Disruption sparked by COVID-19 looks likely to play itself out relatively soon. But each area’s story will probably be dramatically different.
Joe Seydl, Senior Markets Economist
Dan Alter, Mortgage Solutions Team Lead
The end of the U.S. housing market’s pandemic-induced volatility seems in sight.
A recent and telling stabilization of single-family home sales—surprising in the face of high mortgage rates—has us expecting prices to finish bottoming out across the nation by the end of this year or in early 2024.
By the second half of 2024, we believe, the national housing market will return to a pre-pandemic historical norm in which residential home prices climb slowly and steadily, keeping modestly ahead of the rate of inflation.1
However, we also expect each city’s path back to normalcy to be very different.
What do all these factors add up to for buyers and sellers? We think it would be helpful to:
- Understand that the current, ahistorical housing market favors neither buyers nor sellers. Buyers may benefit from timing the continuing downturn, but prices are only coming down from inflated pandemic-inspired highs—and the stock of existing homes is still in (mostly) short supply.
- Know that valuations may slip, but the slide could be limited. Sellers, for their part, may see their home valuations dip further through the course of this year. But, again due to limited supply, the market bottom may not be as low as they fear.
- Do your homework. Keep in mind that there are significant nuances to each area’s housing markets in the best of times—and these are strange days indeed. So whether you are a buyer or seller, it is important to gather as much information as possible about ongoing trends at the national and local levels.
The nation’s housing market has begun to stabilize
The COVID-19 housing upheaval began in March 2020 when lockdowns inspired people to move out of densely populated urban areas in favor of areas with larger homes better suited to remote work.
Many local housing markets saw dramatic price surges. For the nation as a whole, home prices rose a startling 39% from June 2020 to August 2022—igniting fears that a new housing bubble was emerging.
As inflation also started to rise significantly, the Federal Reserve responded by instituting the fastest series of interest rate hikes in the past 40 years. In 2022, the Fed drove its policy rate interest rate from 0% to 4.5%, with the 30-year fixed mortgage rate rising from 3% to a peak of more than 7% by the fall.
These interest rate increases quickly deflated the nascent housing bubble in the United States, where market activity plummeted last year as mortgage rates spiked. Now, though, we’re seeing some tentative but clear signs that suggest this period of intense volatility may soon be ending, and that stability may be on the horizon. Pending sales of existing homes recently ticked up. Moreover, homebuilders’ confidence about future sales (a leading indicator) is on the rise.2
The road to recovery starts with the recent uptick in pending home sales
U.S. Pending Home Sales Index (2001 = 100)
A housing recovery may come by 2024
These recent, positive indicators are a little surprising, given the 30-year mortgage rate remains at levels (6% to 7%) that many would have thought too high to promote any stability.
Today’s improvement seems to be due in large part to the fact there are nationally too few single-family homes relative to population pressures. The good news is, these drivers may continue to act as guardrails for the U.S. housing market:
- Demand—More than 30% of young people aged 18 to 34 are still living at home with their parents, according to the U.S. Census Bureau. Meanwhile, millennials (aged 27–42 in 2023) form the largest generation group in the United States at approximately 72 million. The effort of all these young adults to own homes creates a source of demand that is likely going to be a theme for years into the future.
- Supply—Meanwhile, there’s a shortage of housing stock. Since the bursting of the 2008 housing bubble, new home development in the United States has lagged demand. We estimate that today, this gap totals approximately 2-2.5 million too few housing units. And the recent spike in mortgage rates is, counterintuitively, exacerbating this shortage, as homeowners who refinanced when rates dipped are now disincentivized to move and take a new mortgage at a higher rate (2020 had the largest number of refinances on record; 2021 had the third highest, trailing only 2003).3
We think these drivers will not only cushion how low prices will go over the next year, but can also help shorten the lag between the transaction trough, when sales activity bottoms out, and when prices follow to their nadir.4
Right now, aggregate home prices are down about 4-5% from the national peak in 2022. We believe aggregate home prices will decline, on average, about 10% from peak to trough when the pandemic’s full fallout is complete.
In other words, we expect the decline to be smaller and come sooner than the crash the U.S. housing market experienced after the 2008 Financial Crisis, when single-family home values depreciated by about 30% from peak to trough. That nadir came a full two-and-a-half years after buying activity bottomed out in late 2008.
Also, unlike the 2008 crisis that depressed home prices fairly uniformly across the country, the amount of volatility an individual city is experiencing depends largely on each locale’s unique market dynamics.
Cities are on very different paths back to normal
Last fall, as the housing market was cooling, we said that national statistics, while informative, would fail to tell the whole story. We expected cities to experience widely differing drops in home prices based in large part on how overheated their particular housing markets were at the start of 2022, before interest rates rose.
Subsequent home price drops have been dispersed, and in the general direction we anticipated.
Proven right: Our 2022 expectation that cities would see widely divergent home price drops
Proven right: Our 2022 expectation that cities would see widely divergent home price drops
To see “why” there has been great variety among local housing markets, we think it is instructive to compare and contrast Austin (a city to which remote workers ran during the pandemic) and New York City (one they fled).
When this historically anomalous chapter in the U.S. housing market is finally closed, we expect Austin to have experienced greater price volatility, with a peak-to-trough decline of 15% to 20%. Yet the price drops in Austin sow the seeds for more affordable homes over the long term than in New York, where we expect prices to experience less volatility (probably a 5% to 10% peak-to-trough decline).
The key reasons for these differences are due to demand and supply drivers, which generally apply to any city.
In Texas as a whole, but Austin especially, there has been a quick and substantial supply response to the pandemic-sparked surge in demand for large homes with space for remote work.5 Today, Austin’s inventory is 10%–15% above its level before COVID-19 hit. This rapid response in supply can help keep prices in Austin affordable over the long term.
By contrast, New York’s inventory (much of it space-constrained) is currently significantly below (30%– 40%) its pre-pandemic level. Interestingly, Bridgeport, Connecticut, and its environs (suburbs of New York City) have seen even less of a supply response, with inventory currently about 60% below their pre-pandemic levels!
Austin is building new homes quickly—NYC and its surrounding suburbs are not
Supply deficits in New York (and the surrounding metropolitan region) restrict how much home prices can fall—even as buying activity has slowed considerably. Over the past year, single-family home sales in New York have declined about 38%.
On the flip side, the Big Apple’s inability to build quickly to meet demand—stymied as it is by more regulations, greater regulatory fees and higher labor costs—might continue to stifle its affordability.
One long-term effect of these dynamics can be a continuation of population shifts away from less affordable cities like New York and toward places like Austin. First-time home buyers and young adults in particular won’t wait for supply dynamics to improve in cities like New York; they’ll keep opting for more affordable metro areas that increasingly offer attractive job prospects—and in many cases, a more temperate climate.
Expensive states like New York keep losing population while the more affordable (like Texas) gain
Let’s talk about your city
Are you thinking of buying, selling or building a home? Reach out to your J.P. Morgan team. They can work with you and your realtor to provide comparisons of major metro areas’ housing markets.
1For the 50 years prior to the pandemic, existing home prices grew about 5% a year (one percentage point above inflation), a result, we believe, of population growth and scarcity of land.
2Zonda National Builder Survey asked: “How is January (2023) shaping up so far for your local operation in comparison to your expectations?”; 37% of the 150 to 200 respondents (larger, publicly traded builders) said “stronger than expected” and 36% said “on track,” while only 13% said “slower…but not worrisome” and 14% said “slower…and causing concern.” Importantly, in the same survey, 51% said they saw a positive response by consumers to the drop in mortgage rates from 7% to 6% in January.
4There is anecdotal evidence suggesting developers are offering substantial closing cost credits and funding buyers’ mortgage rate buydowns to move their excess stock. However, new housing completions make up less than 1% of the total housing stock annually. So the overall housing market remains in a state of shortage, even though some builders may have some excess new inventories they are trying to reduce. Sources: Bloomberg and J.P. Morgan.
5These two variables, size and remote share, were key drivers of housing demand across U.S. cities when the pandemic first began. To illustrate, we ran a multivariate regression with these two variables (while also controlling for population growth) against our z-score metric on housing valuation at the city level. The r-square of the regression was 0.42, with both variables of interest (size and remote share) being statistically significant and carrying a positive sign on the estimated coefficients.