When A REIT Dividend Stops Being Income: Test The Engine, Not The Yield

When A REIT Dividend Stops Being Income: Test The Engine, Not The Yield

The version of REIT investing that goes wrong quietly has almost nothing to do with red months on the price chart. It is a dividend that stops being income — a check funded out of your own principal, or by a single tenant who can no longer pay, or by leverage that comes due in exactly the wrong year. Rates and recessions take the blame, but the investor who actually loses to a REIT usually loses to a payout that was never really an income stream in the first place. And this year has handed us a clear view of both the trap and the gift: after nearly a half-decade of underperformance, REITs rebounded through mid-2026 and beat the broad market doing it, even as inflation stayed sticky and the Federal Reserve settled into a higher-for-longer hold. The rebound was real but uneven, and the screeners are, of course, advertising yields up to 22%.

None of that should panic an income investor. It should make us pick up a better ruler. Because the single most common check on a REIT dividend — the payout ratio you find on a stock screen, dividends divided by reported accounting earnings — is close to useless for deciding whether the income is real, in both directions. It can panic you about a payout that is perfectly sound, and it can bless a payout that is quietly dying. Let's test both sides, because that is where the dark side actually lives.

Why the standard REIT screen is the wrong ruler

Start with the false alarm. Realty Income pays monthly, has raised its dividend 24 straight years, and just last quarter produced revenue near $1.5 billion, ahead of Wall Street consensus. It also shows a trailing payout ratio to accounting earnings of about 287%. To anyone raised on ordinary-company dividends, 287% screams "cut incoming." The number is a mirage, and seeing why is the whole lesson. Real estate accounting is distorted by depreciation — a REIT that owns buildings for decades books enormous non-cash depreciation charges, so GAAP earnings permanently understate what the properties actually throw off. The correct ruler is AFFO (adjusted funds from operations), the cash-based earnings measure REIT dividends actually run on. Against AFFO, Realty Income pays out roughly 73% — safe, and even a little conservative — which is why the company can count 670 consecutive monthly dividends and still raise its 2026 AFFO guidance at mid-year. Its reported free cash flow is negative, but that is the machine buying more properties; operating cash flow is over $4 billion a year against roughly $30 billion of net debt. A 5.1% yield today is buying an income stream with a wide cushion, not a teetering one.

Here is the part the screens hide. The same ruler that falsely condemns Realty Income silently waves through the opposite case: a REIT whose payout looks fine on the page while the income underneath is coming apart. The numbers that matter are not in the income statement at all. They are in the tenant list, the balance sheet, and the claim that the dividend is being funded by operations and not by something else.

The engine that actually broke: Medical Properties Trust

Medical Properties Trust is the textbook case, and it hurts because it was once the safe pick. The business was simple and reassuring: MP Trust bought hospitals and leased them back to the operators, collecting rent like any landlord. But it had one of the most concentrated tenant books in REITdom — a handful of operators, and ultimately one tenant, Steward Health Care, carried an outsized share. In May 2024, Steward filed for Chapter 11, pulling 166 affiliated debtors in with it. Medical Properties wrote down its 9.9% stake in Steward from $125.6 million to $35.7 million, absorbed $693 million of impairment charges tied to the tenant in the first quarter alone, and then spent the rest of the year in crisis management: a global settlement, taking control of its real estate back from Steward, and re-leasing to replacement operators across a portfolio of hospitals.

Now watch what happened to the income, because this is the part the yield screen cannot show you. The dividend, which had run at $0.29 a quarter for years, was cut to $0.15 and then sliced to $0.08 within a matter of months in 2024. It has crept back up to $0.09 a quarter by late 2025 — an annual dividend of $0.36 today, which works out to a yield of about 8.7% at the current price. Read that again: an 8.7% yield that any screener will happily list as "high income," sitting on a payout roughly two-thirds smaller than the one that supported the old price. The investor who buys it for the yield is not buying the dividend of 2019; they are buying the much smaller dividend of a company still paying for its worst tenant decision. A concentrated tenant base is the quiet killer in REITs, and no payout ratio on any screen is going to tell you the rent is in danger until the rent is already gone.

The eleven-percent yield that is partly your own money coming back

The mortgage REIT is where the deception is structural, not accidental. Annaly Capital yields about 11.6% and recently hiked its payout again, pushing the annualized run-rate toward 12%. The mechanism is worth spelling out: an mREIT borrows at short-term money-market rates and buys long-dated agency mortgages paying higher coupons, then runs that whole book on roughly two dollars of debt for every dollar of equity. A double-digit mREIT yield is really an interest-rate spread, amplified by leverage. Nothing wrong with that in principle — but it changes what the distribution is. Part of what lands in your account is actual interest income; the rest can be a return of capital, which is not income at all but the company handing back a piece of the principal you already invested. Industry-wide, roughly 12% of REIT dividends are classified as return of capital; in mortgage REITs the share can run much higher.

The number that decides whether that 12% coupon is a gift or a slow withdrawal is book value — the marked-to-market value of the loan book behind each share. Agency mREIT book values have already declined substantially once in this cycle as rates repriced, and I could not get a clean current book-value figure for this piece, which is exactly why I flag the mechanism instead: a mortgage REIT paying double digits while book value erodes is not compounding your income, it is paying you back out of what you already own. That does not make Annaly a bad holding — as a small, understood slice of a portfolio, an 11.6% coupon has a job to do. It makes it a component, not a core, and book value per share — not the yield — is the number to track.

Leverage: the slow-growing dark side

The third dark corner is the balance sheet, and it sneaks up on even the boring net-lease giants. W.P. Carey, a landlord to industrial and retail tenants, yields a modest 5%, carries debt at roughly 102% of equity, and this past year produced about $1.2 billion of operating cash flow while spending over $2.5 billion on new properties — hence negative free cash flow. Like Realty Income, its dividend is covered on the AFFO basis that matters, and the negative free cash flow is growth in progress, not distress. But it is also a machine that only keeps running if the capital markets stay open on reasonable terms. In a higher-for-longer regime, every roll of that machine's financing costs more, and the refinancing window — not the income statement — is where REIT disasters actually get made. The payout can look identical a year before and a year after a dividend cut; what changed was the price at which the debt got rolled.

The mood-break that is actually a bargain: VICI Properties

Now the flip side that makes the whole framework worth having. The market this year has mostly refused to distinguish between the Medical Properties Trusts of the world and the VICI Properties of the world, and that sloppiness is where the income opportunity lives. VICI is the landlord to casinos — it owns the real estate that Caesars, MGM, and their peers operate — and it has fallen about 20% over the past year to a five-year low, near its 52-week low. Yes, the price dropped. But the engine? Its cash-based earnings, measured by AFFO, came in at $0.62 per share in the second quarter, up 4.6%, and management has been raising full-year 2026 guidance while the usual suspects on the tape fixated on a reported per-share earnings number that missed consensus. The headline number startled the chart; the cash that funds the dividend improved.

The payout has now grown seven straight years — about a 7% dividend CAGR over the past eight — and the dividend consumes only about 61% of earnings. So the stock sits at roughly book value — the market is pricing the properties at marked-to-market worth, no more, no less — while paying about 6.6%, against an industry average closer to 3.9%. The bear case is legitimate: a landlord to leveraged casino operators is exposed to tenant credit and to rate duration, and this is a high-class problem to have in a higher-for-longer world. But look at what the price already expresses. If the income engine is intact and the payout is covered at 61%, then a 20% markdown is not a reason to fear the dividend; it is a chance to buy more future income on better terms. That is the reinvestment logic that separates an income investor from someone watching a screen. This one passes the test — and it is the kind of name that belongs in a diversified income machine, not as a conviction bet on the direction of gaming.

Three questions before any REIT dividend, and the portfolio that survives the dark side

So before any REIT yield earns a thumb on the scale, run three questions. First, what funds the payout — cash from operations and AFFO, not accounting earnings, and not debt. Second, who pays the rent and how concentrated is that rent — can the failure of one tenant end the dividend, the way Steward ended Medical Properties'? Third, what does the leverage do at the refinancing window — is it fixed or floating, and what happens to the dividend when the debt is rolled in a higher-for-longer regime?

And then the fourth question that ties it together: did the price fall because the income engine broke, or because the market's mood turned? The whole dark side of REITs collapses into that single distinction. When the engine broke, like it did at Medical Properties Trust, the dividend is a promise that has already cracked, and yield chasing just gets you a lower price later. When only the mood broke, like it did at a 61%-covered VICI at six-and-a-half percent and book value, the lower price is the market doing your reinvestment work for you.

The portfolio is the yield machine, and that is what makes the dark side survivable. No single REIT check should ever be the retirement; spread the income across net-lease, experience, healthcare, and other sub-sectors, keep mortgage REITs as a small, understood slice judged on book value, and treat any double-digit yield as suspect until the coverage and return-of-capital tests pass. On mood-breaks with intact engines, add. On broken engines, do not wait for a price rebound to decide — the income already told you. Timing the Fed is a fool's errand, and I can be fantastically wrong about the harmless part of the calendar; the part I can test is the engine, so that is where the conviction goes. A dividend that lands in your account every month is the only guaranteed return in this game. Everything else is a promise, and a REIT's structure decides whether the promise holds.