriate housing has to a large extent driven household formation. Inadequate new construction in and near places that created lots of jobs, such as New York City and the San Francisco Bay Area, drove prices so high that it was hard for even well-paid young workers there to strike out on their own. Meanwhile, the houses that were being built around the country grew increasing large and out-of-range for first-time buyers, with the share of new single-family homes with four bedrooms or more rising from 18% in 1985 to an all-time high of 47% in 2015 even as average household size fell.
The pandemic did make all those extra bedrooms seem a bit more sensible, both as home offices for remote work and as refuges for adult children fleeing locked-down cities. It’s also possible that the grand experiment in working from home that Covid-19 unleashed will break through some of the housing logjam for millennials, both by reducing real estate prices in super-expensive cities and making it easier for young professionals to do big-city jobs while living in less-expensive locales.
There are even some extremely modest signs that this is happening: The percentage of 18-to-29-year-olds living with their parents has been rising since the 1960s, according to a Pew Research Center analysis of census and CPS data, and reached an all-time high (going back to 1900) of 52% over the spring and summer. But more recent CPS data-crunching by Harvard’s Joint Center for Housing Studies found that by October the percentage of 25-to-29-year-olds living with parents had actually fallen below 2019’s levels (the 18-to-24 living-at-home percentage remained elevated because so many colleges and universities were still operating remotely). Maybe some of them got a good deal on a Manhattan apartment!
The story of millennials and housing has so far mainly been one of disappointed expectations, though, so I wouldn’t get too excited just yet. One big issue is inequality. To quote another recent Business Insider headline, “The millennial wealth gap is growing as some flock to buy houses and others give up on homeownership entirely.” The new opportunities presented by remote work are largely reserved to college graduates, while even much-cheaper apartments in expensive cities are still too expensive for most. According to Zumper’s latest rent report, the 24.3% year-over-year drop in asking prices for San Francisco one-bedroom apartments still leaves a median price of $2,650, which over a full year adds up to 75% of the $42,212 median income of U.S. 25-to-34-year olds. Meanwhile, in many cheaper cities away from the coasts, rents are up by 10% or 20% over the past year. For many millennials, the economics of striking out on one’s own remain pretty daunting.
That's based on the September households estimates from the vacancy survey, which I used because the October, November and December ones aren't available for as long on a continuous basis.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
Justin Fox is a Bloomberg Opinion columnist covering business. He was the editorial director of Harvard Business Review and wrote for Time, Fortune and American Banker. He is the author of “The Myth of the Rational Market.”
©2021 Bloomberg L.P.