Is ABR a Buy or a Sell? The Bull and Bear Case (2026)

That depends on which case you find more convincing, and both are on this page. The bull case rests on Working down the legacy bridge book, with distributable eps (h1 2026) at ~$0.17. The bear case rests on credit is the central risk. If you believe the thesis, the real questions become sizing and overlap rather than timing. Walnut is not an investment adviser and this is not a recommendation.

Nobody can answer that for you, and the honest version of the question is narrower: has anything changed in the reason you bought it? The bear case on this page is the place to check. Credit is the central risk. If that risk is what you were worried about and it is now playing out, that is a real signal. If the price simply fell while the thesis held, that is a different situation entirely. Walnut is not an investment adviser.

Working down the legacy bridge book. The biggest driver of the stock is how quickly, and at what loss, Arbor resolves loans written at the 2021 to 2022 peak.

Credit is the central risk. Arbor had ~$429 million of non-performing loans, ~$95 million of other non-accrual loans and ~$387 million of modified loans at June 30, 2026. It also held ~$546 million of real estate owned that it has to operate or sell. Values for the Sunbelt apartments behind much of the bridge book have suffered from heavy new supply and higher insurance and operating costs, and more impairments would push book value below ~$10.95. Leverage magnifies every loss: ~$11.5 billion of liabilities against ~$2.9 billion of equity, much of it in repo and credit facilities whose lenders can make margin calls. With the stock near a quarter of book, issuing equity would be highly dilutive. The dividend could be cut or suspended if Q3 earnings land near the ~$0.05 consensus. Removal from the S&P SmallCap 600 on October 27 shrinks the pool of index holders. Governance questions also linger. A 2023 to 2024 short-seller campaign alleged hidden delinquencies, and Bloomberg reported a federal inquiry in 2024. The class action that followed was dismissed with prejudice in May 2026, but a derivative suit is still pending.

Mortgage REIT that makes multifamily bridge loans and services a large Fannie Mae and Freddie Mac agency loan book.

Decide that in advance, because it is far harder to think clearly once a position is moving against you. The concrete tripwires here: the driver behind the bull case (Working down the legacy bridge book) stalling in the reported numbers rather than in the narrative, the risk above (credit is the central risk) turning from a possibility into a reported fact, or the position growing large enough that a bad outcome would matter to your whole portfolio regardless of who is right.

Arbor is a mortgage REIT. It lends against real estate rather than owning buildings, and it has two segments. The structured segment makes short-term, floating-rate bridge loans, ~$12.1 billion as of June 2026. Most go to apartment owners repositioning properties and to investors building single-family rental portfolios, with smaller amounts in mezzanine, preferred equity and construction loans. The agency segment originates permanent multifamily mortgages for Fannie Mae, Freddie Mac and other government programs. Arbor keeps the servicing rights and earns an annual fee on a ~$36.7 billion portfolio. The two segments feed each other, because bridge borrowers often refinance into agency loans once their properties stabilize. As a REIT, Arbor must pay out most of its taxable income as dividends, which is why it has historically carried a high yield.

No single company announcement explains it. The drop came from a series of hits to confidence in Arbor's earnings and book value. Citizens downgraded the stock on September 29. On October 9, Piper Sandler cut its target to $3.50 with an underweight rating, and Keefe, Bruyette & Woods cut to $3.00 with underperform. Consensus estimates for Q3 and Q4 distributable earnings fell to roughly $0.05 a share, about a third of management's $0.15 to $0.17 guidance and well below the $0.17 dividend, which raised the odds of another cut. On October 22, S&P said Arbor would leave the S&P SmallCap 600 on October 27, which forces index-tracking funds to sell. The stock went from ~$5.25 at the end of August to a low near $2.53 on October 19 and closed ~$2.92 on October 23.

The market is acting as if it is. Arbor paid $0.43 a quarter through early 2025, then $0.30, then cut to $0.17 in July 2026. At ~$2.92 a share, $0.17 a quarter works out to a yield above 20%. Yields that high usually mean investors expect another reduction. Distributable earnings were only $0.17 a share for the whole first half of 2026, and analysts now expect about $0.05 a quarter for the rest of the year. Management had said earnings would run near $0.15 to $0.17 as legacy assets are resolved. The common dividend is normally declared alongside quarterly results, so the Q3 report on October 30 is the next decision point. Arbor's three series of preferred stock rank ahead of the common, and their October 30 dividends were declared on October 5.