By any reasonable forecast made in March 2020, the pandemic should have been a housing recession. Lockdowns, mass unemployment, supply-chain disruption — every component of demand should have softened. Instead, the U.S. housing market entered the most extreme acceleration in its modern history.
Pre-pandemic conditions
The U.S. housing market entered 2020 in a state of mild excess demand. Existing-home sales had run at 5.3–5.5M for four consecutive years (2016–2019), tight by historical standards. Median nominal prices had grown roughly 4% annually since 2014 — well above the long-run average of ~5.5% nominal CAGR but normal-looking after the post-2008 underbuild. Mortgage rates averaged 3.94% in 2019; the 30-year fixed had been below 5% for almost a decade.
The supply side was tight. Post-2008 construction had run well below trend for nearly a decade — the U.S. completed roughly 12 million single-family homes between 2010 and 2019, vs an average of ~16 million per decade in the post-1970 period. By 2019, the U.S. was estimated to be 3-5 million units short of underlying household-formation demand. Vacancy rates for both rentals and owner-occupied stock had fallen to multi-decade lows.
March 2020 — the Fed cuts, the QE response
The Federal Reserve cut the federal funds rate from 1.50–1.75% to 0–0.25% in two emergency moves on March 3 and March 15, 2020. Simultaneously, the Fed announced unlimited quantitative easing — the FOMC committed to buying Treasuries and mortgage-backed securities "in the amounts needed to support smooth market functioning."
The MBS-purchase commitment was the most consequential piece of the policy stack for housing. By April 2020, the Fed was buying roughly $40B of agency MBS per month — and this pace continued through November 2021, with only a brief tapering announcement in late 2021 finally slowing the program. Cumulatively, the Fed's MBS holdings rose from $1.4 trillion in February 2020 to $2.7 trillion by year-end 2021, an unprecedented expansion that pushed mortgage rates structurally lower.
The 30-year fixed mortgage rate fell from 3.4% in January 2020 to 2.65% in early January 2021 (the weekly low of the entire PMMS series). Annual averages tell the story: 3.94% (2019), 3.11% (2020), 2.96% (2021).
Demand-side: stimulus + remote work
Two demand-side forces compounded the rate effect:
Stimulus. The CARES Act (March 2020), supplemental unemployment insurance, the December 2020 stimulus package, and the American Rescue Plan (March 2021) collectively transferred roughly $5 trillion to U.S. households between March 2020 and June 2021. Personal saving rates spiked from 7% (pre-pandemic) to 33% in April 2020, eventually settling at elevated levels through 2021. By mid-2021, U.S. household balance sheets were stronger in aggregate than they had been at any point in the post-WWII period — measured as total assets minus total liabilities relative to GDP.
Remote work. The Census Bureau Household Pulse Survey estimated that roughly 30 million U.S. workers shifted to remote or hybrid work arrangements by mid-2021. For housing, the implications were structural rather than cyclical:
- Many workers no longer needed to live within commuting distance of their employer's office — opening migration corridors that had not previously been viable.
- Within metros, the value of suburban and exurban locations rose meaningfully relative to dense urban cores. The price spread between cities like San Francisco and their suburban hinterlands narrowed sharply.
- The willingness to pay for housing space (extra bedroom for a home office, larger lot for outdoor flexibility) increased materially.
The migration patterns showed up cleanly in the data. The five fastest-appreciating metros of 2020–2022 (Boise, Phoenix, Austin, Tampa, Charlotte) shared two features: high net in-migration from coastal expensive metros, and meaningful housing supply expansion potential. The five slowest-appreciating major metros (San Francisco, New York, Boston, Chicago, Washington D.C.) shared the opposite features: net out-migration during the pandemic period and constrained supply-side response.
Supply-side: existing inventory freezes
The supply-side response to the pandemic-era demand surge was muted by an unusual dynamic: existing inventory froze. Several factors compounded:
First, in 2020 specifically, the lockdown environment made listing a home and conducting in-person showings logistically difficult. Inventory of homes-for-sale fell from roughly 1.7 million units (early 2020) to 1.0 million units by late 2020 — a 41% drop in nine months.
Second, as 2020 became 2021, the rapid price appreciation made many would-be sellers reluctant to list. The mental anchor of "wait three months and the same home will be worth $30,000 more" depressed listing volume even as buyer demand surged.
Third, the inventory drought interacted with rising mortgage rates in 2022 to produce the rate-lock phenomenon (see below). By the time 2022 began, U.S. existing-home inventory was at roughly 800K units — the lowest level recorded since the inventory series began in 1982.
New-home construction tried to fill the gap. New-home sales reached 822K in 2020 and 771K in 2021 — the highest levels since 2006. But residential construction faced its own supply-chain limits: lumber prices spiked from $400/MBF (early 2020) to $1,500/MBF (May 2021), construction labor was constrained by an aging tradesperson cohort and immigration flow disruptions, and entitlement timelines in supply-constrained metros (California, the Pacific Northwest) made it impossible to bring large quantities of new inventory online quickly.
The 2021 price spike
The 2021 calendar year recorded the largest single-year increase in U.S. median existing-home prices in the entire NAR series — 17.5% nominal. The S&P/Case-Shiller National Home Price Index ended 2021 up 18.9% YoY, also the largest single-year reading on record. Roughly half of all U.S. metropolitan statistical areas posted 20%+ YoY price increases in 2021.
For context: the previous record single-year price increase was 13.5% (2005), set during the subprime peak. The 2021 reading exceeded the 2005 reading by 4 percentage points despite happening in an environment of much-tighter underwriting and lower aggregate household leverage. The mechanism in 2021 was not credit expansion (which was actually constrained relative to pre-pandemic) but rate-induced affordability expansion combined with structural demand growth from remote work.
The 2022 unwind
The Federal Reserve began reversing course in March 2022, raising the federal funds rate from 0–0.25% to 4.25–4.50% over the year — the fastest hiking pace since 1980. Mortgage rates responded immediately: the 30-year fixed rose from 3.4% in January 2022 to 7.1% by November 2022, an annual average of 5.34% for the year (up from 2.96% in 2021).
The effect on housing transaction volume was instant. Existing-home sales fell from 6.12M (2021) to 5.03M (2022) to 4.09M (2023) — the lowest reading since 1995. The pace of price appreciation slowed but did not reverse; median nominal prices rose 8.2% in 2022 and 0.9% in 2023, a striking decoupling from the volume collapse.
The rate-lock era (2023–2024)
The 2023–2024 pattern of low transaction volume coexisting with rising prices is the defining feature of the post-pandemic housing era. The mechanism is the rate-lock effect:
By year-end 2023, roughly 76% of mortgaged U.S. owner-occupants held loans below 5%, and 40% held loans below 3%. For these households, the arithmetic of selling and buying a comparable replacement home at prevailing 6.5–7.5% rates is punishing — typically a 60-100% increase in monthly principal-and-interest cost for the same loan balance. The result has been a near-complete listing freeze: U.S. existing-home inventory averaged 1.0 million units in 2023 and 1.1 million in 2024, vs the pre-pandemic norm of ~1.7M and the long-run average of ~2.5M.
The rate-lock effect resolved the apparent paradox of falling sales and rising prices. Demand had cooled but was still meaningful — household formation continues, life events still force transactions, foreclosure-driven distress is essentially absent. Supply, however, fell faster than demand — and prices set at the margin reflect the much-reduced supply curve.
The 2024 housing market is not 2008. The defaults aren't there, the supply isn't there, and the structural credit fragility that drove the GFC was dismantled fifteen years ago. What 2024 actually is: an affordability crisis driven by the highest home-price-to-income ratio in U.S. history.
What it changed
The pandemic-era surge produced three durable structural changes:
- The locked-in mortgage stock. Roughly 50 million U.S. mortgages outstanding at sub-5% rates create a multi-year drag on existing-home transaction volume that cannot resolve through any path other than (a) Fed cuts that bring rates back below 5% or (b) the slow accumulation of life-event-driven listings over time.
- The migration flows. The top-quartile-versus-bottom-quartile metro price-spread narrowed meaningfully during 2020–2022 and has only partly reverted. Some of that narrowing is permanent — the price-of-Boise relative to the price-of-Austin is structurally different than it was in 2019 because the migration path between them has been reset.
- The affordability stress. The U.S. price-to-income ratio reached 5.4× in 2024 — the highest in the modern record. Resolving that stress requires either prices to compress, incomes to grow, rates to fall, or some combination. None of those mechanisms are running fast enough as of 2026 to materially change the picture.
Year archives covered by this explainer
- 2019 — pre-pandemic
- 2020 — Fed cuts to zero, QE begins
- 2021 — 17.5% YoY price spike, 2.96% record-low rate
- 2022 — Fed hikes, rates spike to 7%
- 2023 — rate-lock takes hold (4.09M sales, lowest since 1995)
- 2024 — affordability low, $408K median