Extend and pREITendThe maturity wall in the filings, and the signal that flags who sells next

"Be fearful when others are greedy, and greedy when others are fearful." — Warren Buffett

"You've got to know when to hold, know when to fold up, know when to walk away, know when to run." — Not Warren Buffett

Every CRE acquisition team knows the fantasy: the phone call that lands a few weeks before a distressed REIT announces a disposition, the term sheet that arrives while competitors are still reading the press release.

The problem has never been identifying distress. Debt ratios, falling NOI and stress scores have flagged troubled REITs for decades. The problem is that distress tells you almost nothing about timing. Plenty of stressed REITs hold assets for years, and plenty of healthy ones sell constantly. The question that actually generates deal flow isn't who is in trouble? It's who is selling in the next few weeks below market?

Those are different questions, and they need different evidence. This piece walks through both halves: what the public filings already tell you about the pressure building on REIT balance sheets (every number below is live and you can push it around yourself), and then what it takes to turn that standing pressure into a timing signal.

The pressure is on the public record

Every US equity REIT states exactly when its debt falls due. It sits in one footnote table of the 10‑K, shaped a little differently in every filing. Read all of them and you get the sector's maturity ladder: the clearest standing measure of who has to do something, and roughly when.

What that costs depends entirely on the rate it gets refinanced at. The exhibit below reprices it: choose a rate and a window, and every maturing dollar moves from the issuer's in-place rate to yours. Nothing else is assumed: no asset sales, no equity raise, no change in the debt balance.

Two things to read it correctly. The figure is a fully-repriced annual run rate, not next year's income statement: debt maturing in year three only starts costing more in year three. And the in-place rate is each issuer's whole-book average, interest expense over total debt, because the coupon on the specific tranche that is maturing is not something filings disclose. Since maturing debt is usually the older, cheaper debt, that makes every increase here a conservative one.

A decade of pushing it out, and the year it stopped

A maturity ladder is only interesting against its own history. Rebuild it from the 10‑K of every year since 2013, holding the cohort fixed at the 32 REITs that filed a tagged ladder in all thirteen years so the trend isn't just composition, and the shape of the last cycle is unmistakable: refinancing risk was pushed steadily further out through the cheap-money years, bottoming in 2017, and has climbed every year since 2021.

Pressure alone still isn't a name on a call list, though. For that you have to put it against what each REIT is actually doing.

Who has to act, and who already is

Two numbers, both straight out of the filings, put every REIT in one of four positions. Up the side: how much of its dated debt falls due inside three years, the pressure to do something. Along the bottom: how much it actually sold for, as a share of its own balance sheet. Split each at the panel median and you get the four readings that matter to an acquisitions desk.

The interesting quadrant is not the busy one. A REIT selling hard from a comfortable maturity position is recycling capital, and it will run a competitive process. A REIT carrying a wall it has not started addressing is the one worth calling.

And this is where the public record runs out. It tells you the pressure is real, rising and unevenly distributed, and it tells you who has and hasn't moved. It cannot tell you what any one of these REITs will do next month, which is the only thing an acquisitions desk is actually paid to know.

Distress is a level. Selling is an event.

Everything in Part one is a level. It describes standing pressure, and standing pressure is not a date. Traditional REIT tracking has the same shape: quarterly filings tell you what happened three months ago, stress scores measure past performance, and almost none of it is built to answer when. The question that matters is the when: not how distressed is this REIT but what is the probability this REIT sells an asset in the next thirty days.

That is a forecasting problem, and it rests on two observations.

The market moves first

The equity market is a prediction machine. Investors sell REIT shares before the REIT announces asset sales, pricing in the earnings misses, dividend pressure and covenant strain management hasn't disclosed yet. A stock falling hard against the index is the market saying something the filings won't say for another quarter, and relative stock performance leads disposition announcements by weeks.

Selling comes in waves

REITs in disposition mode sell in waves, not one-offs: a sale in the last thirty days raises the odds of another in the next thirty. On its own that is ambiguous, because healthy REITs transact constantly too, which is exactly what Figure 3 shows. Crossed with stock underperformance it stops being ambiguous, and that interaction is what separates a fire sale from routine portfolio rotation.

The model

A panel of 87 public REITs observed fortnightly through time, each observation labelled with whether that REIT sold anything in the following thirty days, and a gradient boosting classifier trained on transaction-behaviour and stock-performance features. It scores every REIT in the universe and outputs one number: the probability of a sale inside the window.

A 1.9× lift means the top-ranked decile of observations sold at nearly twice the base rate: a ranked outreach queue, refreshed weekly, with the features that flagged each name attached.

One name per sector

Put the two halves together and you can already draft the first call sheet. Within each sector, take the REIT with the heaviest near-term wall among those that have also written assets down: the filings' own two-signal proxy for the seller most likely to accept a discount in the next few months. Two sectors put no name forward under that rule, which is a finding in itself.

If one of these names sits inside your mandate, give us a call. We run the timing model across the full REIT universe every week, and we can show you which of that REIT's properties best fits your demonstrated investment thesis, before the process starts.

Data through FY2025 10‑K filings. Source: SEC XBRL companyfacts API (data.sec.gov). 92 REIT filers pulled; 56 equity REITs with a machine-readable ladder in their latest 10‑K, in-place rate for 50 of them. Median in-place rate 4.26%. Built 4 August 2026. Figures are as-filed and not adjusted for events after each balance-sheet date. Transaction data underlying the timing signal courtesy of MSCI Real Capital Analytics. Header photograph: the Rosslyn skyline from Hains Point, by Wei Feng on Unsplash.