ORC — Q3 2026 Financial Report Analysis

Book value fell 17% in one quarter as a bond sell-off hit the portfolio

Orchid Island Capital lost $178.9 million, or $0.90 per share, in the third quarter of 2026 (July to September), against a profit of $72.1 million ($0.53) a year earlier. The cause was not the day-to-day business: Orchid's net interest income (the interest it earns on its mortgage bonds minus what it pays to borrow) more than doubled to $59.1 million. The loss came from a sharp rise in US interest rates, which knocked $567.1 million off the market value of its bond portfolio. Gains of $335.4 million on its hedges (positions that profit when rates rise) covered only about 59% of that.

For a company like this, the figure that matters most is book value per share: what shareholders' equity is worth per share if every bond is priced at today's market value. It fell to $6.02 at September 30, from $7.22 at June 30 and $7.33 a year earlier.

How the business works, briefly

Orchid is a mortgage real estate investment trust (mREIT). It buys agency residential mortgage-backed securities (agency RMBS): bonds made of bundles of home loans, where Fannie Mae or Freddie Mac guarantees the principal, so the risk is not borrowers defaulting. The risk is interest rates. Orchid finances these bonds with repurchase agreements ("repo"), which are very short-term loans secured by the bonds themselves. At September 30 the average repo had 26 days left to run. It owned $10.7 billion of bonds against $1.17 billion of shareholders' equity.

Two things follow from that structure:

1. Profit comes from the spread. That is the gap between what the bonds pay (5.91% on average in Q3) and what the borrowing costs (3.92%).
2. Book value swings with bond prices. Orchid carries its bonds at market value and runs every price change through the income statement. When long-term rates rise, its fixed-rate mortgage bonds fall in value, and with roughly nine dollars of bonds for every dollar of equity, a small percentage move in bond prices becomes a large move in book value.

The headline "revenue" shown for this report ($164.1 million) is GAAP interest income, the closest thing this business has to sales. Total interest expense was $105.0 million.

Key figures

Book value per share is computed from the balance sheet: equity divided by shares outstanding ($1,171.8 million / 194.7 million shares; $1,086.1 million / 148.2 million a year earlier). The company does not state the figure in the 10-Q.

What happened in the quarter

Rates jumped to levels last seen in the early 2000s. The 10-year Treasury yield rose from 4.47% at June 30 to 5.29% at September 30, and the 2-year from 4.18% to 4.89%. Management names the drivers as supply shocks from the Iran war, the Bank of Japan buying fewer US Treasuries as it normalises policy, and widespread deleveraging and forced selling by leveraged bond investors. The Federal Reserve also raised its policy rate by 0.25 points in September, to a 3.75%–4.00% range. The 30-year fixed mortgage rate in Freddie Mac's survey rose from 6.43% to 7.28%.

Mortgage bonds did worse than plain Treasuries. The filing says the agency RMBS sector returned -4.1% in Q3 and -0.8% relative to interest-rate swaps of the same duration. Duration is a measure of how much a bond's price moves when rates move. Conventional 30-year mortgages, which make up most of Orchid's holdings, returned -4.4%. Management's explanation for why mortgages lagged: as rates rose, homeowners became much less likely to refinance. That stretched out the expected life of the bonds just as rates went up, so their prices fell further. By quarter-end "nearly all Agency RMBS are trading at a discount to par." The portfolio's effective duration rose to 4.33 at September 30 from 2.51 at December 31, 2025. In plain terms, a 1-point rise in rates would now cost the bonds about 4.3% of their value, before hedges.

The hedges worked, but only partly. Orchid hedges with short positions in Treasury futures and borrowed Treasuries, plus swaps, swaptions and other derivatives. Those gained $335.4 million in Q3, against $567.1 million of unrealized losses on the bonds. The remaining shortfall of $231.7 million is the mortgage-specific underperformance described above. It is what the filing calls "spread risk": the bonds fall by more than the hedges tied to benchmark rates gain.

The income side kept improving. Net interest income rose because:

• The portfolio is much bigger. Average bond holdings were $11.10 billion, against $7.67 billion a year earlier (+44.7%), funded largely by share sales in 2025.
• Asset yields rose and borrowing got cheaper. The average yield rose to 5.91% from 5.65%, while the GAAP cost of funds fell to 3.92% from 4.45%.

On management's "economic" basis, which counts the current-period effect of the hedges as part of interest expense, cost of funds rose slightly to 3.43% from 3.25%. Economic net interest income was still $72.1 million, up 47.9%. Prepayments slowed to an 8.4% annualised rate, from 10.9% in Q2 and 10.1% a year earlier. For bonds bought below face value, slower prepayments mean the discount is earned back more slowly.

What the headline numbers hide

• The loss is mostly paper, but it is real equity. Net earnings excluding realized and unrealized gains and losses were $52.8 million ($0.27 per share). Everything else in the result, $231.7 million, was price moves on bonds and hedges. Nothing was realized from bond sales in Q3, so no loss was locked in by selling. But Orchid marks its bonds to market, so the drop is already in book value, and book value is what the leverage ratio and repo lenders' margin calls run on.
• The dividend is not covered by the cleaner earnings measure. Orchid paid $0.30 per share in Q3 ($0.10 a month), while earnings excluding gains and losses were $0.27. The gap was similar in Q1 and Q2 ($0.26 each against $0.30 or more paid). The monthly dividend was already cut from $0.12 to $0.10 starting with the April declaration, and October's declaration stayed at $0.10. For the first nine months, operating cash flow of $195.3 million did cover the $188.7 million of dividends declared.
• Most of the hedge gain belongs to future periods. Of the $335.4 million derivative gain, management attributes $303.7 million to future periods. It is the market value of funding costs locked in ahead, not cash earned this quarter. Only $13.0 million was attributed to Q3's own funding cost.
• Per-share growth trails total growth. Net interest income rose 120%, but the weighted share count rose 45.5% (to 198.5 million) after the 2025 stock sales. Earnings excluding gains and losses rose from $0.16 to $0.27 per share, not by the full 146%.
• The buyback is now running, below last quarter's book value. Orchid repurchased 4.95 million shares in Q3 at an average of $6.30, under an authorization the board raised by 25 million shares in June. After quarter-end it bought another 1.67 million shares at an average $5.05. That is below the $6.02 book value at September 30, so those later purchases add to book value per share for remaining holders, if book value held up into October. The 10-Q does not give an October book value.
• Liquidity looks adequate. At September 30 Orchid had $462.4 million of unrestricted cash and about $53.3 million of unpledged securities. Lenders' haircuts (the extra collateral they demand) were stable at about 4.1%. The filing describes repo funding markets as "otherwise stable" despite the rate move.
• No accounting one-offs to strip out. Realized gains on bond sales were zero in Q3. Operating expenses rose to $6.3 million from $5.4 million, mostly from higher management fees ($4.1 million against $3.3 million) on a bigger equity base, and remain small next to the portfolio.

Takeaway: Orchid's income engine is the strongest it has been in this filing's history: the GAAP spread was 1.99%, against 1.12%–1.43% through all of 2025. But for a leveraged agency mREIT, book value is the scorecard, and it fell 17% in a quarter in which hedges covered only about three-fifths of the bond losses. Leverage rose to 8.6x without any new borrowing. A strong carry of $0.27 a share is no help if rates keep rising and mortgage spreads keep widening.

Outlook

Management gives no earnings or book-value guidance. Its own market read is cautious. It notes that rates kept rising "during the third quarter and early fourth quarter of 2026". Fed Funds futures priced about 3.8 more quarter-point hikes over the next year as of September 30. Management also flags new stress in French government debt after quarter-end. The MOVE index, a gauge of expected bond-market volatility, rose from 71.96 to 110.45 during Q3. High volatility usually works against mortgage bonds, because it raises the value of the homeowner's option to refinance and the cost of hedging.

The company's own sensitivity table at September 30 estimates that an instant further 1-point rise in rates would reduce equity by about 5.6%, and a 2-point rise by about 18.4%, including hedges. It estimates that a 1-point fall would also reduce equity, by about 3.4%. That is the "negatively convex" profile the filing describes, where losses are possible whichever way rates move. That model assumes mortgage spreads do not change, and spreads widening was exactly what hurt in Q3.

Our read on what to watch:

• Book value per share in the Q4 release. The post-quarter buyback at $5.05 suggests the stock traded well below the September 30 book value in October. Whether that reflects a further decline in book value is not stated.
• Whether the $0.10 monthly dividend holds. Earnings excluding gains and losses have run at $0.26–$0.27 a quarter against $0.30 paid.
• Leverage. At 8.6x, another quarter like this one would push it higher unless Orchid sells bonds into a weak market or leaves borrowing unchanged while the equity base shrinks.
• Funding costs if the Fed keeps hiking. Repo costs reset within weeks, while most of the bond coupons are fixed (5.51% weighted average).

The full-year 2026 annual report (10-K) is the next filing; Orchid's fourth-quarter results usually come out in late January.