NextFin News - Bess Freedman, chief executive of luxury brokerage Brown Harris Stevens, said the United States needs more housing in the market, calling supply the biggest challenge facing buyers and sellers alike. Her message, delivered in a television interview on September 16, lands in a market that should not be able to look this way: inventory has rebuilt to its highest level in more than a decade, yet sales have fallen to a 14-month low and the median home price keeps rising. More homes are available, but fewer are changing hands. That contradiction is the clearest sign that the housing market's problem is not a temporary inventory shortage - it is a structural deficit that no single interest-rate cycle will fix.
The numbers set up the tension. Existing-home inventory climbed 3.2% in August to 1.62 million units, the highest level since November 2019, and supply now stands at 4.9 months - the most in over ten years, and nearly inside the five-to-six-month range that defines a balanced market. Yet existing-home sales fell 2% in August to a seasonally adjusted annual rate of 3.98 million, the weakest pace since June 2025, even as the median price edged 1.6% higher year over year to $429,100. Inventory is normalizing. Transactions are not. And with the 30-year fixed mortgage rate back near 7% - averaging 6.71% in the week of September 10, the highest in more than a year, and 6.76% as of that date, up from 6.35% a year earlier, according to Freddie Mac - the lock-in effect keeps owners with sub-4% mortgages from listing, while first-time buyers, who made up only 30% of purchases, remain priced out.
The Inventory Rebound Is Real, but It Is Not Unlocking Transactions
On the surface, the market appears to be healing. The National Association of Realtors reported on September 10 that existing housing inventory rose to 1.62 million units in August, up 5.9% from a year earlier. At August's sales pace, it would take 4.9 months to exhaust the current supply, up from 4.6 months in July and the same month a year earlier. By the textbook definition, the United States is almost back to balance.
But the mechanism behind that rebuild matters more than the level. This inventory is not accumulating because a wave of eager sellers has arrived. It is building because demand has weakened faster than new listings. Every quarter-point rise in the mortgage rate removes tens of thousands of buyers from qualification range, and the lock-in effect - the gap between a 3% legacy mortgage and a 7% replacement - keeps would-be sellers tethered to their current homes. The result is a market that is simultaneously better supplied and less liquid. August sales fell 1.2% year over year, ending a four-month streak of annual gains, and first-time buyers accounted for just 30% of purchases, well below the roughly 40% share that characterizes a healthy market.
"We need more housing. That's the biggest challenge right now," Freedman said in the interview.
She is right about the direction of the cure. But the diagnosis needs a second layer: the United States does not merely need more homes listed this quarter. It needs a sustained increase in the housing stock that survives the interest-rate cycle.
The Shortage Is Structural, Not Cyclical
Here is the central judgment of this analysis: the housing shortfall is structural, and rate relief alone will not close it without a step-change in construction. Freddie Mac estimates the U.S. housing stock - currently about 147 million units - is 3.7 million units below the 150.7 million units the market needs given the current population. That deficit persists despite the addition of 5.8 million units over roughly four years, because households increased by 6.3 million over the same period and target, or latent, households by 6.9 million. Construction has been running behind demand for more than a decade, and the gap is measured in millions of units, not months of listing inventory.
Other estimates run wider. Research published by a major listing portal in March 2026 put the supply gap at more than 4 million homes in 2025 and found it is geographically concentrated, with the Northeast facing the most severe deficit relative to pre-pandemic construction while many Southern and Western metros now hold more for-sale inventory than during the 2017-19 period. The Congressional Research Service, summarizing the academic literature, noted that shortage estimates vary by methodology but agree on direction: the United States has underbuilt for years, and the burden falls hardest on the lowest incomes. The National Low Income Housing Coalition estimates a shortage of 7.1 million rental homes affordable and available to renters with extremely low incomes.
Why didn't builders fill the gap when the demand was obvious? Three structural constraints, not cyclical hesitation. First, land and zoning: in the metros where jobs and migration concentrate, zoning rules and approval timelines cap density before a shovel hits the ground. Second, construction costs and labor: the NAHB/Wells Fargo Housing Market Index stood at 35 in August 2026, well below the 50 threshold that separates builder optimism from pessimism, reflecting elevated regulatory and material costs. Third, development financing: even well-capitalized builders face higher carrying costs that push them toward higher price points, which is why new supply skews toward the upper end of the market.
This is why the current inventory rebuild is not the same thing as a structural fix. A 4.9-month supply of existing homes is a flow variable that can evaporate quickly if rates fall and locked-in owners suddenly list, or if buyers return en masse. The 3.7 million-unit deficit is a stock problem, and stock problems are solved only by cumulative construction.
The K-Shaped Market: Who Can Buy and Who Cannot
Freedman has described the current environment as a "K-shaped" housing market - a split where the upper arm moves independently of the lower. The data support the framing. In Manhattan, Brown Harris Stevens' own first-quarter 2026 market analysis showed the average resale price rising 5% year over year to just under $2 million, while the median climbed 13% to approximately $1.13 million. Luxury transactions are being sustained by equity-rich buyers for whom mortgage rates are a secondary consideration; cash and large down payments insulate that segment from financing costs.
Below that tier, the math is brutal. A buyer financing the national median price of $429,100 with 20% down at a 6.75% rate carries a monthly principal-and-interest payment of roughly $2,227, before taxes and insurance. The same home at a 3.5% rate - the level that prevailed for much of the early 2020s - would have carried about $1,541. That is a difference of roughly $685 a month, or 44%, on the identical house. No wage growth over the past three years comes close to bridging that gap, which is why first-time buyer share has not recovered and why existing-home sales remain at a 14-month low even as inventory builds.
The K-shape has a second-order consequence that most commentary misses. It means the housing market can look healthy in the aggregate - prices rising, inventory building, luxury sales firm - while the entry-level market, which drives household formation and broader economic mobility, remains dysfunctional. Aggregate data flatter the reality. The recovery is real for the top decile of buyers and largely imaginary for the bottom half.
Renters Show What Supply Actually Does
There is one segment of the housing market where a construction-led supply response has already worked, and it is instructive. The rental market absorbed a record wave of multifamily construction over the past two years, and the result is visible in the data: the typical U.S. asking rent reached $1,965 in June 2026, up just 2.2% from a year earlier, and nearly two in five rental listings carried a concession. Single-family rents, where new supply has been far thinner, rose 3% year over year to $2,320 - roughly double the 1.5% gain for multifamily units, which now average $1,789. The same good - shelter - is getting cheaper where builders were allowed to build at scale, and more expensive where they were not.
This is the strongest empirical support for Freedman's position, and it exposes the flaw in the counter-argument. Building more market-rate housing does not instantly hand affordable units to the lowest-income renter. What it does is relieve the pressure one income tier at a time: the household that rents a new luxury apartment vacates a slightly older unit, which becomes affordable to the next tier down. Filtering is slow and imperfect, and it works far better in the rental segment - where purpose-built supply can be delivered at volume - than in the for-sale market, where each unit typically requires a separate lot and a longer approval chain. But the rent data prove the mechanism exists. The question is not whether supply works. It is whether the for-sale market can replicate what multifamily construction did for rents.
The policy implication follows directly. A supply solution that relies only on market-rate single-family homes will take a decade to reach the bottom half of the market. A solution that pairs market-rate construction with zoning reform - allowing duplexes, townhomes, and mid-rise apartments near jobs and transit - compresses that timeline. The Northeast's persistent shortage, concentrated in metros with the tightest land-use rules, is not an accident of demand. It is the predictable output of a system that makes new housing illegal to build at the price most households can pay.
The Counter-Thesis: Maybe Building More Market-Rate Homes Does Not Help
The strongest case against Freedman's position is not that supply is adequate. It is that building more market-rate housing will not reach the buyers who need help, and could simply bid up land and labor costs without improving affordability where it matters. Critics of supply-side solutions point to the 7.1 million-unit shortfall of rental homes affordable to extremely low-income renters - a gap that market-rate construction, by definition, does not fill. If new units are absorbed by higher-income households and investors, the marginal buyer at the median sees no relief, and the political case for "just build more" collapses.
There is force in that objection, but it narrows the argument rather than overturning it. A market-rate unit built today relieves pressure on the household that would otherwise outbid a middle-income buyer for an existing home. Filtering - the process by which older homes become affordable as newer stock absorbs demand at the top - is slow and imperfect, but it is the only mechanism that operates at scale without direct subsidy. Evidence from metros that have permitted the most construction shows price growth moderating relative to constrained peers. The counter-thesis is correct that building alone cannot solve the affordability crisis at the bottom of the market. It is wrong to conclude from that truth that building does not matter.
The falsifying signal for the structural-shortage thesis is specific and observable. If annual housing starts sustainably exceed 1.7 million units for three consecutive years while household formation slows below 1.2 million annually, the stock deficit would begin closing on its own and the "chronic undersupply" argument would be wrong. Current data point the other way: building permits stood at 1.433 million in July 2026, and builder sentiment at 35 signals no imminent construction boom.
What Comes Next: Three Horizons
Short term (next three to six months): The market stays range-bound. Mortgage rates near 7% keep the lock-in effect intact, inventory continues to rebuild toward five to six months, and sales hover near the 4 million annual pace. Any signal that pushes the 30-year rate below 6% would unlock a wave of pent-up demand and quickly absorb the inventory overhang - the apparent balance would prove illusory. For sellers, that is the risk of waiting: the moment rates fall, the competition that has been held back by financing costs returns all at once.
Medium term (one to three years): Affordability, not inventory, remains the binding constraint. The beneficiaries are cash-rich buyers and sellers who can wait; the exposed are first-time buyers, trade-up owners carrying low-rate mortgages, and builders whose entry-level projects cannot pencil at current rates and costs. If rates grind lower toward 6%, expect a volume recovery before a price correction; if they hold near 7%, expect continued price resilience with depressed transaction volume - the worst combination for market liquidity. The rental market, by contrast, is likely to keep absorbing new multifamily supply with modest rent growth, keeping the rent-versus-buy math tilted toward renting for households without substantial equity.
Long term (five years and beyond): The structural deficit persists unless zoning reform and construction productivity improve. Demographic demand from millennial household formation and immigration does not reverse on a policy cycle. Markets that liberalize density and streamline permitting will attract capital and moderate price growth; markets that do not will see the K-shape harden into a permanent feature. The divergence between them will be the defining story of the next decade: two housing markets in one country, one functioning and one frozen, separated by a zoning map.
The base case is a slow, two-speed market: inventory normalizes, volume stays subdued, and prices drift modestly higher in nominal terms while falling in real terms for overleveraged buyers. The upside case requires a rate decline that unlocks both buyers and sellers simultaneously, producing a volume-led recovery with stable prices. The downside case is rates holding near current levels while construction costs keep new supply expensive - a prolonged freeze with a slowly rising vacancy overhang in markets that overbuilt the wrong product.
Freedman's call is correct, but incomplete. The United States does need more housing in the market - and it needs it built in the places where people actually want to live, at price points the market can reach without subsidy. More listings will not fix a shortage measured in millions of units. Only construction will. The housing market is not short on inventory; it is short on homes, and those are not the same thing.
Explore more exclusive insights at nextfin.ai.
