CoStar Says Office Vacancy Is Falling. The Office REIT Dividends That Remind You Why.

If you follow headlines, the U.S. office story is getting better. CoStar says vacancy has ticked down since its mid-2025 peak, there have been four consecutive quarters of rising occupancy, and the vacancy rate is projected to gradually decline to around 13% by 2030. The narrative sounds like a recovery. The dividend story from the companies that own these buildings tells a different one.

The vacancy improvement is not what the headline implies. CoStar calls it an "unprecedented period of shrinking supply" — buildings are being demolished, not leased. New office construction has become rare. Demand growth is described in the same report as "lukewarm." Four quarters of occupancy gains happen when the denominator is getting smaller, not when tenants are flooding back. CoStar's own February update said vacancy would plateau at 14.1% through 2026. Rent growth is expected to remain below 1% this year. That is the definition of a market where cash flow is not improving, even if the vacancy number looks a shade better.

Now look at what the three largest public office REITs are actually paying and how they fund it.

Boston Properties (BXP), the largest publicly traded office landlord, slashed its quarterly dividend by 29% in September 2025 as part of a "strategic reset." The current quarterly dividend is $0.70 per share, which annualizes to a forward yield of roughly 5.7%. The payout ratio on that dividend sits at 203% — meaning the company pays out more than twice what it earns in free cash flow. Total debt is $17.4 billion against $7.7 billion of equity, for a debt-to-equity ratio of 202%. Free cash flow has declined 16% year over year to $275 million. The dividend is not funded by the business. It's funded by balance sheet capacity that was already stretched before vacancy peaked.

Vornado (VNO) has already done the same arithmetic. Its quarterly dividend was cut to $0.3375 per share, yielding about 1.9% forward — far less compelling than the headline numbers you see on some summary pages that still reference the pre-cut payout. Free cash flow is negative $416 million over the trailing twelve months. Operating cash flow is only $391 million, barely covering a dividend base that was already reduced. Total debt of $8.9 billion carries a debt-to-equity ratio of 111%. The company that once carried New York's trophy office portfolio has quietly surrendered most of its yield because the rents no longer support it.

SL Green (SLG), Manhattan's largest office landlord, announced an annual dividend of $2.47 per share in March 2026, down from $3.01 the prior year. That's a cut of roughly 18%. The forward yield from that reduced payout is 1.9%. Free cash flow is negative $980 million. The company is spending over $1 billion in capex while pulling in only $92 million in operating cash flow. Debt-to-equity sits at 98%. SL Green reported a net loss of $0.38 per share in the second quarter of 2026. The dividend is being maintained at a level the company can barely afford, and its own guidance suggests the cash-flow gap is not closing fast.

The pattern across all three is the same: the vacancy headline is stabilizing on paper because buildings are disappearing, but the cash that flows from the buildings that remain is not enough to support the old dividend levels. Each of these companies has already cut. BXP's payout ratio at the new level still exceeds 200%, which tells you the cut wasn't enough to make the dividend safe — it was enough to buy time while management looks for exits, repositioning, or further reductions.

Here's the question for the income investor that the vacancy chart doesn't answer: if rent growth is below 1% and vacancy is structurally above 13.5% — a full percentage point higher than the post-Great Recession peak, by CoStar's own words — what is supposed to make the cash flow better? The answer from these companies is usually the same: dispositions, repositioning, patience. None of those things produce dividend coverage in the near term.

Some readers see the 5.7% forward yield on BXP and think of volatility as a reinvestment opportunity. This is where you have to be disciplined about the difference between price pain and business pain. A lower price is a reinvestment opportunity when the income engine is intact. Here, the income engine is structurally impaired. Rent growth is negligible. Demand is tepid. Leverage is high. Three of the largest players have already cut dividends because the math stopped working. That is not a temporary market mood shift. That is credit deterioration.

CoStar's August 2026 note did flag one positive development: several occupiers have committed to additional space in anticipation of higher office attendance. That helped produce four quarters of occupancy gains. But CoStar simultaneously warned that hiring in traditional knowledge industries continues to lag, per-employee space needs are structurally lower than pre-2020 levels, and AI-driven productivity gains could enable firms to expand output with fewer employees — dampening space demand further. The near-term occupancy gains and the long-term demand destruction are pulling in opposite directions. The net result is the plateau at 14.1% that CoStar itself projects through the end of 2026.

So what does the income investor do? The honest answer is that these names don't belong in a portfolio built to fund a retirement through cash flow. The yields are either already cut to unattractive levels (VNO and SLG at roughly 1.9%) or remain elevated only because the payout exceeds what the business generates (BXP at 5.7% on a 203% payout ratio). Neither scenario is a sustainable income source.

If you already hold one of these, the decision depends on whether you're collecting the dividend while you can or whether you're waiting for a recovery that the fundamentals don't yet support. If you're looking to allocate income capital, the office sector is where you look last. The vacancy decline is real. It just isn't the kind of decline that makes a dividend safer.

The real story in office right now is that the market is working exactly as it should: vacancy adjusts through supply destruction when demand won't return. The problem is that the REIT dividends were set when vacancy was lower and rent growth was higher. The dividend cuts that have already happened are not mistakes. They're the cash-flow engine catching up to the reality of the buildings. The ones that haven't happened yet are the risk.