Stop Waiting for Someone Else to Go Dark

CRE Daily reported on September 25 that Coresight Research "counted a net gain of 127 store openings through Aug. 28, representing 28.6M SF of newly occupied space," and that "through the same point in 2025, retailers had closed a net 1,954 stores." The same brief puts national retail vacancy at about 4.4%, JLL's second-quarter net absorption at 10.2 million square feet against roughly equal negative absorption a year earlier, and CBRE's second-quarter completions at 5.2 million square feet — a record low, with construction volume roughly 40% below 2016 levels.

Our position: the closure count is the number that moved, and it moved against anyone whose site strategy depends on someone else going dark. A swing of roughly two thousand stores in the same eight-month window is a supply event, not a sentiment reading. If you need a box in 2027, plan to build it.

The dark box was a channel, and it has narrowed

A net 1,954 closures through late August of last year is a lot of space arriving already zoned, already served, already paved. Waiting for it was a workable way to get a box. Per the CRE Daily summary, that flow has reversed this year.

Note also what the two figures say together. A net gain of 127 stores paired with 28.6 million square feet of newly occupied space points to openings much larger than the closings. In our read, the closures that are still happening are not producing the box sizes most pad and small-shop users want.

Construction is not covering the gap

Per the same summary, average retail rent is up 19% since the second quarter of 2020, to just under $25 per square foot, while the Associated General Contractors of America puts cement, lumber and sheet metal 30% to 60% above February 2020 levels. Costs moved further than rents. That is most of the explanation for a record-low completions quarter, and it is why we do not expect new supply to reopen the channel on its own.

The Southeast is the exception, and that changes the job here

Per JLL's second-quarter retail report, summarized by Institutional Real Estate in August, construction is on pace to stay historically thin and concentrated in a handful of Sun Belt metros. Several of those are the markets we work — Georgia, the Carolinas, Florida.

That is a different problem than a built-out market has. Here, new pads are still getting delivered. Our competition is other developers bidding the same corners, not an absolute shortage of dirt.

What is still coming loose is small

Restaurant Dive reported on September 22 that Yardbird Southern Bar and Table filed for Chapter 11 with three company-owned and two licensed locations remaining, about $25 million in debt, and closures in Denver, Los Angeles and Miami already behind it. Per that report, a stalking-horse asset purchase agreement is in place with higher offers to follow.

That is one shape of a 2026 restaurant closure: a handful of boxes, full service, in large metros. It does not restock a drive-thru pad market in Georgia or the Carolinas.

What we watch for

We underwrite the ground-up path first and treat a found second-generation box as upside, not as the base case.

We track the closure count monthly rather than the vacancy rate. Vacancy at 4.4% is the result. The closure count is the input that changes first.

We price seller flexibility as part of the number. Shopping Center Business reported in June, citing Colliers, that the market "remains structurally tight due to limited new supply and steady backfill demand," with retailers favoring 1,500 to 2,000 square foot formats. When competition for small second-generation space is that deep, the seller of a clean raw pad has more leverage this year than last, and we would rather pay for a long feasibility period than assume another box appears.

And we would want two or three more monthly readings before calling the closure cycle finished. One eight-month count is a data point, not a trend.