Office prices move to occupied square feet

A buyer paid $47.4 million for 410 Townsend, a San Francisco building that is 65 percent leased, and the print is changing the unit in which office buildings get quoted. PWD's tracking shows the buyer paid for the occupied square feet and took the empty third as an option: the in-place rent roll is the asset, and the vacancy is a free option the buyer can exercise by leasing it. On a total-square-foot basis the print looks like a distress number; on an occupied-square-foot basis it lands as the nearest thing to a clean cross-market office comp.

Occupied square feet divides the price by the footage actually leased and paying rent, rather than by the entire building, which turns vacancy from a discount to a hypothetical full value into a separate, zero-cost option. The buyer of 410 Townsend did not pay for the empty third; it paid for the space that generates cash flow, and the core math still works if the empty space never leases. If it does lease, the buyer collects the upside without having paid for it, which is why the San Francisco number is closer to a Dallas comp than any total-area denominator ever was.

Montclif and FCP are doing the same arithmetic on a Dallas office at 63 percent leased, where the trade reads as buying a construction schedule as much as a building—another way of saying the buyer is not paying for vacant space. The 63 percent leased figure is now the headline comparable, not the building area; the price is set by the occupied rent roll while the remaining vacancy rides as the option. A buyer who pays for occupied footage can compare San Francisco and Dallas without pretending that vacancy in one city means the same thing as in another, which is why occupied square feet is the only denominator that lets vacancy be compared at all.

MetLife and the New York City pension fund have brought the same test to Houston with a whole-asset offering on a 70.6 percent-leased tower, where nearly three in ten square feet are open and the winning bid will tell the market what it costs to fill the rest. As the coverage put it, the price will be set by what it costs to fill half a million square feet, not by a recovery multiple on the building's total area. Houston's CBD has lacked a clean office mark; if this offering prints, it will print in occupied-square-foot terms because no buyer is underwriting the empty space at anything but zero.

The debt market arrived at the occupied-square-foot denominator before equity did. New York Life sized its $386 million refinancing of 200 Madison Avenue to Havas's lease, not to a Midtown East recovery, and the three-year floating loan is a bridge across a tenant commitment rather than a bet on the building's total square footage. Hudson Bay's $85 million loan against a Microsoft lease in SoHo is the same transaction in miniature, with one tenant's rent roll as collateral and the rest of the building not what the lender is underwriting. Lenders stopped paying for empty floors years ago; the three-year floating structure against Havas shows they now price the occupied income alone, treating the lease roll as the asset, and the equity side is only now catching up.

The San Francisco, Dallas, and Houston prints converge on the same unit: a buyer paid for occupied square feet at 65 percent, another bought a construction schedule at 63 percent, and a seller is asking bidders to price a 70.6 percent-leased tower by the cost of filling it. None of these disclosures gives a clean per-square-foot price on total building area. What they give is a denominator that lets an allocator compare a 65 percent-leased building in a weak market with a 70.6 percent-leased building in a stronger one. The vacancy is no longer a discount rate to be argued about; it is a free option, and the price is for the occupied part only.

Swickard's Portland tower purchase is the exception that proves the rule, because the buyer now controls 90 percent of downtown Portland's fourth-largest building but the price is not an office comp. The buyer is pricing the floors that are not office; the office floors are the part of the building that has not yet found a bid. That is precisely what the occupied-square-foot denominator says: space with no tenant and no use gets no price until it gets one. A mixed-use buyer can assign value to hotel, residential, or retail floors, while the office component remains the option, so the Portland trade does not print an office mark and should not be read as one.

Nashville's second print tells the same story from a different angle: LBX paid $54.3 million for 134,113 square feet, but the headline price is not the comparable; the residual between the disclosed price and the building's features is. The market's inability to disclose a clean per-square-foot price for the smaller building is itself information, because once buyers are transacting on occupied square feet, the residual becomes the signal rather than the total square footage.

Four years of arguing about what a building is worth have led here: the $47.4 million San Francisco trade did not buy a building at 65 percent leased; it bought the income in place and the empty floors for nothing. Dallas and Houston are now testing whether the rest of the market will adopt that unit. Owners still quoting price per total square foot are solving yesterday's problem with a denominator nobody is underwriting anymore.