Three years into a prime office recap that actually paid what it said

Since I've posted two losses in this room I owe the other one.

In early 2023 I put $60k into a recapitalization of a newer office tower in a strong downtown submarket. Built 2016, roughly 240,000 sf, 91 percent leased at the time, weighted average lease term a little over seven years. The existing owner needed equity to pay down a loan ahead of maturity, so new money came in at a basis well below what the building cost to build.

What I liked before I invested, and I checked each one in writing:

Rent roll expirations. Under 8 percent of rent rolled in the first three years. That was the whole reason I looked twice, after what happened to me on a suburban deal.

Lease structure. Almost all of it was net or modified gross with expense pass-throughs and annual escalations, so operating cost increases didn't land entirely on ownership.

Debt. Fixed rate, five years remaining after the recap, DSCR comfortably above 1.3 at close. No rate cap to expire.

Reserve. $6.1M for leasing and capital, against a building where a full floor retenant was modeled at about $1.9M all in. Three retenants covered. That was the number that convinced me.

What nearly broke it: one tenant on floor nine, about 11 percent of the building, went into a restructuring in 2024 and stopped paying for five months. Distributions were cut in half for two quarters while the sponsor worked it. They ended up assigning the space to a larger tenant already in the building who wanted to expand, at a higher rent, and the reserve covered the gap. If that expanding tenant hadn't existed, we'd have been looking at 26,000 sf on the market in a year when nobody was signing.

Where I am: $60k in, about $14,200 distributed over three years, position marked slightly above cost on the last statement. Not a home run. It paid what it said it would pay and that's what I wanted.

What I'd keep: the expiration schedule check first, the reserve sized to multiple retenants, and no floating rate exposure I can't see the end of.

Answer

2monica_tran3SF4CPRO6

I want to press on the reserve, because that's the number marlow says convinced him and it's the number most people skim.

$6.1M against a modeled $1.9M per floor retenant. That's three floors covered. In a 240,000 sf building at, say, 20,000 sf a floor, three floors is 25 percent of the building. So the reserve covered a quarter of the building turning over.

Here's what I'd check if I were reading this deal cold. Was the $1.9M figure all-in? Tenant improvement allowance, leasing commissions, free rent months, and the downtime carry where the space earns nothing but still pays taxes and insurance. Most sponsor retenant numbers I've seen include TI and commissions and leave out free rent and carry, which can be 30 to 40 percent of the real cost. If that $1.9M was missing free rent, the true number is closer to $2.6M and the reserve covers two floors, not three.

That still would have been enough here, because only one 11 percent tenant went sideways. But it changes what a bad case looks like. Two floors of coverage in a market where nobody signs is a different deal from three.

marlow, the other thing you did that I'd copy is checking the expiration schedule before anything else. Under 8 percent rolling in three years on office is the single strongest fact in your post. Everything else, the reserve, the fixed debt, the pass-throughs, is defense. The low rollover is why you didn't need much defense. If that number had been 35 percent, no reserve size would have made me comfortable.

23 replies

1hannah_west29 Get yours

$14,200 on $60k over three years is about 7.9 percent a year on average, call it a high 7 handle if the timing was even. That's a fine number for the risk, but I want to know how much of it came in year one versus after the tenant restructuring. If years two and three were half rate, the trailing yield is a lot less smooth than the average makes it sound and anyone reading this should see the shape, not the total.

2angela_burkeServicePRO6

The part that made this work is the expanding tenant already in the building. That's luck with a structural component, because a 240,000 sf building with a diverse roster has internal demand a 60,000 sf building doesn't. Worth naming that as a size effect rather than a sponsor skill.

3tonyv88SO4CPRO6

Managing the property side of a few office buildings, I'd say the assignment to an in-building tenant is more common than cobalt is giving credit for, but only in newer well-run buildings where tenants actually want to stay. In older stock the expanding tenant leaves for better space instead. So the vintage did some of the work here too.

1marcus_hale41CG4A

thatch's free rent point is the one I'd hold onto. I asked a sponsor to break out a per-floor retenant assumption line by line and the free rent was three months on a seven year lease, which nobody had priced into the reserve. They added it after I asked and the reserve went up 18 percent.

Ask for the retenant build-up as a table, not a number.

2dmitri_v6CapitalPRO6

On the assignment, marlow, do you know whether the lease permitted assignment without landlord consent, or did ownership have to approve it? Those are very different positions when a tenant is restructuring, and the answer sits in one clause. If the tenant could have assigned to anyone, you got a good outcome from a weak clause.

-1aaron_fitz10 Get yoursPRO6

Reading this next to your two loss posts is the most useful thing in this room for me. Same person, same type of asset, opposite outcomes, and the difference is almost entirely rollover and lease structure. That's a decision rule I can actually use.

1stephen_ok9CO2CPRO6

What was the basis per foot at the recap versus replacement cost? You said well below, but the number matters. Buying newer prime at a discount to what it would cost to build is a different bet from buying older stock cheap, and I'd like to see the spread.

1luis_carrera Get yours

meridian, that spread is the thing I keep circling. I've built a spreadsheet comparing entry basis to estimated replacement cost on eleven office deals and the correlation to outcomes is not as clean as I expected. Low basis on obsolete buildings didn't protect anyone, because the building still couldn't compete for tenants at any rent. Prime at a discount seems to work. Secondary at a discount just delays the problem.

1chriswalton8CC1CPRO8

windlass, that's a genuinely useful finding and it explains why the same phrase, bought below replacement cost, means opposite things depending on the building. Nobody rebuilds a 1985 suburban three story either way, so replacement cost isn't a floor there.

2wendy_park48CapitalPRO7

warrant asked the right question and I'll extend it. In a tenant restructuring, whether ownership gets to approve an assignment often depends on how the proceeding treats the lease, and that's a matter for a lawyer with the actual documents, not a forum. What I'd want out of the file is the consent standard in the assignment clause and whether there was a recapture right letting ownership take the space back instead of accepting an assignee. Recapture is the clause that turns a bad assignment into a leasing opportunity. It's usually two sentences and it's usually skipped.

3carlos_nunezActive

Five months of a tenant not paying and distributions only halved. That implies the reserve or the other rent was carrying real weight. Most deals I've been near, five months of an 11 percent tenant going dark stops distributions entirely.

1jessreyes12AU1APRO8

Basic question I would not ask out loud in a room. What's a recapitalization actually doing? Is new money buying the building or buying part of the old owner's position?

4raj_mehta14CapitalPRO6

compass, I had to look this up too. As I understand it the building doesn't change hands. New equity comes in, the existing owner's stake gets reduced, and the cash usually goes to paying down the loan or funding leasing. So you're buying into an existing capital stack rather than buying an asset. Someone correct me if that's off.

-1joe_delgado5AP4CPRO6

vellum has it close enough. The thing I'd add is that the price you come in at is negotiated, not set by a market sale, so you're relying on somebody's valuation of the building rather than a bid. That cuts both ways. marlow got in below replacement cost because the old owner had a loan maturity pressing on them.

-1mikeoconnell7SP2APRO6

The loan maturity pressure is the part of this whole cycle that produces deals like marlow's. A lot of office debt is coming due through 2026 and owners who can't refinance at the same proceeds have to bring equity or sell. That's where the below-cost entry comes from.

1bmoreno23AB1B

Starting from nothing here, and what I'm taking from this is that the good office deal and the bad office deal looked similar on the summary page. Occupancy in the 70s versus the 90s, and rollover in three years of 8 percent versus 19 percent in fourteen months. Those are two numbers. Is it really that concentrated?

1karen_whitfieldPG4APRO7

ember, mostly yes for screening. Those two numbers plus the lease type plus the debt maturity get you past a large share of the bad ones. They don't tell you the building is good, they tell you it isn't obviously broken. The rest takes walking it.

1brandonlee36Passive

ledger's question about the shape of the distributions still hasn't been answered and it's the one I'd want. marlow, was it roughly $6k, $4k, $4k, or was it front loaded?

1paulgrant17PassivePRO5

The tenant credit side is what stands out to me. On a house I screen one person. Here a single tenant at 11 percent of the building goes into restructuring and everyone's distribution gets cut. What does screening even look like for an office tenant that size, financial statements?

1ehernandez74PO2BPRO5

wren, from the passive seat you're screening the sponsor's underwriting of the tenant rather than the tenant. What I'd want to see is whether the sponsor got financials or a guarantee, how much security deposit or letter of credit was posted, and what the tenant's industry looks like. marlow, did that floor nine tenant have a letter of credit? Five months unpaid with distributions only halved makes me wonder if something was drawn.

2kevin_lombardPG1BPRO5

I went through this thread twice before responding. What lands is that the reserve wasn't there to make the returns better, it was there so one tenant failing didn't end the deal. I'd been thinking of reserves as money sitting idle.

1tylerbrooks9PassivePRO5

This is the version I wanted to see. Newer building, long leases, fixed debt, big reserve, and a return in the high single digits that showed up. It's less exciting than the deals that lost money, which seems to be the pattern in this room.