The Latticework Letter — Issue #008

Last issue ended with Mr. Market delivering champagne on a record Tuesday. This week he delivered a resignation letter — the labor market's.

Friday's July jobs report printed negative 23,000 — the first outright monthly job loss of this cycle — against a consensus of +83,000. Revisions took another 103,000 out of May and June. The unemployment rate ticked down to 4.1%, but for the wrong reason: people leaving the labor force, not finding work. Wage growth slipped to 3.2% year-over-year, the slowest since May 2021. And the market's response was to hit another record close that same day — because in this tape, bad news for workers is good news for rates.

The data kept cooperating with that trade all week. Wednesday's CPI came in at 3.4% headline, 2.5% core — the coolest core reading since March 2021. Thursday's PPI printed flat against expectations of an increase. Put the three together and the September hike we've been carrying at roughly two-in-three odds for a month has collapsed: by Thursday, futures put hike odds around 35–40%, making a hold the base case for the first time since June. The 10-Year spent the week in a 4.64%–4.70% band, around 4.67% Thursday; the 30-Year sat stubbornly above 5.2% all week.

The catch — there is always a catch — is oil. The Hormuz interim deal that ignited last week's euphoria did not get signed. Iran published a restrictive counter-draft, talks went to "deadlocked" by mid-week, a Houthi missile attack on a commercial vessel killed six, and Brent retraced the entire de-escalation rally: from below $80 last Wednesday to within sight of $90 by this Wednesday, easing to roughly $88 Thursday. Traffic through the strait Monday: about 10 transits, versus roughly 130 a day before the war.

So the rate relief is real, but read the label. The hike didn't die because inflation was defeated — headline CPI still starts with a 3 and energy is +14.7% year-over-year. It died because the economy cracked first. For landlords, that distinction is the whole story this week, and we'll come back to it twice below.

Say "net lease" and most people picture a pharmacy pad with a drive-thru window. The money has moved. CBRE's Q2 figures, published last week, show total net-lease investment of $12.8 billion, up 13% year-over-year — and $8.1 billion of it, 63%, was industrial, up 28% year-over-year and up from a 56% share a year ago. Retail was $2.9 billion. Office, $1.8 billion and falling.

This week put faces on that statistic. Global Net Lease closed its Modiv Industrial acquisition Wednesday (shareholders approved Monday with roughly 94% of votes cast): a roughly $535 million portfolio at a 7.6% cash cap rate, 15-year weighted lease term, 2.4% annual escalations — taking GNL's industrial exposure to roughly half of straight-line rent. Realty Income's $2.6 billion Q2, reported last week, was about 65% industrial, and its industrial renewals recaptured 105.8% of expiring rent. Broadstone's marquee deal is a roughly $303 million build-to-suit powered shell for a Fortune-20 tenant. Add the data center JV we covered last issue and the pattern is unmistakable: every large check in net lease is chasing big single-tenant industrial boxes.

The why is mechanical. A REIT that must place $10 billion a year cannot do it $3 million at a time; industrial trades in nine-figure chunks. Build-to-suits and sale-leasebacks come with 15-plus-year leases and real escalations, against retail's flat-lease legacy stock. And the retail that is on the market is mostly the wrong retail — Boulder's Q2 data showed investment-grade tenants under 10% of listed retail inventory, while asking caps sit at 6.60% retail versus 7.25% industrial (+10 bps this quarter). The professional bid goes where the spread is.

The caution is residual value, same as ever. A vacated retail pad on a hard corner is the most liquid distressed asset in commercial real estate — the 1031 market will always price the dirt. A vacated 500,000-foot single-tenant box built to one occupier's spec is a project. What this week really tells you is that net lease is splitting into two markets: a retail-pad market owned by 1031 buyers and priced like bonds, and an industrial-and-infrastructure market owned by institutions and priced like credit. The averages you read — 6.82%, 6.9% — increasingly describe neither.

The jobs report has a restaurant-shaped echo.

Wendy's withdrew its 2026 outlook. U.S. same-restaurant sales fell 7% in Q2 — the sixth consecutive quarterly decline — and the company now targets 289–358 U.S. closures in the first half, with the CEO signaling more beyond plan where weak units drag franchisee health. One more wrinkle for landlords: Nelson Peltz is reportedly preparing a bid for the company (one outlet, unconfirmed — watch, don't trade on it). A Wendy's ground lease is still fine paper; a Wendy's ground lease under a leveraged franchisee in a declining system is a different instrument, and the market prices them the same until it doesn't.

Moe's Southwest Grill's largest franchisee, Quality Fresca, filed Chapter 11 Monday. It bought 67 units in March 2020; 38 remain, and the filing seeks to reject at least 16 leases with more rationalization planned. Roughly $52 million of liabilities against $44 million of assets, and about 600 employees. Jack in the Box, for its part, closed 17 more units this quarter on its ongoing cull. Against all this, Brinker posted Chili's 21st consecutive growth quarter — the consumer hasn't stopped eating out; she has stopped eating out at concepts that lost her. Operator selection is tenant selection.

Sunday, August 16: Verizon's 274 corporate stores become franchise-operated, as covered in #006. The credit substitution we described stops being a forecast this weekend.

Not fifty million. Fifty dollars. That is the price Dutch Bros — the coffee chain itself, via its operating LLC — agreed to pay for all fourteen of Salad and Go's shuttered Texas and Oklahoma leases, in the same asset purchase agreement that pays $105.0 million for the 51 operating Arizona and Nevada locations. Same buildings. Same lease paper. Roughly $2.06 million per lease with a business inside; $3.57 per lease without one.

We wrote last week that a 15-year lease on a new building is worth what the operator's business is worth. Here is the market quoting it to the dollar. The Texas boxes aren't worthless as real estate — but the leases are worth nothing to a buyer who'd be assuming rent obligations on dark stores, so they clear at a courtesy price. Landlords holding rejected leases get a claim capped under §502(b)(6) of the bankruptcy code — typically the greater of one year's rent or 15% of remaining rent, not to exceed three years — on a lease they underwrote at 15-plus.

Two live caveats for anyone tracking the docket: the sale is proposed as a private sale with no auction (a $3.8 million breakup fee and $10 million minimum overbid protect it), it had not been approved by the court as of Thursday, and Dutch Bros can exclude any Arizona or Nevada lease before closing, cutting the price about $2.1 million per excluded site. If your pad is on that list, you are short a put you didn't know you sold.

The ecologist Garrett Hardin condensed systems thinking into seven words: "You can never do merely one thing." Every intervention in a connected system produces effects beyond the intended one — and the first question of the intelligent operator is always and then what?

The market spent this week doing first-order arithmetic. Hike odds collapsed; rates rallied; REITs caught a bid. If you own net lease, the temptation is to celebrate: the denominator in every valuation — the discount rate — just got friendlier, and the 241-basis-point spread over Treasuries gets room to breathe.

And then what? Run the causation backward. The hike died because payrolls printed negative and wage growth hit a five-year low. The same consumer weakness that rescued the denominator is attacking the numerator: it is the -7% at Wendy's, the 16 rejected Moe's leases, the fifty-dollar bid for fourteen drive-thrus. A landlord's asset is a fraction — tenant cash flow over a discount rate — and this week the bottom of the fraction improved because the top of it is eroding. That is not a gain; it is a transfer between the two halves of your own position, booked as a win by everyone who only reads one line.

Munger warned about first-conclusion bias: the mind, like a fertilized egg, closes to the second idea once the first arrives. The first conclusion this week is "rates are our friend again." The second — the one worth money — is that you should now underwrite tenants as if the jobs report continues, and structure debt as if the rate relief doesn't. If both go your way, you're pleasantly surprised. If neither does, you already own the only thing that survives both: a durable tenant, on a hard corner, paying contracted rent.

The Texas landlords of Salad and Go did their first-order math too — new building, 15-year lease, healthy growth story. The one thing they assumed could never happen, happened. Ask what your portfolio assumes can never happen — then ask and then what?

The Latticework Letter is published weekly on Friday mornings. It is independent analysis — we have no brokerage relationships, no listings to push, and no financial products to sell. Our only interest is giving you a clearer picture of the market.

Forward this to someone who owns NNN assets and is tired of getting their intelligence from people with something to sell.