Ellington Financial Inc. Q3 FY2025 Earnings Call

Good morning, everyone. Thank you for joining us. Welcome to the Ellington Financial Third Quarter 2025 Earnings Conference Call. This call is being recorded. I will now hand it over to Alaael-Deen Shilleh to start.

Thank you. Before we begin, I'd like to remind everyone that this conference call may include forward-looking statements within the meaning of the safe harbor provisions of the Private Securities Litigation Reform Act of 1995. These statements are not historical in nature and involve risks and uncertainties detailed in our annual and quarterly reports filed with the SEC. Actual results may differ materially from these statements, so they should not be considered to be predictions of future events. The company undertakes no obligation to update these forward-looking statements. Joining me today are Larry Penn, Chief Executive Officer of Ellington Financial; Mark Tecotzky, Co-Chief Investment Officer; and J. Herlihy, Chief Financial Officer. Our third-quarter earnings conference call presentation is available on our website, ellingtonfinancial.com. Today's call will track that presentation, and all statements and references to figures are qualified by the important notice and end notes in the presentation. With that, I'll hand it over to Larry.

Thanks, Alaael-Deen. Good morning, everyone, and thank you for joining us today. I'll begin on Slide 3 of the presentation. Ellington Financial delivered another quarter of strong performance with strategic execution, highlighted by the continued growth of our adjustable distributable earnings, the continued growth of our investment portfolio, and the continued strengthening of our balance sheet. For the third quarter, we reported GAAP net income of $0.29 per share and ADE of $0.53 per share, which set a new quarterly high for this metric since we started reporting it in 2022, and which once again significantly exceeded our $0.39 per share dividends for the quarter. The increase in ADE over the past several quarters is the direct result of higher net interest income from our loan portfolios, reflecting both the growth of those portfolios and their strong ongoing credit performance, combined with sizable proprietary reverse mortgage securitization gains at Longbridge, where we're now completing roughly reverse mortgage securitization per quarter. Our quarterly results also benefited from robust gains from securitizations of non-QM loans and closed-end second lien loans. We priced a total of 7 securitizations during the quarter. That's a record for us. And including transactions completed subsequent to quarter end, have now priced a total of 20 securitizations year-to-date. That's more than triple last year's pace. The EFMT securitization franchise has become a tremendous asset for Ellington Financial, allowing us to access liquidity for many of the largest fixed income investors in the world. Finally, in addition to all these drivers, we also had excellent performance in our securities businesses and strong earnings from our affiliate loan originators such as LendSure. Shifting to our balance sheet. Our total portfolio holdings grew by 12% during the quarter as we continue to deploy capital to our highest conviction loan businesses. Portfolio growth was led by non-QM, proprietary reverse mortgage, and commercial mortgage bridge loans and complemented by opportunistic additions of other residential mortgage loans and CLOs as well. Notably, Longbridge delivered a record quarter for proprietary reverse origination volumes. Demand from borrowers for Longbridge's proprietary products continues to grow. But just as importantly, demand from investors remains strong for the resulting securitization debt tranches. I believe that Ellington Financial, with our Longbridge subsidiary and our securitization franchise, is uniquely positioned to profitably intermediate between proprietary reverse mortgage borrowers and investment-grade debt investors. Moving to the financing side. We further strengthened our balance sheet by increasing our long-term non-mark-to-market financings in two key ways: first, by accelerating our pace of securitizations, and second, by expanding our access to long-term unsecured financing. First, as I mentioned earlier, we priced 7 securitizations during the quarter. Importantly, we are close to pricing our inaugural securitization of residential transition loans, which would reduce reliance on short-term financing in that strategy as well, unlock capital for redeployment, and create high-yielding retained tranches for our portfolio. Securitizations are important because they provide stable non-mark-to-market funding while reducing reliance on repo. This greatly enhances capital efficiency, and I'll elaborate on that later in my concluding remarks. Securitizations also allow us to manufacture high-yielding retained tranches with valuable call options. These retained tranches are generating some of the most attractive returns across our platform, and they contribute strongly to ADE even without repo financing. Second, on the financing side, on the final day of the third quarter, we successfully priced $400 million of 5-year senior unsecured notes rated by Moody's and Fitch and broadly distributed to institutional investors across more than 80 accounts. We utilized more than half the proceeds to reduce repo borrowings, with the remainder funding new high-yielding investments. The unsecured notes priced at 7 3/8%, representing a 363 basis point spread over the 5-year treasury at the time. We were pleased with the execution and the strong investor demand and encouraged to see the bonds trading well following issuance. We expect pricing to improve on future transactions as we become a more established unsecured note issuer and as we migrate a greater share of our borrowings to long-term financing, creating a virtuous cycle. With that, I'll turn the call over to JR to walk through our financial results in more detail. JR?

Thanks, Larry. Good morning, everyone. For the third quarter, we reported GAAP net income of $0.29 per common share on a fully mark-to-market basis and ADE of $0.53 per share. On Slide 5 of the deck, you can see the portfolio income breakdown by strategy, $0.42 per share from credit, $0.04 from Agency, and $0.09 from Longbridge. And on Slide 6, you can see the ADE breakdown by segment, $0.59 per share from the Investment Portfolio segment and $0.16 from the Longbridge segment. In the credit portfolio, net interest income grew sequentially, and we also had net realized and unrealized gains on residential transition loans and other loans in ABS. Partially offsetting higher net interest income were net realized and unrealized losses on non-QM retained tranches, CLOs, forward MSR-related investments, and residential REO. We continue to benefit from solid credit performance in our loan portfolios and from strong earnings at our affiliate loan originators. I'd like to highlight a new slide in the earnings presentation. Please turn to Slide 15. This slide illustrates the strong credit performance of our loan portfolios over time, reflected in the exceptionally low realized credit losses across our residential and commercial loan strategies since each business's inception. Note that the realized credit loss rate is shown on a cumulative inception-to-date basis. If presented on an annualized basis, these percentages would be even lower. This metric captures the quality of our loan underwriting, incorporating both loans that performed as expected and paid off at maturity and loans that require individual workout efforts. On the top right, you'll see just 13 basis points of cumulative realized credit losses on approximately $14.7 billion of residential mortgage loan fundings, spanning non-QM, RTL, home equity, and proprietary reverse mortgages. And on the bottom right, cumulative losses totaled only 47 basis points on more than $2 billion of commercial mortgage bridge loan originations dating back to before COVID. The combination of the strong credit performance and the high yield of these loans has been a key driver of EFC's sustained growth in ADE over time. Moving to Agency. That portfolio also generated strong results in the third quarter with net gains on both our long Agency RMBS and associated interest rate hedges. Lower interest rates and reduced volatility, together with tightening Agency yield spreads created a favorable environment that was broadly supportive of portfolio performance. The Longbridge segment had another excellent quarter with strong contributions from both originations and servicing. Origination profits were driven by higher origination volumes of proprietary reverse mortgage loans, higher origination margins for HECM reverse mortgage loans, and net gains related to the proprietary loan securitization completed during the quarter. Meanwhile, base servicing net income, strong tail securitization executions, and a net gain on the HMBS MSR equivalent, primarily due to tighter HMBS yield spreads, drove the contribution from servicing. These gains were partially offset by a net unrealized loss on the retained tranches of proprietary reverse securitizations due to faster prepayment speed assumptions, lower home price appreciation projections, and higher applied discount rates. Turning now to portfolio changes during the quarter. Slide 7 shows an 11% increase in our adjusted long credit portfolio to $3.56 billion quarter-over-quarter. Our portfolios of non-QM loans, commercial mortgage bridge loans, other residential loans, and CLOs all expanded, as did our portfolio of retained non-QM RMBS in that case from the securitizations we executed during the quarter. These increases were partially offset by the impact of loans sold into securitizations, net sales of non-Agency RMBS, and the smaller residential transition loan portfolio, with principal paydowns in that portfolio exceeding new purchases. For our RTL, commercial mortgage bridge, and consumer loan portfolios, we received total principal paydowns of $352 million during the third quarter, which represented 21% of the combined fair value of those portfolios coming into the quarter, as those short-duration portfolios continue to return capital steadily. On Slide 8, you can see that our total long Agency RMBS portfolio decreased by 18% to $221 million due to net sales. Slide 9 illustrates that our Longbridge portfolio increased by a substantial 37% to $750 million, driven by a record quarter of proprietary reverse mortgage loan originations, partially offset by the impact of a proprietary reverse securitization completed during the quarter. Please turn next to Slide 10 for a summary of our borrowings. At September 30, the total weighted average borrowing rate on recourse borrowings decreased by 8 basis points to 5.99% overall, with a notable 17 basis point decline on credit borrowings. Quarter-over-quarter, the net interest margin on our credit portfolio increased by 54 basis points, reflecting both that lower cost of funds as well as higher asset yields, while the NIM on Agency decreased slightly by 2 basis points. At September 30, our recourse debt-to-equity ratio was 1.8:1, up slightly from 1.7:1 as of June 30, while our overall debt-to-equity ratio was down slightly to 8.6:1 from 8.7:1. During the quarter, we improved financing terms on two Longbridge-related facilities. On September 30, we priced $400 million of 5-year senior unsecured notes at a fixed coupon of 7.3%, which was a 363 basis point spread to the 5-year U.S. treasury at the time. Consistent with our goal of staying neutral to the path of interest rates, upon pricing, we immediately swapped the fixed coupon to a floating rate, thus locking in that 363 basis point spread. The notes issuance closed in early October and will appear on our balance sheet beginning in the fourth quarter. As of October 31, nearly 20% of our recourse borrowings are unsecured. And of equal importance, the percentage of those borrowings subject to mark-to-market margining declined to 61% from 74% month-over-month. We expect our notes offering to increase our overall cost of funds by approximately 17 basis points. Keep in mind that this figure does not capture the expected accretive benefit of adding more assets at yields above the cost of this debt, as well as the release of capital reserves related to the repo paydown, which Larry will elaborate on later. In keeping with our mark-to-market philosophy, we'll elect the fair value option on these new unsecured notes as we have for our other notes and mark them to market through the income statement. As a result of this election, we expensed all associated deal costs in October rather than amortizing them over the life of the notes. At September 30, combined cash and unencumbered assets were about $1.2 billion, or about two-thirds of our total equity. Book value per share was $13.40, and economic return for the third quarter was 9.2% annualized. With that, I'll pass it over to Mark.

Thanks, JR. This quarter has been crucial for EFC's development. As Larry noted, we are making progress with ADE and enhancing our company's resilience. Repo financing markets are currently operating smoothly with plenty of liquidity and competitive terms. However, EFC's balance sheet is more robust when we diversify our funding sources and decrease our reliance on short-term repo. Our recent debt deal and the seven securitizations we've completed, where we shifted from repo funding to non-mark-to-market debt, are significant steps in this direction. Our resilience is also attributed to the downside protections we've implemented, such as our credit hedges. Although these hedges affected our returns this quarter, we see them as a vital safeguard, especially as we notice some potential issues in the economy emerging. For instance, there have been two prominent bankruptcies in the corporate credit markets, and job growth has weakened considerably compared to earlier this year. These market risks could become broader, and our credit hedges are structured to guard against that. Additionally, our strategy of focusing on higher FICO scores, lower LTV loans, and our purchasing activities further strengthen our portfolio's resilience. We’re committed to investing in proprietary technologies that help our affiliate loan originators and partners originate and deliver loans more efficiently to us. These technology investments are yielding positive results through increased purchase volumes, as we have significantly broadened our network of originators selling loans to us. We are also optimistic about technology's potential to automate and enhance various aspects of loan underwriting. These initiatives contributed to our 12% growth in the portfolio this quarter, attributed to our loan strategies and our effective securitization process. With rates inching lower, we expect this trend to persist or even accelerate. Alongside increasing our loan purchases, we are broadening our reach. We have recently started buying certain types of loans eligible for purchase by Fannie Mae and Freddie Mac, which have been increasingly acquired by private investors instead. We plan to launch a securitization for these loans in the upcoming months. This agency-eligible mortgage segment is particularly promising, as the current administration seems open to private capital stepping into areas that were traditionally dominated by the GSEs. This presents a unique opportunity for EFC. We are also focusing on acquiring seasoned mortgage loan portfolios from banks. With lower rates and narrower spreads, many bank portfolios that were previously significantly underperforming are now much healthier, prompting many banks to divest what they view as noncore assets. Since the beginning of October, we have noticed an uptick in loan packages being offered by bank sellers, and we have already acquired two of these. We believe this could represent a meaningful growth opportunity for us moving forward. Overall, we are following the same EFC strategy while also expanding it. Our approach involves using proprietary sourcing methods to buy a wide array of mortgage products, securing financing through securitizations, and retaining high-yielding tranche investments along with potentially valuable call options. We began this approach in 2017 with non-QM deals and subsequently added second liens and proprietary reverse loans. Now, we are entering the market with an RTL deal and plan to securitize agency-eligible mortgages soon. As we widen the range of products we buy and securitize, we are also offering our affiliate loan originators more ways to increase their earnings by expanding the types of products they can provide to customers. These synergistic relationships promote our portfolio growth while also boosting our affiliate loan originators' profits, which in turn enhances our earnings and book value per share through our equity stakes in these originators. This strategy has been pivotal in driving strong ADE growth this year, especially given the substantial contributions from our retained tranches and our stakes in originators such as Longbridge. Expanding our presence in the securitization markets has created a virtuous cycle. Over the past eight years, we have established our EFMT securitization brand, starting with our first deal in 2017. Each new transaction and loan product further solidifies our market reputation and grows our investor base, leading to improved execution levels. Looking forward, we see a broader array of opportunities than we did in the first half of the year. Loan volumes have increased as mortgage rates are more than 80 basis points lower than earlier this year. With our expanding market presence, our securitization volumes have risen significantly. However, the overall credit environment has weakened. Home price growth has stagnated, many consumers are facing financial stress, and numerous companies are not only slowing hiring but are actively cutting jobs. We will continue to employ a data-driven, model-based investment strategy to seek high returns while managing downside risk. Now back to Larry.

Thanks, Mark. To sum up, this was an excellent quarter for Ellington Financial on a number of fronts. We achieved earnings growth and meaningful portfolio expansion, and we marked a significant inflection point in evolving our financing base, all of which we think positions us well for continued earnings strength and dividend coverage in the quarters ahead. I'm pleased to report that this momentum has continued into the fourth quarter, with securitization activity remaining robust and origination volumes strong at Longbridge and at our non-QM loan originator affiliates, LendSure and American Heritage Lending. Of course, we've also been hard at work deploying the proceeds from our unsecured note issuance. We've used a portion of the proceeds to grow the investment portfolio by more than 5% in October alone, and we've used most of the remainder of the proceeds to pay down repo as planned. Our ADE generation power is very strong. So it's good that we have some ADE to spare going into the fourth quarter, as we expect to experience a modest near-term drag on ADE as we deploy the proceeds from our notes issuance. But even after we deploy those proceeds, we expect to realize additional, more subtle benefits from our notes issuance over time, as I'll now explain. The first additional benefit is through increased capital efficiency, which is a byproduct of replacing mark-to-market financing like short-term repo with long-term non-mark-to-market financing. Specifically, and consistent with our disciplined risk management approach, we maintain extra cash and capital reserves against our repo and other mark-to-market facilities to guard against potential market shocks. We can reduce those reserves when we replace repo with long-term unsecured notes, which frees up capital that can be redeployed into higher-yielding assets, thereby further amplifying long-term earnings potential. As an aside, similar benefits apply to our securitization financings. The second additional benefit from our notes offering is a much longer-term benefit. We view our shift toward a greater proportion of long-term unsecured and securitization financing and a lesser proportion of shorter-term repo financing as a fundamental evolution of our capital structure. This shift is fortifying our balance sheet, enhancing risk management, and supporting earnings stability. Including our existing $263 million of unsecured notes, nearly 20% of our recourse borrowings are now unsecured as of October 31, and we intend to increase that proportion over time. And as this evolution progresses, we expect that we'll see upgrades in our credit ratings, which should enable us to issue more unsecured debt at even more attractive economic terms, setting off a virtuous cycle. As a result of these multifaceted dynamics, I believe that our unsecured notes program will enable us to both build a more resilient balance sheet and expand our earnings power. As always, our goal is to deliver durable, high-risk-adjusted returns to shareholders across market cycles. Looking ahead, with conservative leverage, ample liquidity from our recent unsecured notes issuance, and a steady pace of securitizations, we believe that Ellington Financial is well-positioned to continue delivering strong and sustainable dividend coverage. And with that, let's open the floor to Q&A. Operator, please go ahead.

We will go first to Crispin Love with Piper Sandler.

First, just on the loan originator platforms. We've started to see a more conducive mortgage rate environment. Can you just discuss how this has changed valuations in your stakes and then overall operating performance, increased flow to Ellington? And then are there any other areas where you're looking to add capacity in other platforms or new platforms from an originator level?

Okay. JR, do you want to talk about valuations sort of generally speaking, how we value, and how the recent tailwinds are helping valuations?

Yes, sure. Thanks, Crispin. So the stakes are third-party valued twice a year, and then the other two times we adjust based on interim P&L. And the valuation providers are typically looking at, broadly speaking, three data points, kind of trailing earnings, forward earnings, and then multiples relative to the market. And so I think there are a few factors at play. There have been some publicized trades in the market at multiples to book, premiums to book. But at the same time, book value has built up because earnings have been so powerful. And in many of these cases, we're also getting distributions of those earnings. So we get to return cash and then redeploy it. And so I'd say that the strong earnings performance has reflected through higher book value, but it has also led to some more liquidity for these platforms and higher multiples. We have a table in our 10-Q where we go through the multiples of earnings that our valuations reflect, and they're not at the same levels as one particularly notable transaction that happened within the last couple of months. So at a premium to book, reflecting the earnings power of the platform, but not at the premium reflected in that transaction. So if that helps answer the question, I mean, earnings have driven book value, which has driven values, and just interim P&L is reflected in our mark-to-market as we capture the benefit of those earnings.

Yes. And then in terms of new products, I think, right, Crispin, was that your question is are we looking at adding stakes in any new products? Is that right?

Yes, that's right.

Yes, Mark, I’m not aware of any new products. We are considering a slight increase in servicing capacity, but I don't have any specifics. Mark, do you know of anything?

Yes. I mean the one thing I would say, Crispin, is that you are starting to see adjustable-rate mortgages taking an increasing share of the new origination market. I think for some originators in the agency space, it's as much as 10%. And if I go back to the early days in talk in 2015, 2016, 2017, that product was 100% adjustable rate. It used to be all 7-year ARMs, right? So we are starting to see some demand for adjustable-rate mortgages. And part of that is a consequence of the steeper yield curve where you can offer a little bit lower rate on a 7-year ARM than you can on a fixed rate. So that's a new product. That's one thing that we've been working with some of our affiliates and some other originators with.

And Mark, I found your comments on buying loans from banks interesting. Can you just dig a little bit deeper on that opportunity? Is it primarily commercial real estate loans? Is there any residential in there? And then just what types of banks are you dealing with? Is it more community banks or are there larger ones as well?

The two transactions I mentioned earlier were both residential mortgage loans, one of which turned out to be adjustable rates. These involved smaller banks rather than the large global systemically important banks. Many banks are looking to restructure their portfolios. This is particularly relevant in the Agency MBS market, where several banks still hold substantial portfolios of Fannie 2s and Fannie 2.5s, which have negatively impacted net interest margin. It's noteworthy that most mergers and acquisitions involving banks over the past couple of years have been followed by significant portfolio restructurings. If M&A activity rises, we can expect more banks to sell off loans they've been keeping in their portfolios for some time. The current market dynamics of lower yields, a steeper yield curve, and tighter credit spreads are increasing the value of certain loans on balance sheets, making it more feasible for banks to accept losses in favor of reorienting their portfolios. I anticipate this trend will continue if the rate environment remains as it is now.

And we will move next to Bose George with KBW.

This is actually Frank on for Bose. First question is on credit. You touched on weaker consumer, a little softness in the labor market, and negative HPA. Can you just elaborate on maybe what you're seeing within your portfolio and where you're seeing the best allocation of capital today?

Sure. We have the new slide that JR discussed, which clearly illustrates the credit performance of our loans in both the residential and commercial sectors. If we examine consumer spending by income levels, we notice that the weakness predominantly affects the bottom half, specifically the lower 50% of income levels. This trend is evident in subprime auto loans, lower FICO credit cards, and certain segments of the FHA and VA portfolios, which generally have lower credit quality compared to what is observed with Fannie and Freddie. On the other hand, higher-end borrowers continue to spend and exhibit strong credit performance, which is our primary focus. Additionally, I mentioned in the prepared remarks that there has been an increase in companies announcing layoffs. These layoffs, indicated by some corporations, may potentially affect higher-wage borrowers. Currently, credit performance remains robust, and we have made significant progress this year in addressing some delinquencies in the small balance commercial portfolio. However, I wanted to highlight this situation as it is on our radar and is an area of consideration for us.

Yes. And Mark, just to follow up on that, on Slide 15, I just want to reemphasize what JR said during the prepared remarks, which is that these are non-annualized. These are cumulative numbers. So our commercial mortgage loan business, bridge loan business, which goes back, gosh, 10 years at least, when you see 47 basis points of cumulative losses, that's over 10 years. So obviously, anything on an annualized basis will be much smaller. On the residential side, again, 13 basis points goes back many, many, many years. So we're really pleased with this performance. As Mark says, we've been laser-focused on FICO, which has really helped us. And of course, in RTL, all our loans have personal guarantees. So we feel very good about where we stand. We did have a couple of loans that we talked about on some relatively recent earnings calls that on the small balance commercial space, one of them has been resolved. One of them is actually now going quite well towards resolving, I would say, later next year, but things are looking good. So we really feel good about where we stand.

Great. You had a strong ADE over the past year. Do you anticipate any increase in the dividend level? Also, could you provide details on the drag you mentioned on ADE in the fourth quarter?

In terms of the drag, I believe JR mentioned that our overall cost of funds would be about 17 basis points higher, all other things being equal from our current position. As I mentioned, we have some solid ADE coming in. We still believe we will be able to continue covering the dividend. I want to clarify that we have no plans to lower the dividend. Right now, with a yield around 11, it's a good dividend. We just want to keep covering it as we have been.

And we will move next to Trevor Cranston with Citizens JMP.

A follow-up question on your comments on sort of the general credit backdrop and credit performance. Looking at the credit hedge portfolio, it looks like the size of the hedge positions declined somewhat in 3Q. So I was wondering if you could just maybe provide some commentary on how you guys are thinking about the risk of sort of spread widening and how you guys are approaching the credit hedge part of the portfolio right now?

Yes. The decrease in the credit hedge was more of a minor issue, as we had just priced that deal at the end of the quarter on December 31. With a significant amount of cash coming in, we decided to temporarily reduce the credit hedge. The hedge is meant for protection during market fluctuations, but the influx of cash reduced the immediate need for a larger hedge. However, I do anticipate that the hedge will increase as we invest that cash into high-yielding opportunities that are more aligned with overall market risks. Therefore, I see the recent decline as a minor moment in the larger context.

And then you talked about working towards deploying the capital from the debt issuance at the end of the quarter. Obviously, you guys have been able to do some issuance on the common equity side through the ATM plan as well. Should we think about the debt issuance sort of decreasing your appetite for common equity in the near term? Or is the sort of amount of issuance there small enough that it doesn't really move the needle enough to change things?

Yes, we can deploy capital fairly quickly. I want to emphasize that our ATM issuance has been beneficial, which is important for us to maintain. We've had a strong quarter for ATM issuance, and due to the various opportunities we've discussed previously, we're able to deploy capital efficiently. We have a range of strategies we can utilize, and we select the ones that seem most advantageous at any given time. We don't view them as alternatives; additional ATM issuance is currently beneficial and also supports our general and administrative ratios. There are many good reasons to continue that approach. Of course, we prefer to see our stock trading at a premium, and we want to avoid taking any reckless actions that might jeopardize that situation.

Yes. To add to that, we noted that in October, the portfolio increased by more than 5%. While we didn’t specify the exact amount, that represents about $200 million based on a $4 billion base. We raised $400 million and indicated that we are using more than half of that to pay down repo. Additionally, we achieved 12% portfolio growth in Q3. The key takeaway here reinforces Larry's point that we haven’t faced any issues with deployment and portfolio ramp in recent months. We also raised funds in the ATM during the quarter in a way that was favorable compared to where the book value per share ended up at $30.

And we'll move next to Timothy DeAgostino with B. Riley Securities.

On Longbridge and the proprietary reverse mortgage product, you had mentioned that it was like record volume origination. I was wondering, within that market, what does competition look like for you all?

Yes, there's not much competition. In the proprietary space, there are more HECM issuers. Longbridge is the second largest in this area and sometimes ranks first in terms of volumes by certain metrics. There are really two other competitors; one is a public company and the other is not. The challenge for others to originate this product is that they lack the capital base and the integrated outlet we have, particularly with Ellington Financial, which has funds to invest. Our originator can effectively generate this product for us. Consequently, there is definitely less competition in this area than in others, putting us in a strong competitive position. Our successful securitization has allowed us to offer better terms to borrowers, which has resulted in higher volumes for us. Better terms lead to increased volumes, and we hope this trend will continue.

Okay. Great. Yes. And then just quickly flipping to the credit portfolio. Quarter-over-quarter, it seems like non-QM was the biggest piece, like had the most investment quarter-over-quarter. I was just wondering what you're seeing in that market? And why do you like it so much?

Mark?

Non-QM isn't a single category; it encompasses loans for investors, owner-occupied loans, and various types of documentation from complete verification to bank statements and others. We have been involved in the non-QM market since 2014 when we made our initial investment in our first portfolio company, LendSure. Over the last 11 years, we have dedicated significant time and resources to enhancing our credit modeling, prepayment modeling, and strengthening relationships with both our affiliated originators, in which we hold equity, and other origination partners, where we may not have ownership but actively engage in discussions regarding underwriting guidelines. We have a dedicated team that visits originators regularly, conversing with underwriters and appraisers to establish best practices for loan origination. This has given us a profound understanding of the market, which has expanded as Fannie and Freddie have tightened some of their guidelines. Approximately 10% to 15% of homebuyers are underserved by the GSEs, which is central to the non-QM space. Recently, we have observed a general increase in FICO scores, which we view positively. We have maintained strict discipline regarding loan-to-value ratios, with most of our portfolio being under 70% LTV on average. The volume of non-QM securitizations has risen significantly this year, leading to increased liquidity and commoditization, which has attracted a larger pool of investment-grade bond buyers. All these factors have contributed to tighter spreads, allowing us to procure well-underwritten loans for borrowers with elevated FICO scores and securitize them at significantly tighter spreads than what was available in 2024. The increased market liquidity has led to greater assurance in execution, making this an asset class where we can achieve considerable scale, resulting in highly appealing retained investments in our portfolio. Thus, it is the combination of higher volumes, disciplined underwriting, and tighter investment-grade spreads that makes this sector particularly enticing for us.

And we will move next to Eric Hagen with BTIG.

We actually have a follow-up on the Longbridge portfolio. I mean when we think about the total upside potential in Longbridge, does it require more leverage to get to its target returns? And how do you think about the amount of leverage in that portfolio? And what's both objective and sustainable for leverage in that portfolio?

Thanks for the question. I believe additional leverage isn't necessary. Most of Longbridge's equity comes from its servicing, particularly the technology and servicing portfolio, which provides a high yield without the use of leverage—significantly higher than forward servicing. Regarding the proprietary reverse loans that appear on EFC's balance sheet, there will be some leverage while we're accumulating for securitization. However, post-securitization, when we keep only the residual interest, it doesn't require additional leverage. Although we consolidate those loans, which may make it seem like there's more leverage, this financing is long-term and not subject to market fluctuations. Essentially, we have a small retained interest that doesn't necessitate more leverage. More leverage would only be needed as origination volumes rise, which requires more warehouse financing, but that situation is more temporary.

Yes. Okay. That's really interesting. Going back to non-QM for a second, here. I mean, what's your perspective on convexity risk in the space right now? I mean, to your very point, it feels like so much progress has been made among brokers and loan officers. The asset class is higher quality, more transparent, more liquid. At the same time, I mean, we wonder how it would respond to lower rates, even meaningfully lower rates, and your ability to kind of backfill that portfolio if rates were to fall.

It was interesting to see that in the last remittance cycle, jumbo speeds increased significantly, while non-QM did not experience a similar rise. However, looking back to 2020 and 2021, non-QM loans demonstrated the capacity for rapid prepayment speeds, often exceeding 40 CPR. The current rate changes have certainly highlighted the importance of understanding and valuing prepayments. In a securitization, a large part of the investment is retained, which exposes us to prepayment risk. The decline in rates is also adding value to call options, allowing us to keep the ability to secure loans at par that were originally at 103, with the potential for them to be worth 105 after a rate shift. These call options perform well in a declining rate environment. Additionally, we have the excess spread component that is linked to quicker prepayment speeds. Thoughtfully managing prepayment risks through payment penalties, which about 30% of the market utilizes, is one of our key strengths. We have extensive knowledge in understanding the servicing portfolio acquired from Ellington, along with insights into IOs and inverse IOs, and how different borrower types respond to prepayment options. Prepayment speeds are indeed a priority for us. Regarding portfolio size, during a refinancing wave, the market doesn't shrink; borrowers usually replace existing loans with new ones. Often, when borrowers refinance, they might be cashing out and acquiring a slightly larger loan, which can lead to an increase in market size. We are concentrating on the prepayment risks associated with various loans and ensuring they are properly hedged across the yield curve. In terms of volumes, I anticipate a significant increase, generating plenty of opportunities.

Yes, I want to highlight that we meticulously model our approach. While we warehouse loans pending securitization, particularly non-QM loans that have negative convexity, we manage this by using short positions on TBAs and other interest rate hedging products against those loans. This creates a short position on a negatively convex instrument that balances the long position we hold. Additionally, if you look at Page 14 of the presentation, you'll see our interest rate sensitivity analysis. In our quarterly report, we extend this analysis to 100 basis points. Overall, despite the prepayment risk associated with certain IOs in our non-QM retained tranches, the negative convexity throughout the company remains well-controlled. Page 14 shows some negative convexity, but the potential declines in equity for both a 50 basis point drop and increase are quite modest. This is an area we analyze very closely, drawing on over 30 years of experience.

And we will move next to Doug Harter with UBS. It's actually Marissa on for Doug today. Just one for me, more broadly on the reverse mortgage space. How are current market conditions, notably the outlook for moderating HPA and the evolving regulatory environment, how are they impacting your outlook for the ongoing opportunity in the space?

Sure. On the regulatory side, there isn't much happening. There had been discussions about some improvements labeled HMBS 2.0, but that seems to have stalled. So, the regulatory situation remains quiet. Home Price Appreciation is important for us. In our proprietary reverse securitization, we retain the residual, which means we are exposed to long-term home price appreciation. We mentioned in our prepared remarks that due to some stalling in home price appreciation, we adjusted down the valuation on those retained pieces in our proprietary reverse mortgage securitizations. However, this effect was limited and offset by many other factors within the portfolio. We monitor this closely as it affects the portfolio's value. It's important to note that there is a significant cushion; all reverse mortgages, including proprietary ones, are originally issued at very low loan-to-value ratios. Thus, we are not significantly exposed to shorter-term home price appreciation but rather to ultra-long-term home price appreciation. In the short term, the loan-to-value ratios are well below 50%.

That was our final question for today. We thank you for participating in the Ellington Financial Third Quarter 2025 Earnings Conference Call. You may disconnect your line at this time, and have a wonderful day.