Asset Classes
Retail has quietly become one of the tightest major property types in the country, and almost nobody is building to fix it. US retail construction completions fell to 4.7 million square feet in the first quarter of 2026, the lowest quarterly figure since CBRE began tracking the metric in 2005, and down from a peak above 25 million square feet in late 2015.
The demand side did not go anywhere. National retail vacancy sat at 6.0 percent in the second quarter, a hair above its record low, with net absorption (the change in occupied space over the period) positive at 708,000 square feet and the active development pipeline down to less than 0.3 percent of existing inventory, per Cushman and Wakefield. Only 2.3 million square feet of new space delivered in the quarter. Asking rents pushed to $25.65 per square foot, up 2.2 percent year over year. When you deliver almost nothing into a market that is still absorbing space, the tenant, not the landlord, is the one who runs out of options.
This is a supply strike, not a demand miracle. Ground-up retail pencils badly right now: construction and land costs are high, anchor tenants are cautious about opening new footprints, and lenders treat speculative retail as roughly the last thing they want on the books. So the space that already exists is the space you get. Availability, the share of space being marketed for lease whether or not it sits empty today, ticked up just 10 basis points to 4.9 percent nationally, still historically low. Capital has noticed: retail investment volume over the trailing twelve months reached $62 billion, up 31 percent year over year, per JLL, even as most other property types stayed quiet.
We have argued for years that the boring version of retail, grocery-anchored centers and small-bay strip in growing suburbs, is a better risk than the market gives it credit for. The supply picture is the reason. A center leased today faces almost no new competition down the road, because the competing center is not being built and, at these costs, will not be for a while. That is a durable moat you did not have to pay for. The real risk in retail now is tenant credit and consumer health, not a wall of new supply, and we would rather underwrite the former.
Our Backyard
The headline in Southwest industrial is that the correction is ending. The finer print is that it is ending at very different speeds, and the map matters more than it has in years.
Phoenix led the turn. Industrial vacancy there fell to 10.8 percent in the second quarter, a nine-quarter low, down 240 basis points from a year earlier, on 6.1 million square feet of net absorption in the quarter alone, per Cushman and Wakefield. Las Vegas told the same story more quietly, vacancy down to 10.1 percent, off 70 basis points year over year. Both markets built too much in 2022 and 2023, and both have now spent enough quarters leasing it up that the trend has turned.
Boise is the exception, and it is worth understanding why. Treasure Valley industrial vacancy rose to 9.6 percent, up 180 basis points year over year and 80 basis points in a single quarter, per Cushman and Wakefield. The level is not alarming. The direction is the point. Boise is a smaller, later-cycle market still digesting its own deliveries while the larger metros have moved past theirs, and a single large building swings the number. Its demand anchors, Micron's expansion and the data center cluster out in Kuna, are lumpy by nature. This is roughly where Phoenix stood earlier in the same cycle the big markets are now exiting.
Reno sits in between. Northern Nevada industrial vacancy was 11.7 percent, still elevated but down from its late-2024 peak, and net absorption has stayed positive, 1.9 million square feet year to date, per REBusinessOnline. The supply overhang is real, with 1.6 million square feet under construction and another 15.8 million in planning, but the demand to chew through it is showing up.
For an operator, this is the difference between buying the recovery and waiting for it. In Phoenix and Las Vegas the vacancy math already works in the owner's favor, and pricing reflects it. In Boise and Reno the same fundamentals sit a few quarters behind, which is exactly where a patient buyer wants to be if the demand thesis holds. A 9 or 11 percent vacancy rate that is falling does not worry us. One that is rising with no anchor behind it does, and neither of these markets is that.
One retail footnote from the same backyard, because it rhymes with the national picture. Boise retail vacancy rose to 4.4 percent, its highest in more than two years, per Cushman and Wakefield. Nationally, retail vacancy is climbing almost nowhere because nothing is being built. In Boise it climbed a little because something finally was. In a market starved of new retail, about the only way vacancy rises is if a developer actually delivers, and most are not.
Sources
- CBRE, US retail construction completions (via Bisnow)
- Cushman and Wakefield, US Retail MarketBeat Q2 2026
- Cushman and Wakefield, Phoenix Industrial MarketBeat Q2 2026
- Cushman and Wakefield, Las Vegas MarketBeat Q2 2026
- Cushman and Wakefield, Boise MarketBeat Q2 2026
- JLL, US retail investment (via Yahoo Finance)
- REBusinessOnline, Northern Nevada industrial market
Frequently Asked Questions
- Why is so little new retail being built?
- Ground-up retail pencils badly right now. Construction and land costs are high, anchor tenants are cautious about new footprints, and lenders treat speculative retail as high risk. So completions hit a 20-year low in early 2026 even as existing space keeps leasing.
- Is low retail construction good or bad for owners?
- Good for owners of existing, well-located centers. With the national pipeline under 0.3% of inventory, a leased center faces almost no new competition, so occupancy and rents hold. The risk shifts from oversupply to tenant credit and consumer health.
- Is Southwest industrial recovering everywhere at once?
- No. Phoenix fell to a nine-quarter-low 10.8% vacancy and Las Vegas to 10.1%, both tightening, while Boise rose to 9.6% and Reno sat at 11.7%. The correction is ending, but at very different speeds by market.
- Why is Boise industrial vacancy rising while Phoenix falls?
- Boise is a smaller, later-cycle market still leasing up recent deliveries, and single large buildings swing its numbers. Demand anchors like Micron and the Kuna data center cluster are lumpy. The level, 9.6%, is not alarming; the direction is the point.
- What does this mean for how VAC invests?
- In Phoenix and Las Vegas the vacancy math already favors owners and pricing reflects it. In Boise and Reno the same fundamentals sit a few quarters behind, which is where a patient buyer wants to be if the demand thesis holds.