Armour 住宅房產信托 (ARR) 2025 Q4 法說會逐字稿

Good day. Welcome to the ARMOUR Residential REIT's fourth-quarter 2025 earnings conference call. (Operator Instructions) Please note this event is being recorded. I would now like to turn the conference over to Scott Ulm, CEO. Please go ahead.

Good morning, and welcome to ARMOUR Residential REIT's fourth-quarter 2025 conference call. This morning, I'm joined by our Chief Financial Officer, Gordon Harper; as well as our Co-Chief Investment Officer, Sergey Losyev and Desmond Macauley.

I'll now turn the call over to Gordon to run through the financial results.

Thank you, Scott. By now, everyone has access to ARMOUR's earnings release, which can be found on ARMOUR's website, www.armourreit.com. This conference call includes forward-looking statements, which are intended to be subject to the safe harbor protection provided by the Private Securities Litigation Reform Act of 1995.

The Risk Factors section of ARMOUR's periodic reports filed with the Securities and Exchange Commission, described certain factors beyond ARMOUR's control that could cause actual results to differ materially from those expressed in or implied by these forward-looking statements. Those periodic filings can be found on the SEC's website at www.sec.gov.

All in today's forward-looking statements are subject to change without notice. We disclaim any obligation to update them unless required by law. Also, today's discussion refers to certain non-GAAP measures. These measures are reconciled with comparable GAAP measures in our earnings release. An online replay of this conference call will be available on ARMOUR's website shortly and will continue for 1 year.

Q4 was a strong quarter for ARMOUR with a total economic return of 10.63% for the quarter as we benefited from MBS spreads tightening, lower MBS volatility and a lower interest rate environment. The market momentum we saw in Q4 has continued so far into Q1. ARMOUR's Q4 GAAP net income available to common stockholders was $208.7 million or $1.86 per share. Net interest income was $50.4 million. Distributable earnings available to common stockholders was $79.8 million or $0.71 per common share.

This non-GAAP measure is defined as net interest income plus TBA drop income, adjusted for interest income or expense on our interest rate swaps and futures contracts, minus net operating expenses. Quarter end book value was $18.63 per common share, up 6.5% from September 30. Our most recent current available estimate of book value as of Tuesday, February 17 was $18.37 per common share, which reflects the payment of our January dividend of $0.24 and the accrual of the entire February common dividend payable on February 27, 2026, again, up $0.24 per common share.

During Q4, ARMOUR raised approximately $3.8 million of capital by issuing approximately 183,000 shares of preferred stock through an at the market offering program. Through February 11, 2026, we raised approximately $138 million of capital under our common at the market program by issuing approximately 7.5 million shares of common stock, which is mildly dilutive. We also issued $4.8 million of capital from the issuance of 230,000 shares of preferred stock under our preferred at the market program. Our repaid monthly common dividends per share of $0.24 per common share per month for a total of $0.72 for the quarter. As we have stated previously, we aim to pay an attractive dividend that is appropriate in the context of stable over the medium term.

On January 29, we paid a cash dividend of $0.24 per outstanding common share to the holders of record of January 15, 2026. We have also declared cash dividends of $0.24 per outstanding common share, payable on February 27, 2026 and March 30, 2026 to holders of record on February 17 and March 16, respectively.

I will now turn the call back over to CEO, Scott Ulm, to discuss ARMOUR's portfolio position and current strategy.

Thank you, Gordon. ARMOUR REIT delivered a robust fourth quarter, marking a 6.5% increase in book value in the fourth quarter. The strong growth extended to our balance sheet. The portfolio grew for a second consecutive quarter, increasing by more than 10% from the end of the third quarter of 2025, driven by roughly 22 basis points of spread tightening while maintaining moderate leverage throughout the quarter. ARMOUR's mortgage assets now total over $20 billion, supported by a strong capital liquidity position of approximately 54% of total shareholders' equity as of the end of January.

We viewed Agency MBS as a high conviction opportunity from the onset of the Fed's easing cycle in the third quarter of 2024, and the backdrop for 2026 has now turned materially more supportive. Despite spreads tightening meaningfully so far in 2026, the market's appeal remains anchored in declining rate volatility and easing funding costs, supported by the Fed's efforts to lower rates and maintain ample banking liquidity. While prepayments have moved off their cyclical lows of recent years, they remain contained with primary mortgage rates still anchored around 6%. Adding a steeper yield curve and the result is a market that we expect to continue to favor MBS with compelling returns relative to returns in corporate credit where spreads are trading at historically tight valuations.

Technical supply and demand dynamics are now working with us, not against us. The administration's focus on lowering mortgage spreads reinforces a clear North Star for a stable mortgage market an objective, we expect Fannie Mae and Freddie Mac to support through [FIFA's] $200 billion MBS purchase mandate. The GSEs have posted strong monthly purchases of mortgage assets throughout last year while net issuance of conventional MBS remained negative in the fourth quarter. The imbalance has reminded attractive returns in the TBA roll market, creating a liquid carry environment and expanding the buyer base for Agency MBS.

I'll now turn it over to Desmond for more detail on our portfolio.

Thanks, Scott. ARMOUR's most recent net balance sheet duration stands at 0.14 years with a modest positive bias to the front end of the curve, consistent with easing monetary policy. Implied leverage, excluding treasury loans is 7.9 turns, a balanced posture that reflects tighter spreads and a lower volatility backdrop versus the prior year. The portfolio remains nearly 100% Agency MBS, Agency CMBS or DUS and U.S. treasuries to target specific yield curve exposures.

Consistent with our balance sheet growth, we added over [3 billion] of MBS pools and DUS across the fourth quarter and early first quarter. And our purchase mix has evolved as recent spreads have moved.

Early in the fourth quarter, we determined it was most attractive to overweight premium dollar MBS, which offer the most attractive spreads and yields. Anticipating that GSE purchases would most likely concentrate in [new par] coupons where the impact on primary mortgage rates is most direct, we added over [1 billion] of 4.5 and 5 coupon MBS ahead of Trump's GSE announcement in early January. As [belly] coupons titled to historically rich levels to near single-digit OES, we shifted towards lower coupons and seasoned collateral, where affordability initiatives aim that on [freezing] the housing market could drive higher turnover speeds whilst reserving higher yields in [deeper] discount MBS.

Within premium bonds, we focus more on [call] protection in higher-tier maximum non-sized pools, while keeping payoff targets at 24 [tics] or lower. In Agency CMBS, our 5-year DUS position experienced extreme spread tightening. On a relative value basis, 10-year DUS bonds now screen more attractive, particularly when hedged with longer-dated [super] swaps with [pay] fixed rates still cheaper than treasury hedges. Roughly 86% of our hedges are in OIS and (inaudible) waps with the balance in treasury futures. The benchmark 10-year (inaudible) swap spread has normalized back to its pre-liberation day average of approximately negative 37 basis points and we anticipate further gains will likely hinge on the path of policy debate around the Fed's desired balance sheet size and banking deregulation.

Aggregate portfolio prepayments averaged 11.1 CPR through Q4 2025 and Q1 2026 to date, versus 8.1 CPR in Q3 2025, stable, but running at a somewhat higher level versus the prior year. Despite tighter mortgage spreads, the 30-year mortgage rate has remained in a tight 6% to 6.3% band, though it has recently shifted towards the low end of that range. The administration's push for affordability without sacrificing home price appreciation, leaves mortgage rates and spreads as the 2 primary levers to accomplish that. However, the easy work has already been done, the mortgage rate spread for the 10-year treasury is now below its 15-year average. Further declines in mortgage rates will therefore require lower long-end pressures, which have not declined in sync with front-end rate cuts since the start of the easing cycle in 2024.

Still, we remain mindful that many originators have built significant capacity to ramp up refinancing, which could be triggered by a sustained move below 6% and may accelerate speed in part and premium coupons in coming quarters. Refi activity has proven to be highly sensitive to marginal mortgage rate declines keeping prepayment risk in TBAs and the generic premium MBS elevated. Coupon selection and specified collateral remained the key to containing the prepayment risk. We are positioned accordingly. Nearly 30% of assets are in prepayment protected agency CMBS pools and discount MBS while specified MBS pools with some form of prepayment protection, comprise over 92% of ARMOUR's portfolio.

Funding markets have also turned the corner. 2026 REPO conditions have improved materially versus last year. Markets are liquid and financing levels have eased with REPO rates averaging roughly so [far] plus 15 basis points. the (inaudible) to Fed fund spread has also normalized to near flat. As REPO rate backstop in late 2025, the Fed moved quickly to contain intra-month funding pressures tied to [falling] reserves and elevated [T-bill] supply.

First, the Fed continues to implement a policy of easing the overnight Fed funds rate. Second, it has shifted its reinvestment by directing paydowns of its treasury and MBS Holdings back into the treasury market. Third, it initiated outright purchases of up to $40 billion per month in treasury bills and other short-dated treasuries to stabilize reserve balances and maintain ample system liquidity.

This response reinforces the systemic importance of REPO markets as the foundation for liquid financial conditions and underscores the Fed's low tolerance for a repeat of the September 2019 episode well reserved scarcity and balance sheet frictions contributed to a sharp dislocation in secured funding. While the incoming share has signaled an appetite for a smaller Fed footprint and a reduced balance sheet over time, we expect the Central Bank's focus on orderly funding markets to remain the highest priority with the willingness to respond preemptively ahead of any emerging stress. As of today, we financed the portfolio across 23 active REPO counterparties approximately 80% of our repo principle is financed at a 3% haircut or lower and the weighted average haircut across the REPO book is approximately 2.75%. BUCKLER Securities accounts for roughly 40% to 60% of our REPO financing.

Back to you, Scott.

Thanks, Desmond. We continue to set our dividend with a medium-term outlook. While acknowledging relatively tighter spreads versus the prior year, we expect the backdrop of a steeper yield curve and lower volatility remains supportive for a consistent and predictable return profile for our assets. Our approach remains unchanged, stress test their liquidity, apply systematic hedging and deploy capital when opportunities present themselves. Overall, we're confident in our positioning, our strategy and our ability to deliver value for shareholders in 2026.

Before we open the line for questions, we'd also like to highlight that we've launched a new investor presentation now available on ARMOUR's website. It provides additional insight for investors, including how our portfolio is transformed over time. Thank you for joining today's call and for your continued interest in ARMOUR.

(Operator Instructions) Timothy D'Agostino, B. Riley Securities.

I was wondering on the portfolio interest-bearing assets. By my estimates, it increased year-over-year around like 49%. I was wondering the outlook in '26, do you see potential for similar growth or maybe a little bit less given the increase in 2025?

I think there are a couple of elements there, but certainly, one of the most important is capital raising. And we are -- when we see an opportunity to raise capital, combined with investment opportunities we like, we'll execute on that. But we are -- we discriminate a fair amount in terms of what we -- what is going to be attractive or not. So I'm afraid I got to tell you, it depends on how the market behaves, both on the investment side and the equity side of whether we will be similar or smaller or in some other relationship to what we're able to do last year.

Okay. Great. And then just to confirm, book value as of Tuesday was $18.37 per share?

Correct. And that's after the accrual of our full February dividend and the payment of our January dividend.

Trevor Cranston, Citizens.

Can you guys talk about where you're seeing incremental returns on new investments today given the spread tightening that's occurred and how you view the incremental level of return compared to the dividend you're currently paid?

This is Desmond Macauley. So on a primary basis, the level yield on [305], which are currently production coupon is around the mid-teens, let's say, about 15%. This assumes 8 turns of leverage, hedged to 0.5 duration using swap hedges. And it's a static framework over a period of just about 3 months. It doesn't assume any more spread tightening.

Now we think, at least in the medium term, we could see a bit more spread tightening. So let's say we get another 10 basis points of [OES] tightening. That adds about 4% to that return. And also the curve [with] stiffness are more. So if we have another -- if we see another 50 basis points in curve stiffening, particularly made by the front end through more [pair] cuts, which is we anticipate that can also add about another 1% or so.

So those are all parts of the full to return framework, some of that would accrue to our book value.

Now in terms of marginal capital raise, we see that, that hurdle rate is about 16%. So that would be dividend used to common and the management fee is just 75 basis points on new equity. So you add that together, that's roughly about 16%. So you can see that for production coupon, the base case returns are close to that level already. And with just a little bit more [steeper] and if we see more tightening, it would surpass that by a couple of more points.

Does that answer your question?

Yes, that's very helpful. And then I guess, in general, can you guys talk about how you're thinking about the likelihood of further actions driven by the government to attempt to lower mortgage rates things such as increasing the GSE portfolio limits further or potentially doing other things like lowering [GPs], et cetera?

This is Sergey. Yes. So around the [weak] and (inaudible), we were expecting maybe a few more announcements on the affordability push that the administration has announced with the GSE purchases. We haven't gotten anything. It feels to us that maybe the lowest hanging fruit has been picked in terms of pressuring spreads and mortgage rates lower, but without affecting home prices.

I think the next steps kind of have both positives and negatives for that push in terms of GSE, the [GC] cuts for the GSEs take away some of the profitability, make them less of a private enterprise, profitable enterprise and more of a policy tool. It will introduce negative complexity to investors who may demand wider spreads.

So some of the further steps may work counter to what administration has called the North Star in terms of keeping mortgage spreads nice and stable. We do expect more announcements. Obviously, there have been announcements on portability, assumability of mortgage loans, 50-year loans have been taken off the table. So there's a lot of announcements that we made. But once you get to the implementation stage of it, things have been quite slow.

Having said that, we definitely expect in the midterm year for these announcements to be quite active.

Dave Storms, Stonegate Capital.

I wanted to start with just asking for a little more [cuts] on your current liquidity. It looks like quarter over where you put a little more work than it looks like it's back up as of last month end. I guess how do you think about this in the near term?

So yes, I think our liquidity, we mentioned is about 54% of the total equity at the month end. It's a really good spot reflects our moderate leverage kind of where we have been steady in terms of liquidity. So we don't foresee any sharp changes given our current position in the portfolio.

Understood. And then I also know you mentioned in the prepared remarks that about maybe 30% of your portfolio is payment protected with mortgage rates hovering around to 6%. Do you -- I know the market like nice round numbers. Do you see any risk of a tipping point? Or is more maybe a linear situation as mortgage rates may continue to take lower?

Yes. I mean, look, prepayments have increased from Q4. So far in Q1, we noted in our script. We're definitely towards the lower range of the mortgage rates that we have been over the last couple of years. February prevailing mortgage rate will be lower after the GSE announcements as well.

So the risk of faster prepayments has increased, right? And I think in sync with that, our portfolio has morphed over the last couple of quarters to protect us more from lower mortgage rates, 30% in discounts and thus, specified pools make up 92%. Within the 92%, almost 40% is in the loan balance stories, others, credit and [geo] story. So we feel like there's -- there are faster refinances are in the future but we've built our portfolio to -- for that environment.

(Operator Instructions) Eric Hagen, BTIG.

I think you guys mentioned in the opening remarks, haircuts for MBS have come down as kind of an interesting comment. Can you maybe frame kind of where that level is relative to like the historical levels? And then if the GSEs are helping reduce volatility in the market, could we see that haircut level come down even further potentially?

We would hope so. I mean, a lot of the guidance on haircuts comes from FICC. But in terms of our bilateral counterparty REPO haircuts, we have worked with a lot of our counterparts to bring down the maximum haircuts closer to our weighted average of [2.75] I think a lot of -- almost 80% of our repo book is closer to a 3%.

Okay. Following up on the conversation around just where you are in the coupon stack. I mean you mentioned originators have been really able to leverage some of their tools to be aggressive on refi. I mean how does that drive the appetite for the current coupon specifically? And like the OAS that's in the current coupon, how do you compare that to some of the lower coupons and just where you feel comfortable taking prepayment risk?

Yes. We've been looking away from current coupon because that's kind of where the biggest impact from the announcement has been really all throughout the Q4. We did add in Q4, a little over [1 billion and 4.5 and 5]. But since then, probably are more looking at the wings, deeper discount coupons where we can see some of the housing activity perhaps reignites with any of these affordability measures. In terms of premium coupons, they're still our core holding.

If you look at the OAS spread difference between [102] price and current coupon MBS we're at close to 2 centers deviations and not spread historically speaking, right? So a lot of the fares and prepayments and [GP] cuts have already been priced into the premiums. So it's really looking at kind of a barbelled approach in the coupon stack at this point. But even within the [bell] as the coupon stack, you can find stories which pick OAS versus [TBA] specifically, maybe like season collateral, things like that.

How many Fed cuts do you feel like are currently priced into the mortgage basis?

How many Fed cuts?

Yes, how many Fed cuts for the rest of this year, do you think are priced into the mortgage basis?

The market is expecting by the end of December, a little bit over 2 cuts. And from our perspective, we think it's reasonable. We think that normalization will continue this year. It looks like when we get to around June, the probability is about 100%, getting close to 100%. And that will be a very good environment for the MBS market and mortgage spreads, we think that the curve is already steepened.

If we do see more cuts, then funding costs will come down, the curve would stiffen even more. And that makes the entire space more attractive and it adds to overall total return.

This concludes our question-and-answer session. I would like to turn the conference back over to Scott Ulm for any closing remarks.

Thank you very much for your interest in ARMOUR REIT. If there are follow-up questions, don't hesitate to call the office, and we will get back to you soon as we can. Thanks so much, and good morning to you.

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.