MFA Financial 2026年第三季度业绩电话会议：账面价值下降，可分配收益改善

核心要点

• MFA Financial公布2026财年第三季度GAAP净亏损为1.117亿美元，即每股普通股基本亏损1.19美元，主要是由于利率上升和抵押贷款利差扩大导致约1.53亿美元的按市值计价净额下降。
• GAAP每股账面价值环比下降12.5%至11.12美元，而经济账面价值下降13.3%至11.44美元。管理层估计，季末后经济账面价值已回升约5%。
• 可分配收益从第二季度的每股0.12美元增至0.26美元。扣除已实现信用损失前，每股可分配收益为0.44美元；若剔除0.05美元的非经常性收益，则约为每股0.39美元。
• 已实现信用损失降至1890万美元。存量多家庭住宅组合的未偿本金余额降至3亿美元以下，管理层预计信用损失将较当前水平显著放缓。
• MFA继续以更高的收益率部署资本。第三季度收购的贷款平均票面利率为7.8%，管理层表示当前投资可带来中高两位数（15%-19%）的股本回报率。
• 约13亿美元利率为1.4%的支付固定利率互换协议将于2027年第一季度末到期。这些互换目前每季度为每股可分配收益贡献约0.08美元。

关键财务数据

业务与运营表现

非合格抵押贷款（Non-QM）仍是MFA最大的资产类别，规模接近60亿美元。该公司在第三季度收购了超过4.8亿美元的Non-QM贷款，平均票面利率为7%，贷款价值比（LTV）为70%。季末后不久，MFA完成了其第25次Non-QM证券化，发售了3.6亿美元的债券。

MFA还赎回了其2024年不良贷款证券化资产，并发行了3.79亿美元平均票面利率为6%的新债券。该交易释放了4000万美元的现金和融资能力，用于新投资。

机构抵押贷款支持证券（Agency MBS）投资组合总额约为40亿美元，接近投资组合的三分之一。MFA将较低票面利率的证券置换为较高票面利率的特定资产池（specified pools），并继续转滚其4.78亿美元的TBA头寸。管理层表示，当前的机构证券利差可支撑中高两位数（15%-19%）的股本回报率。

Lima One在该季度发放了2.88亿美元的贷款，包括2.14亿美元的短期过渡性贷款和7400万美元的30年期租赁贷款。该公司出售了9500万美元的租赁贷款，并录得240万美元的出售收益。

Lima One在2024年春季暂停多家庭住宅贷款业务后，也恢复了该业务的发放。首批贷款总额约为500万美元。该公司计划发放这些多家庭住宅贷款用于出售，而不是保留在资产负债表上。

存量多家庭住宅风险敞口继续下降。剩余本金降至3亿美元以下，公允价值为2.63亿美元，逾期余额降至7100万美元。

MFA还概述了其管理层交接计划。Craig Knutson计划于2027年6月底退休卸任CEO，Bryan Wulfsohn将于2027年7月1日出任总裁兼CEO。Andrew Kal预定于2027年1月1日加入公司并担任首席投资官。

管理层展望

管理层预计，随着公司完成对存在问题的存量多家庭住宅贷款的处置，已实现信用损失将大幅放缓。管理层表示，大部分相关损失已反映在可分配收益中，并认为迈入2027年之际，多家庭住宅信用问题已基本解决。

公司预计，在当前水平上的新投资将带来中高两位数（15%-19%）的股本回报率。收益增长的速度将取决于市场波动性、利率、资产部署情况以及MFA回收低效资本的能力。

管理层表示，Lima One恢复的多家庭住宅业务放贷规模可能会超过1.00亿美元，但也强调其具体时间和规模仍难以预测。

董事会将继续根据长期收益展望、应纳税所得额、流动性、市场状况及其他因素，按季度审议普通股股息。管理层未承诺维持当前每季度0.36美元的股息水平。

风险与关注事项

利率的大幅上升和抵押贷款利差的扩大拖累了第三季度大部分的账面价值下降。在该季度，2年期美国国债收益率上涨超过70个基点至4.89%，而10年期国债收益率上涨近85个基点至5.29%。

MFA将其投资组合套期保值至约一年的久期缺口，但管理层表示，国债收益率上升、抵押贷款利差扩大以及互换利率涨幅小于国债收益率，共同给账面价值带来了压力。

约13亿美元利率为1.4%的支付固定利率互换到期是一个已知的收益不利因素。这些互换每季度为每股可分配收益贡献约0.08美元，预定于2027年第一季度末到期。

信用风险仍集中在不断缩减的存量多家庭住宅组合中。管理层还指出，Lima One的新建建筑贷款是对房价进一步下跌最敏感的住宅业务领域，不过管理层表示尚未观察到借款人放弃项目的情况。

分析师问答集锦

• 底层收益：管理层澄清，在剔除0.05美元的非经常性收益后，已实现信用损失前的可分配收益约为每股0.39美元。投资者还应考虑到即将到期的互换合约带来的每股0.08美元贡献。
• 净利息利差：报告的利差改善至1.63%。管理层预计高收益的新投资将支撑利差，尽管即将来临的互换到期将减少报告的套息收益。
• 证券化市场状况：MFA观察到第三季度利差有所扩大，但表示市场随后趋于稳定，部分领域利差有所收窄。管理层继续认为证券化市场具备深度与韧性。
• 住宅信用：MFA在承销时不假设房价会出现大幅上涨。管理层表示，房价平稳或适度下跌对Non-QM信用表现的影响有限，而Lima One的新建建筑贷款对更深幅度的下跌更为敏感。
• 股票回购：回购活动受到优先股定价的制约，此前MFA一直在发行优先股来为普通股回购筹集资金。管理层表示，普通股交易价格较账面价值存在大幅折价，但正在权衡回购与流动性、业务规模及新投资机会之间的关系。

业绩电话会议完整文字记录

完整财报电话会议逐字稿

管理层陈述

Operator

Greetings, and welcome to the MFA Financial Third Quarter 2026 Financial Results Call. [Operator Instructions] As a reminder, this conference is being recorded. I would now like to turn the conference over to your host, Hal Schwartz, General Counsel. Thank you. You may begin.

Harold Schwartz

Thank you, operator, and good morning, everyone. The information discussed on this conference call today may contain or refer to forward-looking statements regarding MFA Financial, Inc. which reflect management's beliefs, expectations and assumptions as to MFA's future performance and operations. When you use statements that are not historical in nature, including those containing words such as will, believe, expect, anticipate, estimate, should, could, would or similar expressions are intended to identify forward-looking statements.

All forward-looking statements speak only as of the date on which they are made. These types of statements are subject to various known and unknown risks, uncertainties, assumptions and other factors, including those described in MFA's annual report on Form 10-K for the year ended December 31, 2025, and other reports that it may file from time to time with the Securities and Exchange Commission.

These risks, uncertainties and other factors could cause MFA's actual results to differ materially from those projected implied in any forward-looking statements it makes. For additional information regarding MFA's use of forward-looking statements, please see the relevant disclosure in the press release announcing MFA's third quarter 2026 financial results. Thank you for your time.

I would now like to turn the call over to MFA's CEO, Craig Knutson.

Craig Knutson

Thank you, Hal. Good morning, everyone, and thank you for joining us for MFA Financial's Third Quarter 2026 Earnings Call. With me today are Bryan Wulfsohn, our President and Chief Investment Officer; Mike Roper, our Chief Financial Officer; and other members of our senior management team.

Before I get to the quarter, I'd like to say a few words about the announcement that we made on October 8. As you sure saw, I plan to retire from my role as Chief Executive Officer at the end of June next year and Bryan will become President and CEO on July 1, 2027. Bryan will also join our Board on January 1 of next year, and I expect to remain on the board after I retire, so I'm not going to disappear. This transition is the result of a deliberate succession process on which I have been working with our Board since early 2023.

Bryan has run our residential whole loan portfolio since we began buying loans in 2014 and since late 2024 has overseen Lima One and our Agency MBS and BPL portfolios. Bryan has played an active and central role in shaping both the MFA of today and the strategy that guides us. I've worked with him for over 20 years, both at MFA and previously, and I can't think of anyone better suited to lead the company than Bryan.

We also announced that Andrew Kal will join us on January 1 as Chief Investment Officer. I've known Andrew for over 15 years, and he brings over 25 years of mortgage finance experience, most recently as Head of Residential Lending at Starwood Capital. Andrew's arrival will let Bryan spend the first half of next year preparing to step into the CEO role while our investment team gains the benefit of another very experienced leader with deep expertise in our target assets. Bryan will say more about this in a few minutes.

It's been a privilege to work at MFA for the last 18.5 years and to lead the company for the last 9. I'm sure I'll have more to say on that topic in the coming months. But for now, there's plenty of work to do. So let me turn to the quarter. The third quarter was a difficult 1 for rates, particularly in September. During the quarter, the market began to come to terms with a more hawkish Fed as inflation remains sticky and the economy and payroll stayed strong.

In September, the Fed raised the Fed funds rate by 25 basis points their first increase since 2023, and the accompanying message made clear that more increases will likely follow. Although the Fed left rates unchanged at their meeting last week, market expectations are overwhelmingly anticipating a hike at the December meeting. Treasury yields moved sharply higher in the third quarter. The 2-year yield rose over 70 basis points during the quarter from $4.17 to $4.89. The 10-year rose almost 85 basis points from 4.46% to $5.29, the highest level in 2 decades.

The last time the 10-year closed this high was in 2007. Most of that move came in September and much of it in the final week. Oil prices climbed with continued tension with Iran, rate volatility spiked with the Move index rising from the high 70s at the start of September to above 110 at month end, its largest 1-week jump since April of last year. Against that backdrop, credit spreads were relatively resilient, but mortgage spreads were not.

Agency MBS underperformed meaningfully in September. Non-QM securitization spreads, which tightened into mid-September widened back out to about 140 to 150 basis points on AAAs by quarter end. Mortgage rates rose above 7.25%, up about 50 basis points in September alone. We felt the impact of these moves our portfolio carries interest rate risk that we hedge to an approximate 1-year duration gap, but the combination of a sharp increase in rates, wider mortgage spreads and swap rates that rose less than treasuries collectively worked against us. It is worth noting that most of the book value decline was due to unrealized losses from marks.

Mike will walk through the drivers in more detail. While book value was disappointing, we continue to make progress on the operating priorities we laid out for you over the past year. First, we continue to work through our legacy credit book. realized credit losses improved from the second quarter, as we said they would. Our overall portfolio 60-plus day delinquency rate improved again to 6.7% and our legacy multifamily portfolio shrank to under $300 million UPB and $263 million fair value.

With a clear path to resolution on most of the remaining multifamily loans, and most of these losses already reflected in DE, we believe our multifamily credit issues are largely behind us, particularly as we look forward to 2027. Second, the earnings power of the portfolio continued to improve. Distributable earnings rose to $0.26 per share from $0.12 in the second quarter and DE prior to realized credit losses was $0.44 per share.

Third, we kept our expense base tight. G&A for the quarter was $23.6 million, well below the $33 million quarterly average of 2024. Fourth, we continued issuing in the securitization market. We issued a new $447 million NPL securitization in September after calling our 2024 NPL deal, unlocking an additional $40 million of capital and issued a non-QM deal shortly after quarter end.

As we have discussed previously, the securitization markets are remarkably deep and resilient even in the face of general fixed income market volatility. None of this makes the third quarter book value decline easier to accept, and we still have more work to do. We need to finish resolving the legacy credit book and turn improving earnings power into consistent results. But the same market that hurt us in the third quarter has also created opportunity.

Loans that we acquired in Q3 carried an average coupon of 7.8%, and we expect returns on new investments today will produce mid- to high-teen ROEs. We will remain disciplined, but we will continue to put capital to work in this attractive investment environment. I'd like to turn the call over to Mike now to discuss our financial results.

Michael Roper

Thanks, Craig, and good morning, everyone. At September 30, GAAP book value was $11.12 per share and economic book value was $11.44 per share, down approximately 12.5% and 13.3%, respectively, from the end of the second quarter. MSA again paid a common dividend of $0.36 and delivered a quarterly total economic return of negative 10.6%. For the third quarter, we reported a GAAP net loss of $111.7 million or $1.19 per basic common share.

The loss was driven by net mark-to-market declines of approximately $153 million across our various fair value assets, liabilities and hedges as rates rose sharply and mortgage spreads widened, particularly in late September. Net interest income, including TBA dollar roll income was $63.4 million, up from $60.2 million in the second quarter. Our G&A expenses declined $23.6 million, attributable primarily to nonrecurring depreciation expenses in the second quarter and a nonrecurring benefit from a true-up of incentive compensation accruals in the third quarter.

Moving to distributable earnings. DE for the quarter was $26.4 million or $0.26 per share, up from $0.12 per share in the second quarter. Realized credit losses on fair value loans declined to $18.9 million consistent with the expectations we shared last quarter. DE prior to realized credit losses was $45.3 million or $0.44 per share. That figure includes approximately $0.05 per share of nonrecurring benefits related to a refund of disputed servicing costs and a reduction of incentive compensation accruals in the third quarter.

Excluding those items, DE prior to realized credit losses was approximately $0.39 per share, an increase of 11% from last quarter and 30% from the third quarter of 2025. This marks the third straight quarter of significant growth in the underlying earnings power of our portfolio and reflects the progress of the strategic initiatives that we shared a year ago. In light of the significant market volatility over the last few months, we also want to provide an update on how we're thinking about the path of our distributable earnings from here.

Earlier this year, we shared our expectations for elevated credit losses in 2026 and as we work to resolve our legacy multifamily portfolio. This has played out largely as we expected. While these resolutions have been a drag on our -- many of our most troubled legacy multifamily loans have now been resolved or charged off, and we continue to expect realized credit losses to moderate meaningfully from here.

As we look past the near-term, the recent move in rates and spreads has added significantly more variability to the longer-term outlook for RDE. We also have some known headwinds with approximately $1.3 billion of 1.4% pay fixed interest rate swaps that mature at the end of the year and into the first quarter of next year. These swaps contribute approximately $0.08 per share per quarter to our DE today.

We have meaningful options to offset this loss carry and to continue to grow the portfolio's earnings power with new investments coming on of returns in the mid- to high teens, the continued recycling of unproductive capital and a balance sheet capable of supporting significantly more leverage. How quickly this build will depend in part on how volatility, rates and markets evolve.

As always, our Board will continue to evaluate the dividend each quarter, taking into account our long-term earnings outlook, taxable income, market conditions and other factors.

Finally, subsequent to quarter end, we estimate that our economic book value has increased by approximately 5% since the end of September, driven by a rally in interest rates. I'd now like to turn the call over to Bryan , who will discuss our investment portfolio and Lima ONE.

Bryan Wulfsohn

Thanks, Mike. Before I get into the portfolio, I want to add a few words to what Craig said about our announcement last month. I'm honored that the Board has asked me to lead MFA, and I'm grateful to Craig who has been a mentor and partner to me for more than 2 decades. We have a long runway to make this a smooth handoff next summer. I'm also very glad Andrew Kal is joining us in January. I've known Andrew for over 15 years. He has built and run residential lending businesses, and he knows the non-QM business purpose and broader residential credit markets from both sides of the table.

As CIO, Andrew will lead our investment activities alongside the team that has been with us through many cycles. Andrew adds experience, expertise and relationships to our investment team, and we look forward to working with him.

Now to the quarter. Despite a very volatile September, we continue to add assets, securitize loans, reduce our legacy credit exposure and make improvements at Lima One. I'll touch on each. Starting with NonQM, which remains our largest asset class at nearly $6 billion. We acquired over $480 million of loans during the quarter with an average coupon of 7% and an LTV of 70%.

These were sourced through a select group of long-standing origination partners. Just after quarter end, we issued our 25th non-QM securitization, successfully selling $360 million of bonds and locking in nonrecourse nonmark-to-market financing on the underlying loans. We also remain pleased with the performance of our seasoned RPL and NPL loans, which, as a reminder, were loans originated pre-GFC that were acquired at discounts to par between 2014 and 2019. The default rate in that portfolio fell to another all-time low during Q3.

In September, we called our 2024 NPL securitization and issued $379 million of new bonds with an average coupon of 6%. That transaction unlocked $40 million in cash and financing capacity that we can utilize to make new investments. Turning to our agency book. which comprises nearly 1/3 of our investment portfolio at roughly $4 billion. Though we did not sell any bonds due to volatile market conditions, we were active during the quarter rotating lower coupon securities into higher coupon spec pools.

We also continued to roll our $478 million TBA position. Agency spreads widened in September as rate volatility spiked, which hurt book value this quarter. But at current spreads, we believe agencies offer mid- to high-teens ROEs and remain an attractive liquid complement to our credit assets.

At Lima One, origination volume was $288 million in the third quarter, including $214 million of short-term transitional loans and $74 million of 30-year rental loans. Although these figures were a slight drop from Q2 and we have not and will not compromise the underwriting improvements we've made over the past 2 years in order to grow volume. We sold $95 million of rental loans during the quarter, generating $2.4 million of gain on sale income.

And we're pleased to announce that last week, Lima originated and distributed its first multifamily loans since the spring of 2024. This was the culmination of months of hard work by an entirely new underwriting regime. As with Lima's 30-year rental production, our business plan for multifamily will be originated to sell.

Finally, we continue to wind down our legacy assets, our multifamily loan book fell below $300 million of remaining principal and the delinquent bucket fell to $71 million. As Mike noted, credit losses came in below the second quarter, and we expect them to moderate meaningfully from here. Across our entire loan portfolio, 60-plus delinquencies improved to 6.7% from 7% at June 30.

Overall, while the rate environment made for a difficult quarter for book value we were able to add assets at attractive levels. Lima continued to perform well, and our problem asset book continues to shrink. And with that, I'll turn the call over to the operator for questions.

Operator

[Operator Instructions] Our first question comes from the line of Bose George with KBW.

分析师问答

Unknown Analyst

This is Frank on for Bose. Just want to touch on Lima One. You guys mentioned earlier in the year, returning to the multifamily business like can you discuss maybe the pace there and how you're envisioning multifamily being a bigger part of the book going forward?

Bryan Wulfsohn

Yes. So we just originated our first loans last week. Those amounted to around -- somewhere around $5 million in loans. In terms of the expectation going forward and how big it could be, it's hard to really forecast out exactly we think it might be next year. But we think it's possible that, that asset could easily be above $100 million of originations, and we'll see where it goes from there.

Unknown Analyst

Great. That's very helpful. And then just on the dividend. How are you thinking about the current dividend and the current run rate going ahead? You mentioned some headwinds with some swaps turning off on.

Craig Knutson

Sure. So our Board obviously sets our dividend policy taking into account a number of factors, including our longer-term earnings outlook, taxable income, liquidity, peer group dividend policy and other factors. There's no question that the operating environment is more challenging today than it was in August when we held our second quarter earnings call. but considerable uncertainty remains, not the least of which will be the market response to the midterm election results, which obviously we don't know yet.

Our Board will take all of this into consideration when setting dividend policy. For our part, we're focused on the long-term earnings power of the portfolio and prudently deploying capital to make that happen.

Operator

Our next question comes from the line of Melissa Lobo with UBS.

Unknown Analyst

Just looking at net interest spread, which improved to 1.63 this quarter. How should we think about the trajectory from here as the portfolio is repricing and particularly the volatility in the securitization market?

Michael Roper

Sure. So this is Mike. I'm happy to answer the first part of that question, and maybe I'll ask Bryan to respond to your question on securitization market. That number can be pretty sticky, right? For GAAP purposes, you don't get to rerack your assets or liabilities or hedges. The spread is sort of locked in at the time you purchase an asset. I think over the last several quarters, we've seen a bit of a shift, which is really driven in part by the increased allocation to Agency MBS, which is a higher leverage, lower spread product.

In terms of the outlook for spread though, as I think Craig said in his prepared remarks, the assets we're adding today are in the high -- mid- to high teens, which certainly should help that number as we move forward. From a financial reporting perspective, as I mentioned in my prepared remarks, we have these swaps, which are rolling off, which really are not doing much economically given the extremely short duration left in them. However, that decrease in carry will have an adverse impact on the reported spread, notwithstanding the fact that there isn't that big of an impact economically.

Bryan Wulfsohn

And as it relates to securitization markets, we did see widening in Q3. However, the first month of the quarter and recently, we've seen stabilization and some tightening -- so we do think that, that market is very deep, and assets also move wider in terms of spreads when securitization markets, spreads are moving wider as well. So we believe that sort of the NIM, the spread is still able to be maintained there.

Unknown Analyst

Okay. Great. And then on residential credit, seeing the delinquencies improved this quarter. although non-QM very modestly higher. As you look forward, how are you thinking about credit performance if home prices stay flat to modestly lower? And where in the residential portfolio, would you expect to see the most sensitivity?

Bryan Wulfsohn

Mean we -- when we buy assets and originate assets, we don't assume much in the way of HPA growth. So we take a conservative approach when we're looking at projections of home prices as they factor into our credit decisions. So flat home prices are moderately down. We do tend to see right? Like if you were to think about where the pressure could be on the new construction on the Lima side, if things were to go down further, right? You're worried about people potentially walking away from projects, but we haven't seen that at all.

We've seen improvement in that business even in the face of sort of moderating home prices. And on the NonQM side, you look at historical losses we've had there, it's been 0. So we've been sort of wrong to the good, I guess, for the benefit of our portfolio. If we see some flat to moderate, see a basis point or 2 of loss in that portfolio, we're -- it's very minor. We're not concerned there.

Operator

[Operator Instructions] Our next question comes from the line of Jason Weaver with Jones Trading.

Jason Weaver

First of all, congrats to Craig and Bryan on their transition. And it's nice to see the pre-loss number clear the dividend for this quarter. And I wanted to clarify something on that during your prepared remarks that I might have jumbled in my mind. I think you alluded to a run rate of near $0.39 or is that misconstrued?

Michael Roper

It's -- yes, this is Mike. Thanks for the question. We had this quarter sort of nonrecurring benefit from 2 items that amounted to about $0.05. So if you were to kind of strip those out, it's $0.39. Obviously, if you think about sort of the run rate of that number, I also mentioned in my prepared remarks that we have some swaps that are expiring that contributed about $0.08 this quarter. So I think you got to sort of add those 2 together as the portfolio sits today to construct a run rate.

Jason Weaver

Understood. And then on capital priority, it looks like you bought back about 1 million shares in the first half, and it looks like this quarter, you held off a bit and held more liquidity Help me understand how you're weighing the repurchase against keeping dry powder and what would tilt that back towards buying?

Bryan Wulfsohn

Sure. So we announced this quarter last year, a program to effectively swap preferred for common -- so we've been issuing preferred by the ATM for each of the last several quarters and using those proceeds to buy back common. The preferred market has kind of moved away a little bit from levels that we deem to be accretive in terms of where we want to issue that capital which is why the overall volume there was down this quarter. It's really driven by the constraint of how much preferred we can sell at levels we're comfortable selling it more so than a desire to buy back common.

As far as the overall desire to buy back common versus other uses or preserving liquidity, obviously, the stock is very cheap right now and trading at a material discount to book. And we think it's a compelling opportunity there. At the same time, we're focused on maintaining the scale of the business as well as looking at longer-term returns from the assets that are currently yielding in the mid- to high teens today.

Operator

Ladies and gentlemen, there are no other questions at this time. I'll turn the floor back to Mr. Knutson for final comments.

Craig Knutson

All right. Thanks, everyone, for joining our call today, and we'll look forward to speaking with you in February.

Operator

Thank you. This concludes today's conference call. You may disconnect your lines at this time. Thank you for your participation.