Armour (ARR) Q2 2025 Earnings Call Transcript

ARMOUR Residential REIT reported a Q2 GAAP net loss of $78.6 million, or $0.94 per share, offset by $33.1 million of net interest income and $64.9 million of distributable earnings, or $0.77 per share. Book value was $16.90 per share at quarter-end, while liquidity remained strong at 52% of capital and leverage was maintained at 8 turns with a modestly bullish stance on Agency MBS. Management said spreads remain historically attractive, dividends of $0.24 per month are sustainable in the medium term, and the company has raised $104.6 million via ATM issuance in the quarter plus $58.8 million after June 30.

Analysis

ARR is effectively telling you it has turned into a levered carry vehicle again, but with the important caveat that the equity issuance pipeline is still doing part of the funding work. That creates a two-step setup: near term, book value can look resilient because dilution is being partially offset by fee waivers and stable spreads; medium term, if spreads tighten into the expected rate-cut window, the company can redeploy capital at higher ROE and the same equity base should compound faster. The market is likely still underappreciating how much this is a function of funding capacity, not just asset selection.

The more interesting second-order effect is on relative value inside agency MBS itself. Management is effectively signaling a preference for higher coupon specified pools and production paper where prepay convexity is better; that is a negative for generic TBA collateral if the trade becomes crowded and a modest positive for specified-pool lenders and securitization flows that can source better collateral. If bank demand does reappear later this year, the first-order beneficiary is not just ARR's portfolio marks — it is the whole agency complex via spread compression, with the cleaner high-carry names outperforming lower-quality mortgage REITs that are more exposed to book volatility and funding creep.

The key risk is timing: the thesis depends on the Fed easing cycle arriving before mortgage supply normalizes or volatility re-expands. If cuts are delayed into late 2025 or long-end rates back up, ARR's leverage may have to stay capped and the incremental ROE from new capital falls short of the dividend narrative. The contrarian read is that management's comfort with raising leverage is less a bullish macro call than a signal that real-money alternatives are scarce; in other words, the hurdle to look good is low, but the path to materially higher returns still needs a catalyst outside ARR's control.

Key Decisions for Investors

• Long ARR on pullbacks near book value, with a 3-6 month horizon: upside comes from spread compression plus incremental leverage, but size modestly because the thesis is rate-timing dependent and dilution remains a background drag.
• Pair trade: long ARR / short an unhedged mortgage REIT with higher book-vol sensitivity and weaker liquidity posture over the next 1-2 quarters, targeting relative outperformance if volatility stays contained and agency spreads tighten.
• For a cleaner rates expression, buy call spreads on a broad agency MBS proxy or mortgage REIT basket into the next Fed meeting cycle; the risk/reward improves if the market begins pricing cuts before mortgage rates actually fall.
• Avoid chasing generic TBA exposure versus specified-pool beneficiaries: prefer names or instruments tied to higher coupon, better-convexity collateral if you want to express the benign prepay view over the next 6 months.
• Set a risk trigger: if 10-year yields back up and mortgage volatility reaccelerates, reduce ARR exposure quickly rather than averaging down; the model works only while liquidity stays stable.