National Completions Have Cooled Sharply — But the Sun Belt, Northeast, and Midwest Are Living Through Different Cycles
U.S. multifamily completions ran above 500,000 units in 2025, one of the largest supply waves on record. A year later, the U.S. Census Bureau’s August 2026 data shows completions of buildings with five or more units running well below their August 2025 pace, even as permits for the same category remain roughly in line with, or modestly above, year-ago levels. The easy headline is that the supply wave is breaking. The market-by-market data tells a more complicated, and more useful, story for investors.
The National Numbers
According to the U.S. Census Bureau’s August 2026 New Residential Construction report, completions of buildings with five or more units ran at a seasonally adjusted annual rate of 302,000 — 27.1% below the August 2025 pace. Permits for the same category came in at 467,000 SAAR, and starts at 344,000 SAAR, both holding roughly steady with year-ago levels even as completions fell sharply. Across all housing types, total permits were up 3.5% year-over-year against an August 2025 base of 1,347,000 SAAR, while total starts slipped 1.2% year-over-year from 1,291,000 — evidence that the multifamily completions decline is sharper than the broader housing market’s trend, not simply an extension of it.
CBRE reported that U.S. multifamily net absorption reached 167,500 units in the second quarter of 2026, nearly double the first quarter’s 84,300 units, while vacancy fell to 4.3%, down 50 basis points quarter-over-quarter and still above its roughly 5.0% long-term average. New deliveries totaled 77,700 units in the second quarter, down 14% year-over-year. Falling supply and accelerating absorption moving in the same direction, in the same quarter, is the combination behind the “supply wave is breaking” narrative — and the underlying data supports that narrative more than it complicates it, at the national level.
What’s Actually Driving the Number
The mechanics behind the completions decline matter more than the headline figure. The 500,000-plus units completed in 2025 — per the National Apartment Association’s 2026 Apartment Housing Outlook — largely reflect projects that broke ground two to three years earlier, during the low-rate, high-demand development cycle of 2022 and 2023. The current pullback in starts, at 344,000 SAAR for buildings of five or more units, signals fewer deliveries in 2027 and 2028. It is not evidence of an immediate shift already visible in 2026’s completed inventory.
Demand is normalizing alongside supply, not collapsing. The National Apartment Association’s 2026 Apartment Housing Outlook projects absorption of 350,000 to 400,000 units for the year, down from roughly 463,000 in the year ending September 2025 — a forecast, not a reported figure, and one that still represents a solid absorption pace by historical standards. Rent growth is following a similar pattern: CBRE reported national asking rent up 1.5% quarter-over-quarter in the second quarter of 2026, to an average of $2,257, though still only 0.5% higher year-over-year — consistent with the National Apartment Association’s outlook that national rent growth trends toward roughly 2.0% annually in 2026, after slowing through 2025 from 0.9% to 0.6% between the second and third quarters.
“A portfolio approach does not predict which region outperforms next; it reduces the cost of being wrong about any one of them.”
One National Story, Several Regional Ones
The regional breakdown is where “multifamily is not one national market” stops being a slogan and starts being a data point. The National Apartment Association’s 2026 outlook projects rent growth of 4% to 5% in the Northeast, 3% to 4.5% in the Midwest, 2% to 3% on the West Coast, and just 1% to 2% in the Sun Belt — figures broadly corroborated by CBRE’s regional year-over-year rent data as of the second quarter of 2026, which showed the Midwest up roughly 2%, the Northeast up 1.7%, and the Pacific region up 1.4%.
The Sun Belt’s lag is not a coincidence. It is the region that absorbed the largest share of the 2023–2025 supply wave, and markets that took on the most new inventory during that period are, unsurprisingly, the ones still working through it. Markets that saw comparatively little new construction during the same period — concentrated in the Northeast and parts of the Midwest — are seeing pricing power return faster, simply because there is less new supply competing for renters. This is a geography-specific dynamic, not a single national trend, and it is exactly the kind of dispersion that makes market-by-market selection more relevant than a national headline. It is also the kind of dynamic Longview’s market screening is built to account for: markets are evaluated individually, on their own supply pipeline and absorption trend, with an eye toward more balanced supply demand fundamentals, rather than selected because a national narrative happens to be favorable.
What the Data Doesn’t Prove
Two caveats are worth stating plainly. First, Census starts data carries real statistical uncertainty: the August 2026 release notes that the reported 1.2% year-over-year decline in total starts carries a confidence interval of plus or minus 10.8 percentage points, meaning the actual direction of the change is not statistically certain from this release alone. Second, completions data is inherently backward-looking, and the 27.1% year-over-year decline partly reflects an unusually large August 2025 comparison base — a base effect, not solely a change in current building activity.
Falling supply also does not mechanically guarantee rent growth or occupancy gains. Demand — job growth, household formation, and the relative affordability of renting versus owning — has to hold up as well, and the National Apartment Association’s 350,000-to-400,000-unit 2026 absorption figure is a forecast, not a reported result. The regional rent-growth ranges cited above are forecasts too, drawn from the same outlook, and should not be read as a prediction of how any specific Longview asset or market will perform. A slower Sun Belt recovery and a stronger Northeast recovery are both plausible outcomes under current data; neither is guaranteed.
Built as a Portfolio, Not a Single Bet
The dispersion in this data — a Northeast market projected to grow rents 4% to 5% against a Sun Belt market projected to grow them 1% to 2% — is a structural argument for diversification, not a marketing line layered on top of it. Because supply and demand dynamics vary this widely by metro, a portfolio built across multiple markets is structurally positioned to reduce exposure to any single region’s supply cycle — Sun Belt oversupply, for instance — relative to a concentrated bet on one market or one asset. Longview’s selection process reflects that logic: markets are chosen for more balanced supply/demand fundamentals on their own merits, not named here as specific targets, since any such claim should rest on verified, asset-level data rather than the market-level trends this article covers.
That is the practical case behind “Built as a Portfolio. Not a Single Bet.” The national completions and absorption data above describe real, improving conditions in aggregate. But aggregate conditions are not the conditions any single asset actually experiences, and the regional spread in this data is the clearest illustration of why. A portfolio approach does not predict which region outperforms next; it reduces the cost of being wrong about any one of them.
Longview Commercial does not provide tax, financial, or legal advice. This article is for general informational purposes only. Please consult your own qualified tax, financial, or legal advisor before making any investment decision.