The Latticework Letter — Issue #011

Last issue closed with a blank piece of paper. It has writing on it now, and the writing is hawkish.

Kevin Warsh's first Jackson Hole keynote Friday morning did four things the market hadn't fully priced: it named PCE as the gauge he will act on, it owned the Fed's inflation record outright, it declared financial conditions "not restrictive," and it consolidated his no-forward-guidance practice into something like doctrine. The line doing the work: "We must be confident that underlying inflation is moving to our objective, clearly and at sufficient speed. Otherwise, we have work to do." Stocks shrugged. The bond market did not — it started pricing a hike.

The gauge he named promptly cooperated with the hawks. July PCE printed 3.7% headline — slightly above the 3.6% forecast — with core at 3.3%, in line. Still miles from 2%, and drifting the wrong way from the Fed's perspective. The result: CME FedWatch odds of a September 16 hike, roughly 31% when Issue #010 went out, jumped to 66.1% by Monday — nearly double the pre-speech level. The 10-Year, which the Hormuz de-escalation had walked down to the mid-4.60s, sold off for five straight sessions, pushing above 4.81% Wednesday — its highest level since late 2023 — before settling at 4.79% Tuesday and pulling back to 4.77% by Thursday’s close, when Governor Waller said he's leaning toward holding rates steady this month absent an inflation surprise. The Dow rose 624 points on the relief. One governor's caveat is now worth six hundred Dow points; that tells you how tightly wound the September 16 question is.

Meanwhile the labor data spent the week arguing the other side. ADP said private employers added just 38,000 jobs in August — the fewest since January — against a 47,000 forecast, with manufacturing shedding 17,000. JOLTS showed openings flat at 7.3 million. ISM manufacturing came in at 54.6, an eighth straight month of expansion but a full point below July — and its prices index held at an uncomfortable 71.1, a twenty-third consecutive month of rising input costs. Recall that July payrolls fell 23,000; consensus for Friday's August print sits near 58,000–65,000. So the setup into the FOMC is genuinely two-handed: sticky 3.7% inflation and a hawkish chairman on one side, a visibly cooling labor market on the other, and a funds rate at 3.50%–3.75% that Warsh just told you isn't restrictive. For your discount rate, the last two issues said supply worries own the long end. Add a second owner: a Fed chairman who thinks PCE at 3.7% is his problem to fix.

Net lease's original bond-substitute was a drugstore. Twenty-five-year absolute-NNN leases, investment-grade credit, hard corners — for two decades, Walgreens and CVS paper was what 1031 buyers bought when they wanted to stop thinking. The sector now offers the widest same-box credit spread in net lease, and this year's ownership news explains why.

Start with what changed. Walgreens has been private for just over a year — Sycamore Partners closed its take-private in August 2025 — and the rating agencies withdrew coverage on the way out the door. The October 2024 plan to close 1,200 stores over three years has been quietly shelved by the new owners: reporting this summer indicates Walgreens now expects to close fewer than 100 stores in 2026, down from internal projections of roughly 700, prioritizing cash-flow-negative locations, owned sites it can sell, and leases already rolling. Rite Aid, meanwhile, is simply gone — its second Chapter 11 finished the job, and its properties have been exiting the net-lease inventory entirely. CVS is the last investment-grade credit standing in the category.

The pricing tells you exactly what the market thinks of all this. CVS trades around a 6.44% average cap rate. Walgreens paper averaged roughly 7.81% at year-end 2025, with long-dated (10+ year) stores near 6.64% and short-term paper quoted anywhere from 8.6% to 9.5% — consistent with Boulder's Q1 tenant profiles, which put short-term Walgreens at 9.00% or above. The cleanest comp: in Q1 2026, a CVS with 12 years of term traded at 6.50% while a Walgreens with 12 years traded at 8.01%. Same use, same lease shape, same remaining term — 151 basis points of spread, all of it credit and sponsorship. That is the price of a withdrawn rating and a private-equity owner.

Here's the two-sided read. The bull case for Walgreens paper is that the closure pullback is real information: Sycamore's plan evidently needs the store base, the worst locations were culled in fiscal 2025, and an 8% cap rate with a tenant that has stopped shrinking is one of the few places in net lease where you're paid a genuine risk premium. The bear case is that a PE owner's mercy is a policy, not a covenant — the option to close is permanently alive, decided store by store on cash flow, and you will not get a press release before your corner makes the list. The discipline the sector now demands is store-level: pharmacy script volume, the tenant's sales if you can get them, and what the box is worth dark — because the days of underwriting the drugstore sector on the logo are over. The market has already stopped. A 151-basis-point same-term spread says "drug store" is no longer one asset class. Neither was "dollar store," as we saw last week.

Salad and Go produced the ending the last three issues were building toward — and it's a good one for landlords. At auction, 7 Brew won 73 former Salad and Go sites for roughly $143.2 million, beating the Dutch Bros stalking-horse deal ($105 million for 65 locations — the 51 Arizona/Nevada restaurants plus 14 Texas/Oklahoma leases). Dutch Bros elected not to submit a topping bid and walks away with a $3.8 million termination fee plus expenses if the deal closes. The winning portfolio: 41 sites in Arizona, 20 in Texas, six each in Nevada and Oklahoma. A judge still has to approve it at a September 21 hearing. Remember where this started: a debtor trying to hand its whole footprint to one buyer in an exclusive deal, and a rival arguing the estate was leaving money on the table. The rival was right — the auction added roughly $38 million, a 36% improvement over the stalking horse. Process design decided recoveries, exactly as advertised. More on the price itself below.

Wendy's entered September with the U.S. President role officially dissolved — Pete Suerken's departure took effect August 31, with the new COO seat reporting directly to the CEO still being built out. Nothing new on the Trian front since the "no plans to bid right now" reporting; the stock remains in single digits, the 2026 outlook remains withdrawn, and management continues to signal that more restaurant closures are possible as the turnaround plan firms up. The watch item is unchanged: closure lists, not takeover rumors.

Campbell's — not a net-lease tenant, but Thursday's print is a consumer signal worth thirty seconds: revenue missed at $2.1 billion, snacks weak, guidance below estimates, and the dividend cut by more than a third. Another data point for the bifurcated consumer thesis — staples pricing power is thinning at exactly the moment the dollar stores are comping 3%+.

On the calendar: America's Car-Mart's covenant waiver expires September 7 — Labor Day Monday — extendable to September 21 or November 6 only if milestones are met. No new 8-K as of Thursday; watch for a filing early next week. This remains the nearest-dated credit event in net lease. Note the odd symmetry of September 21: it is both Car-Mart's first extension date and Salad and Go's sale-approval hearing. And September 16 is the FOMC.

That is what 7 Brew is paying per site for the core of the Salad and Go portfolio — roughly $125 million for 49 locations. The same auction, the same week, priced the other 24 sites at about $18 million, or $750,000 each. Same debtor, same month, same asset class, same buyer: a 3.4x spread between the boxes 7 Brew had to have and the ones it merely took.

Sit with what $2.55 million buys: a used, purpose-built drive-thru salad kitchen of perhaps 750 square feet, on a leasehold, being purchased to sell coffee. Issue #009 made the case that drive-thru coffee pads are the most expensive dirt in net lease — sub-5.25% cap rates on the sector's thinnest buildings — and this is that thesis showing up in a bankruptcy court, where a coffee chain in a land war paid new-construction money for secondhand corners because the corners, not the kitchens, were the asset. For landlords, the healthy way to read it: your residual value in a drive-thru box is the location's option value to the next concept, and this auction just printed a market-clearing price for that option — at two very different levels. Which is the real lesson. The $2.55 million sites and the $750,000 sites looked identical in the debtor's store list. The distribution — last issue's whole sermon — was hiding inside the portfolio, and it took a competitive auction to reveal it. No appraisal would have found that spread. Only a market did.

Warren Buffett's advice about auctions, which Munger endorsed with his usual economy, was two words: don't go. This week offered the rare spectacle of a company taking that advice mid-auction — and getting paid for it.

Dutch Bros had the stalking-horse position, the inside track, and a growth story that made 73 new corners look strategic. When the bidding passed its number, it stopped. No topping bid, no "just one more round." It leaves with a $3.8 million fee for its trouble, its balance sheet intact, and its 2027 pipeline unburdened by 73 leases priced by an adversary's enthusiasm. 7 Brew leaves with the corners — and with the obligation to make $2.55 million boxes work at drive-thru coffee volumes, forever measured against the price a rival paid to be right.

The mental model here is the winner's curse, and it has a precise mechanism worth internalizing: in any auction for an asset of uncertain value, the winner is, by construction, the bidder with the most optimistic estimate. You don't win because you were right; you win because you were the furthest from the consensus. Which means the moment you win, the auction itself has handed you bad news — everyone else, looking at the same asset, thought it was worth less. The only defenses are a valuation set before the bidding starts and the willingness to look foolish when someone sails past it. That's not timidity. It's the recognition that in bidding, unlike in operating, the discipline is the skill.

It applies well beyond bankruptcy court. The 1031 buyer with a 45-day identification clock is a forced bidder in a permanent auction — the exchange deadline is the enthusiasm, and every broker knows it. The lender stretching to win a deal at 95% of its model is bidding away its margin of safety. Even Warsh's refusal to give forward guidance is the same shape: a man declining to commit to a bid before he has to. The question to carry into any competitive process — an exchange, a marketed deal, a lease negotiation — is the one Dutch Bros evidently asked: what number would I pay if no one else were bidding, and what am I paying for the fact that they are? The first number is the asset. The second is the auction. Only one of them shows up in your rent checks.

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.