Ellington Financial Inc. (EFC) Earnings Call Transcript
February 26, 2026
Earnings Call Speaker Segments
Thanks. Prior to getting started, please note that during this conference call we might make forward-looking statements as defined under the safe harbor provisions of the Private Securities Litigation Reform Act of 1995; such statements aren't historical by nature and carry risks and uncertainties described in our annual and quarterly reports filed with the SEC. Results that actually occur could vary materially from what those statements suggest, and therefore they shouldn't be viewed as forecasts of future events. No duty is assumed by the company to revise these forward-looking statements. With me on the call are Larry Penn, Ellington Financial's Chief Executive Officer; Mark Tecotzky, Co-Chief Investment Officer; and JR Herlihy, Chief Financial Officer. A presentation for our fourth quarter earnings conference call can be found on our website, ellingtonfinancial.com. Today's call will follow along with that presentation, and every statement and reference to figures is subject to the important notice and end notes located at the back of the presentation. Having said that, I'll turn things over to Larry.
Thank you, Alaael-Deen. Morning, everybody, and I appreciate you being with us today. I'll start on Slide 3. Ellington Financial wrapped up 2025 strongly, finishing a year marked by steady, robust results, expansion of the portfolio, and improvements to our liability structure. During Q4, carrying forward the momentum we'd built over the entire year, our adjusted distributable earnings once again came in well above what we paid out as dividends. We grew our investment portfolio further, and we kept making our balance sheet stronger. In Q4, GAAP net income came to $0.14 per share, while ADE reached $0.47 per share, again surpassing the $0.39 per share we distributed as dividends. These results please me even more because we absorbed some near-term drags as we put to work the money raised through our unsecured notes offering and our RTL securitization, both of which I'll get into later. Driving our results were outstanding performance and execution within our loan origination and securitization platforms, with our Longbridge segment contributing at an outsized level once again. Excellent credit results throughout our residential and commercial loan portfolios also bolstered our numbers. Early in October, we closed a $400 million unsecured notes offering — the biggest we've done so far — which represents a major advance in how our capital structure has evolved; we were happy with how it was executed and with the strong investor interest, and we're encouraged that the bonds keep trading today at a meaningful premium. As we'd said we would, we directed part of the offering's proceeds toward paying down short-term repo financing. During the quarter, we also took advantage of securitization markets that remained strong, finishing 7 more securitizations across the quarter. Most notably, in November we closed our first securitization of residential transition loans. That deal has a revolving structure. So when our securitized RTL loans get repaid, we're effectively able to redeploy the securitization debt to fund our pipeline of newly originated RTLs. After the year ended, we closed our first securitization of agency-eligible mortgage loans. Through that deal, our EFMT branded securitization shelf has grown to span 5 distinct residential loan sectors. This lets us term out financing for every one of our major residential loan strategies on a nonrecourse, non-mark-to-market basis, taking the place of repo financing and further boosting balance sheet resilience along with capital efficiency. Since we launched our RTL strategy in 2018, RTLs have delivered consistently robust returns on equity for us; after 2022 and 2023, though, as credit spreads grew wider and the yield curve flipped, securitization economics for RTL were generally unattractive next to plain repo financing. That equation has now changed. With the yield curve back to normal, with securitization spreads fairly tight, and with rating agencies viewing the product more constructively, securitization economics for RTL are now better. What that produces is appealing long-term non-mark-to-market financing, which helps us create high-yielding retained tranches that lift EFC's overall portfolio returns. Regarding the approach we take with loans that qualify for Agency treatment, we launched this effort only a year ago, bringing in roughly $250 million worth of such loans during the latter portion of 2025. That decision fit within a broader pattern we've discussed on earlier earnings calls — namely, shifting toward areas where the government-sponsored enterprises are steadily pulling back, opening space for private investors to come in and earn appealing returns for the risk taken. In our view, the Agency-eligible space — above all those segments where we believe loan-level price adjustments are set too high — offers EFC a potentially sizable opportunity over the long haul, particularly with the clear synergies tied to our underwriting capabilities, where we source deals, and the strength of our securitization platform. We're also of the belief that prospects in the Agency-eligible arena will keep improving, since policymakers seem more and more open to private capital playing a bigger role. Turning now to EFC's balance sheet: our ongoing priorities remain fine-tuning our capital structure and strengthening our resilience; during the fourth quarter, because of our unsecured notes offering, we nearly doubled the share of total recourse borrowings made up of long-term, non-mark-to-market debt, and our unencumbered assets grew by roughly 45%. In addition to those balance sheet improvements, we kept pursuing appealing investment opportunities during the fourth quarter. We put part of the notes offering proceeds to work in new investments, growing our portfolio 9% even once all securitization activity is factored in. Strong origination and acquisition volumes keep benefiting our portfolio across non-QM loans, Agency-eligible loans, closed-end second lien loans, proprietary reverse mortgages and commercial mortgage bridge loans. As the year closed, we had mostly put the entire notes offering proceeds to work, setting the portfolio up for sustained earnings strength as the new year began. That momentum has continued into 2026; during January, with our common stock priced above book value per share, we issued common equity on a basis that was accretive, net of every deal cost. Beyond being accretive, the issuance was sharply focused. We set the offering's size to produce exactly the proceeds required to redeem our priciest tranche of preferred stock — our Series A preferred stock. And right after the offering closed, we announced that tranche's redemption. The coupon on our Series A preferred stock stood above 9%. So beginning tomorrow, once the mandatory 30-day notice period concludes and that tranche's redemption wraps up, our common shareholders will instantly enjoy the advantage of a reduced overall cost of capital. In the meantime, we'll keep watching the preferred equity market, considering a possible refinancing of that capital down the road and at a lower cost; with that, please go to Slide 5, and I'll hand the call to JR so he can take everyone through our financial results in greater depth. JR?
Thanks, Larry. Good morning, everybody. During Q4, our GAAP net income came in at $0.14 per common share, calculated on a fully mark-to-market basis, while ADE was $0.47 per share. Slide 5 breaks down portfolio income by strategy: $0.35 per share from Credit, $0.04 from Agency, and $0.15 from Longbridge. Then Slide 6 shows ADE by segment: $0.61 per share from the Investment Portfolio segment and $0.13 from the Longbridge segment. Within the Credit portfolio, net interest income rose quarter-over-quarter, and we also produced net realized and unrealized gains on non-QM retained tranches along with forward MSR-related investments. Those results were partly offset by net realized and unrealized losses on certain other credit hedges, plus losses on residential REO. We keep enjoying outstanding earnings contributions from our affiliate loan originators, together with robust credit performance throughout our loan businesses — this includes sequentially lower 90-day delinquency rates and ongoing low life-to-date realized credit losses in both residential and commercial loan portfolios, as Slide 15 shows. In the Agency strategy, reduced interest rate volatility and narrowing agency yield spreads generally helped our portfolio during Q4; we delivered strong results, led by net gains on both long Agency RMBS and interest rate hedges. The Longbridge segment also posted another outstanding quarter, with positive contributions from originations and servicing alike. Origination profits came from sequentially higher origination volumes, sustained strong origination margins, and net gains tied to 2 proprietary loan securitizations finished during the quarter. On servicing, steady base servicing net income, robust tail securitization executions, and a net gain on the HMBS MSR Equivalent all added positively. Net gains on interest rate hedges contributed further to results. Now let's move to portfolio changes in the quarter. Slide 7 reflects a 15% rise in our adjusted long credit portfolio, reaching $4.1 billion versus the prior quarter. Non-QM loans, Agency-eligible loans, closed-end second lien loans, commercial mortgage bridge loans, ABS, and CLOs all grew. Our portfolio of retained RMBS tranches likewise increased, in that instance reflecting the securitizations we completed during the quarter. Offsetting these increases in part was the effect of loans sold in securitizations. Our short duration loan portfolios kept returning capital at a healthy pace. For our RTL, commercial mortgage, and consumer loan portfolios, we collected total principal paydowns of $207 million during Q4, equal to 12.7% of the combined fair value of those portfolios entering the quarter. Slide 8 shows our total long Agency RMBS portfolio dipped modestly to $218 million. Slide 9 illustrates that our Longbridge portfolio fell by 18% to $617 million, since continued strong proprietary reverse mortgage loan origination volume was more than offset by the completion of 2 securitizations. Please turn next to Slide 10 for a summary of our borrowings. As of the thirty-first of December, the weighted average borrowing rate across our recourse borrowings fell 32 basis points, landing at 5.67% in total, since the effects of reduced short-term rates and narrower repo spreads outweighed what a bigger share of unsecured notes contributed. In addition, we extended the duration on certain of our bigger warehouse lines, and because of that, the weighted average remaining term on our repo rose 38% versus the prior quarter, reaching close to 9 months, all of which you can see laid out on Slide 24. Comparing this quarter with the last, net interest margin for our credit portfolio dropped 28 basis points, as weaker asset yields outweighed a reduction in the cost of funds. Our average asset yield came down, though the sole cause was a greater share of our assets being loans held in warehouses awaiting securitization. That bigger warehouse portfolio came about from deploying the proceeds of the notes offering. NIM on Agency fell 9 basis points, pushed lower by a decline in asset yields. On December 31, our recourse debt-to-equity ratio stood at 1.9:1, a slight rise from 1.8:1 on September 30. As was said before, during the quarter we issued $400 million of unsecured notes, part of which took the place of repo borrowings. Even so, the rest of the proceeds, deployed together with incremental borrowings into new investments, outweighed the deleveraging effect of repo paydowns, securitizations and greater total equity, producing a small net rise in the overall leverage ratio. That same dynamic pushed our overall debt-to-equity ratio up to 9:1 from 8.6:1. As Larry noted, our balance sheet metrics improved substantially over the quarter. Comparing quarter to quarter, among our total recourse borrowings, long-term non-mark-to-market financings climbed to 30% from 17%, while unsecured borrowings climbed to 18% from 8%. Unencumbered assets likewise expanded substantially, up 45% to $1.77 billion, roughly 90% — 95% of total equity. Going forward, we anticipate carrying on this move toward a bigger share of unsecured, non-mark-to-market and longer-term financings by issuing more unsecured notes, doing more securitizations and replacing our priciest repo borrowings. We see this shift as a basic transformation of our balance sheet that is improving risk management and earnings stability and that we hope will additionally bring about stronger credit ratings for EFC and lower borrowing costs over time. As I said last quarter, we chose the fair value option on our notes, just as we have for our other unsecured debt, and we mark to market through the income statement. Consequently, we expensed every associated deal cost in October instead of amortizing those costs across the life of the notes. And since credit spreads tightened during the quarter, we also booked an unrealized loss on the notes for the quarter. These nonrecurring items, along with some short-term negative carry while we await full deployment of the new note proceeds, amounted to a significant drag on our GAAP earnings for the quarter. Book value per share came in at $13.16 when the year closed, and our economic return during the fourth quarter reached 4.6% on an annualized basis. So with that, I'll hand things over to Mark.
Appreciate that, JR. EFC just wrapped a very strong quarter. Our playbook — originate loans, then securitize them — keeps getting executed, and in Q4 we closed 7 securitizations spanning multiple loan categories, with that same momentum rolling into 2026. During 2025, we pushed our reach far past the non-QM space where we began. These days our securitization platform covers non-QM, second liens, reverse mortgages, residential transition loans, and Agency-eligible loans. Over time, EFC has become far more adept at squeezing out profitability and controlling risk through every stage of a loan's life, starting with the purchase commitment and ending with the securitization exit. To start, we generate a levered return while a deal is ramping, and our aim is a gain on sale profit to the securitization trust, hedging execution risk the entire way. After that, when securitization happens, we focus on building high-yielding retained investments while expanding our growing portfolio of call options. Done right in a cooperative market, this becomes a virtuous cycle, accretive to earnings at every step and a major reason behind the steady results we've posted over time. One more advantage of our sizable securitization platform: it lets us offer steady, competitive pricing to our origination partners and affiliated originators across a wide array of loan types. On top of that, the rising value of our stakes in those affiliated originators keeps delivering strong results for EFC, both in the quarter and across 2025. Yet the asset side of the balance sheet wasn't our only area of productivity. As Larry and JR noted, we're enthusiastic about what being a Moody's and Fitch rated bond issuer means for EFC over the long haul. Pairing the substantial non-mark-to-market financing we've assembled as an active securitizer with our newest bond issuance, which drew very broad institutional participation, is steadily shrinking our reliance on short-term mark-to-market repo financing. I'm not suggesting repo is a flawed financing tool. It isn't. Repo markets worked exceptionally well across 2025, and during the fourth quarter we managed both to extend term and to push our repo financing spreads down even further. That continued compression in financing spreads has mattered for protecting earnings while asset spreads have narrowed; we've also secured tighter spreads on the investment-grade bonds we sell in securitizations, which usually make up more than 90% of any given deal. Several notable government policy announcements arrived this past quarter and across 2025 that matter to EFC. The $200 billion of GSE MBS purchases was likely the headline item. I'll skip the details since there aren't many, but I will note this isn't quantitative easing. Unlike QE, it probably won't meaningfully cut duration or negative convexity in the market. And crucially, bank reserves aren't created by it. The effect has been to establish a lower bound for Agency MBS spreads, and consequently for other AAA-rated mortgage bonds such as non-QM, second lien, and Agency-eligible AAA tranches. Yet what might matter more is that we're navigating a period marked by increased policy uncertainty — potential limits on institutional buying of single-family rentals, GSE reductions, LLPA changes, and mortgage insurance premium cuts are all under consideration, and each one carries consequences for prepayment speeds, for how appealing private label is relative to GSE execution, and possibly even for home prices. Our attention has been on carefully weighing these unknowns and aligning the portfolio with them. Net portfolio growth during the fourth quarter, as Slide 4 shows, came in strong at roughly $400 million, and that's even after factoring in our robust securitization volume; it speaks to years of systematically extending our capabilities to source a wider mix of loan products from a larger group of sellers and in a more automated way, which our loan portal makes possible. We take pride in the technology we've rolled out that lets partners sell us loans easily while we keep cultivating mutually beneficial ties with originators. Competing purely on price isn't our aim; instead we set ourselves apart via service quality and inventive loan programs that adapt to shifting markets. This quarter didn't go entirely as we'd hoped. Bank loans have faced some widely reported difficulties, and our CLO portfolio, though small, was a slight drag. The RTL strategy likewise fell short, hurt by securitization costs and REO workouts. Delinquencies in that area are still very much under control; in fact, January brought strong resolution results. We additionally recorded minor losses in CMBS and ABS, which to my mind are idiosyncratic instead of systemic. And since these sorts of air pockets appeared broadly throughout credit-sensitive markets in Q4, we've really done quite well. If anything, such dislocations are opening up opportunities, and we'll aim to add securities where our analysis shows the price decline far exceeds any shift in fundamental value. Going forward, credit must stay on our radar. The housing market is displaying somewhat wider signs of weakness compared with a year ago, and a growing number of borrowers are struggling to remain current. We've maintained substantial credit hedges, as Slide 20 indicates, and we keep putting money into our technology and sourcing to expand our loan origination footprint, which has been a major contributor to our returns. Now I'll hand it back to Larry.
Thanks, Mark. For Ellington Financial, 2025 mattered a great deal. As I wrap up, I want to walk through our accomplishments and explain how they set us up for 2026. There are 5 buckets I'd put 2025's wins into. To start, adjusted distributable earnings covered our dividend in all 4 quarters of 2025, which makes 6 quarters in a row of coverage; that steadiness counts for a lot when markets swing the way they have, and it speaks both to how durable our earnings engine is and to what diversification brings us. Next, we made our liability structure considerably stronger. Across the year, we finished 25 securitizations, versus only 7 during 2024. We brought $400 million of unsecured notes to market and laid the groundwork for additional notes deals down the road. Terms on our existing secured financing lines got better, and we brought in a number of appealing new facilities. All of that together did more than fund portfolio growth — it drove a significant, intentional shift in how our funding is composed, making it sturdier, more adaptable, and a better fit for our longer-term goals. Third, our loan originator affiliates turned in outstanding results. Origination volume climbed substantially, they picked up market share, and they delivered superb earnings contributions to Ellington Financial's bottom line. The vertical integration we've built keeps handing us a real competitive edge, fueling loan sourcing, enabling securitization scale, and boosting our capacity to earn. Fourth, realized credit losses have stayed remarkably small, which says a lot about our underwriting discipline and how deep our asset management capabilities run. The delinquent inventory we hold is limited in size and is working out well. Keep in mind, we mark-to-market through the income statement. So for any loans we anticipate resolving below par, that hit to income and book value per share has already been taken. And fifth — the heart of our growth story — we grew our portfolio by close to 20% year-over-year, reaching almost $5 billion, all while staying disciplined on credit and risk management. That expansion shows both the return on our technology initiatives and the benefit of new strategic equity stakes paired with forward flow agreements. At Ellington, the flow we're seeing through our residential loan origination portal, launched only 12 months ago, sits at roughly $400 million per month right now and keeps climbing, particularly as our roster of sellers grows more diverse. Our loan portal's success shows vividly how Ellington's proprietary technology can scale EFC's sourcing footprint, sharpen underwriting efficiency, and reinforce EFC's vertically integrated model. Rounding out our investments in <<inaudible>> Okay. My apologies for that. To be safe, let me back up and redo the fifth of our achievements. All right. And fifth — the heart of our growth story — we grew our portfolio by close to 20% year-over-year, reaching almost $5 billion, all while staying disciplined on credit and risk management; that expansion shows both the return on our technology initiatives and the benefit of new strategic equity stakes paired with forward flow agreements. What we're observing with the residential loan origination portal at Ellington, introduced a mere 12 months back, is roughly $400 million each month at present, and that figure keeps climbing, particularly as we keep expanding our varied group of sellers. A strong illustration of how proprietary technology at Ellington can extend EFC's sourcing reach, boost underwriting productivity, and reinforce EFC's vertically integrated approach is the achievement of our loan portal. Alongside our technology investments, during 2025 we brought in 2 additional strategic loan originator equity stakes, every one accompanied by forward flow agreements delivering premium recurring loan volume. These technology and strategic efforts jointly served as major catalysts for our portfolio expansion in 2025, and we anticipate they'll sustain that momentum into 2026; in fact, regarding strategic efforts, I'm happy to share that we've now signed a contract to purchase a modest residential mortgage servicer and are waiting on regulatory approval. Upon closing, this deal will bolster our vertical integration further by shifting additional servicing capabilities internally, particularly for delinquent assets. Though constructing the platform and crafting the servicing protocols will require time, I'm confident this purchase will eventually give us improved command over our servicing results and bolster our capacity to oversee our loan portfolios through varying market cycles. Our goals for 2026 are evident. We're concentrated on expanding our share of the loan origination market while preserving robust credit performance, which, alongside our securitization platform, ought to fuel methodical portfolio expansion. I'm additionally happy to share that 2026 has begun exceptionally well. We project that EFC produced an economic return of about 2% during January, with loan production and portfolio growth staying robust, notably within our non-QM, commercial mortgage bridge, and reverse mortgage loan operations. Across EFC's history of almost 20 years, I think we've reliably shown prudent stewardship of shareholder capital and a readiness to move opportunistically when market conditions prove advantageous. Our choice to redeem our Series A preferred stock via a focused common stock offering mirrors that strategy; we examined numerous alternatives, among them refinancing our Series A preferred with fresh preferred equity; yet because of the enduring wide pricing observed in the preferred market, we determined the selection was obvious. Our common stock offering drew more than 2.5x oversubscription from institutional orders by themselves and was carried out efficiently, highlighting robust market backing for the deal and its reasoning; in sum, I think the achievements of the past year, broadening our loan sourcing, securitizing often and efficiently, reinforcing our liability structure, and refining our capital base, all paired with our careful risk and liquidity management, put Ellington Financial in position to produce resilient earnings and steady dividend coverage over time and through varying market environments. Our team merits considerable recognition for all the effort they've invested to help bring this about. With that, let's open the floor to Q&A. Operator, please go ahead.
I was hoping you might expand a bit on why you chose to acquire the servicer. Does that shift your willingness to purchase the assets? For instance, would MSRs be of interest to you? Or is it primarily a means of fine-tuning the loan portfolio you currently hold?
Mark, would you like to handle that?
Okay, so I'd say a handful of factors were at play here. The first thing is the huge amount of consolidation that's happened among servicers. Mr. Cooper acquired Rushmore, and then Rocket agreed to acquire Mr. Cooper. As a result, the large-scale servicers have gotten larger, and there are fewer high-touch servicing operations available to assist borrowers when they encounter a problem, lose income, or fall behind on a payment. We feel it matters for us to produce the strongest risk-adjusted returns, and we have what you could call top-tier protocols and top-tier technology for managing things such as later-stage collections. So the goal here isn't to scale up a cheap Fannie servicer where everybody pays by ACH and it's all about enormous efficiencies. It's simply an acknowledgment that given all the consolidation among servicers, there aren't many solid servicing options left for helping borrowers who run into any kind of difficulty. So building this out will take some time; as Larry said, it's a small servicer. I'd say the resident expertise — the inherent expertise across Ellington Financial and, more widely, Ellington Management Group regarding the best ways to service and the best ways to help borrowers facing a challenge — runs really, really deep across our NPL and RPL team, our non-QM team, and our RTL team. Several individuals here bring multiple decades of experience. Our intention is to develop the technology alongside the servicer. Then, as Larry noted in his prepared remarks, we'll draw on our knowledge — the knowledge we already have about servicing loans — to devise top-tier protocols and a top-tier workflow, ensuring we achieve the strongest possible results on loans while borrowers receive the finest servicing experience available. We simply determined that reaching that goal required us to build it ourselves. In our view, the marketplace doesn't offer enough of those capabilities for us to assume we could obtain those outcomes without handling it internally.
Understood, that's all clear. Regarding what's possible here — do you see this being applied only to the Ellington portfolio, or might third-party clients also make use of it? And one quick point of clarification: does EFC hold ownership of this entity, or will ownership sit at the broader Ellington level?
EFC holds it. Mark, would you like to address the remaining piece?
Yeah. So the way I look at it, what we need to do at this point is develop the tech, develop the protocols, so this servicer becomes what we consider top-tier, and prove it to ourselves by watching its servicing metrics move to delinquency rates and how you handle borrowers who encounter a bump in the road; plus there are all sorts of champion challenger metrics that people rely on to assess servicers. And first we have to build it, and we have to bring it to the point we want and to how efficiently it runs. After we've done that, I definitely believe other investors in the mortgage space will see that there isn't much capability available today for later-stage collections and may very well want to take advantage of what we're creating.
Brandan here, filling in for Eric. Could you talk about what the environment looks like today for using repo on the retained tranches we still hold from securitization tied to non-QM and RTL? Have the terms gotten better? Also, are there situations in which you might be able to put additional leverage against those retained tranches? And how would the returns shape up?
Mark, would you like to handle that? Or would you prefer that I do it?
Absolutely, I'm able to address that. Yes. So, regarding repo — as I noted during my prepared comments, the repo market performed quite smoothly over the course of this year, correct? You saw the Fed funds rate decline gradually, and at that point the Fed put some reserves into the system because they felt bank reserves were running low. As a result, repo worked very effectively. When it comes to financing spreads on retained tranches, I'd say they're on the lower side. Those retained tranches, in my view, carry a kind of built-in leverage, correct? Either you're working with small tranches sitting at the bottom of a securitization's capital stack, or you're working with tranches where the bulk of cash flow derives from excess spreads; because of the nature of those investments and the leverage they already carry, their prices swing quite a bit. So I don't envision us wanting to layer on additional leverage to those tranches. Our approach to running this company is quite conservative when it comes to repo. What I mean by that is our internal haircuts sit meaningfully above the advance rates that our repo lenders would extend to us. For instance, we might hold loan strategies where lenders would advance 90%, 95% cents on the dollar against the loan. Internally, our thinking is that we'd rather borrow beneath that level to ensure we've got cash available in case of spread volatility and similar situations. So we retain ample capacity to increase leverage should we choose to. Given the price volatility that's inherent to retained tranches, I don't feel that's likely to be an area where we'd want to add it.
Yes. And if I could just add...
Those assets on the road — yes.
I'd like to jump in on that point — so, when you consider it, yes, a portion of that does involve financing. But when you look at where we're headed over the long term with how our financing and liability structures are set up, you know? Take our unsecured notes in particular — we completed a deal at 7 3/8s, and those are currently trading in the high 6s. Our hope is that this trend continues downward. Consider that — those are our unsecured notes; and beyond that, I'd point to our preferred equity as well, which I'd describe as being among the primary financing tools for that. Obviously, those don't carry costs as low as what repo does. But keep in mind, this is — what we're after is that virtuous cycle. Like Mark mentioned, when we're sitting well into the teens on an unlevered yield alone and we're funding ourselves at 6%, 7%, or even somewhere around 8% on preferred, it takes only a modest amount of leverage from those instruments to reach 20%-plus ROE; so a great deal of leverage really isn't necessary. And regarding the sort of repo we mentioned paying down when we completed that notes offering during the fourth quarter in October, it was — that represents precisely the category of higher-cost repo we'd prioritize paying down first.
During the quarter, Mark brought up what the government has announced policy-wise and how that might affect Ellington. Could you go a bit deeper into how you specifically are tackling the Agency-eligible space, given that alterations to LLPAs or G-fees might be coming — possibly at some point this year — and how does that work its way into pricing, plus the convexity risk around prepayments for loans of that kind?
Absolutely. Trevor, excellent question. Here's the thing — none of us can predict the future. We've got plenty of tools at our disposal to keep an eye on policy shifts that might happen. And with the current administration, a great many possibilities exist. So the way Agency-eligible investor loans and second homes started getting securitized through private label — you've witnessed that pattern coming and going across the past 5 years. It has definitely sped up somewhat. The explanation is that when you add loan level price adjustments to G-fees, the total sits so far above what losses are expected to be in those markets that private label has ended up offering superior pricing for that credit risk, and on the whole it means better execution for the people originating those loans. So that's the direction the flow is heading. That said, should LLPAs shift dramatically, the arithmetic might swing back toward Fannie and Freddie, and you might witness volume drop off there. As of today, I'd tell you the execution gap isn't even close. Therefore a minor LLPA adjustment, in my view, won't shift the needle. The bulk of that volume will still land in the higher LLPA category rather than the extremely low LTV loans, and it will still head to private label. But we must keep watching, which is exactly why I chose to include in the prepared remarks that we're already taking certain steps, reacting to the pricing structures that exist in the market today. And should those market pricing structures shift, the economics can shift, the opportunity set will shift, and so will what we do. So as things stand, swinging the pendulum back toward GSE execution for those loans would require a pretty substantial LLPA and G-fee change, though it absolutely could occur. That's something we're able to monitor. Hedging it isn't possible, and controlling it isn't possible either. Then there's the prepayment angle to consider, correct? If an LLPA change is large enough, similar to how prepayment models get discussed in terms of elbow shifts, a change in LLPA amounts to an elbow shift — certain loans essentially become more refinanceable. So whenever we assess premium investments in that area, or IOs, or the floaters we've been doing more of — like on the deal we completed, we had sizable floaters and you build an inverse IO — you've got a prepayment model, and that prepayment model is roughly calibrated to where the market stands currently. And that prepayment model has no awareness that a G-fee cut or an LLPA cut might occur down the road. So to account for that risk, here's what we do. In those sectors today, we're pushing speeds up beyond what a merely calibrated model would give you at present. So we believe that's sufficient for the risk. That's something we aim to manage and factor in, weighting it properly by probability, and we examine how current returns are distributed. That said, I'd note we aren't alone in the market in treating this as an elevated risk. So you're able to raise your prepayment speeds in those sectors and still purchase assets at market levels with highly appealing returns. It's not like other market participants are overlooking this risk or refusing to see it when setting prices.
Frankie Labetti here, filling in for Bose. Origination activity gave you yet another solid quarter. Could you talk about what competition looks like right now and where margins stand so far this year?
Mark, how about you take the forward side? I'll handle the reverse side.
Okay, so looking at the forward space — that's non-QM, second lien, and Agency-eligible investor — my take on how the competitive landscape looked in 2025 is that yes, there was competition, though I wouldn't go so far as to call it cutthroat. When we're thinking through the loan-level pricing we release on a daily basis, what's on our mind is where we'll purchase bulk packages, whether from affiliated originators or simply originators that we partner with. We had the ability to set pricing at certain levels — and I brought this up in my prepared remarks — where in my view we'd have a gain on sale when securitizing those loans, factoring in placing the retained pieces at loss-expected returns that we believe will be quite accretive for ADE. So the market is competitive at all times, but over my career I've definitely witnessed periods when pricing pressure felt cutthroat, felt like you simply weren't being paid for the risk you had to shoulder. That was not what happened in 2025, so my sense is that while it was competitive, you could still set pricing with margin and hold onto assets at appealing yields.
Appreciate it. Now, if I could address the reverse mortgage area. Let's break that down into two categories, okay? You've got HECM originations — that's the FHA-insured offering — and then you've got the proprietary side. Rates remain elevated compared with 2020 and 2021, so HECM volume across the industry has really been pretty flat lately. And I'd say there's a great deal of — if rates were to come down, we'd have substantial room for growth. Longbridge has, at various points, ranked as the number one, number two, and consistently among the top three originators in HECM. Competition does exist. On margins — the volumes are essentially what they are, so to speak, in that segment. Market share clearly plays a role. But when it comes to margins, the gain on sale margins, those are largely determined by marketplace spreads — specifically where you're able to sell the Ginnie Mae certificates, the HMBS. And that was definitely — it's been a favorable factor, margins have been strong, particularly during the back half of last year; and — though it will depend heavily on spreads; but at present, margins look great and volumes are, I'd say, fairly stable with modest growth as we've picked up market share. Over on the prop side, there's even less competition. Competition does exist in prop; and there too, the gain on sale margin will be heavily influenced by securitization exit spreads. And we've observed — I mean, with the most recent deal we completed, we achieved record low spreads on our AAAs. So provided securitization spreads stay tight, which, as I mentioned, we just hit a record low spread on our latest deal, the gain on sale margins in that area, I believe, will remain excellent. And volume there keeps growing as I think the products, the proprietary offerings, are broadening; we're constantly making adjustments to them. And we feel very confident about volumes there continuing to rise for Longbridge.
Excellent, that's really useful. Next, I'd appreciate hearing what you think about possible revisions to bank capital rules and whether that might lead banks to participate more actively.
That's a fascinating point. Mark here. Virtually every mortgage analyst with substantial experience, backed by a capable team and with data at their disposal — essentially the entire universe of mortgage experts — anticipated far greater bank purchasing during 2025 than what actually materialized. Then in Q4, if I recall correctly, banks cut back on their CMBS positions along with their pass-through positions for the first time in quite a while, correct? What they've been doing instead, given those large negative swap spreads, is simply acquiring treasuries and funding them in matched fashion through swaps, right? So bank buying has been limited, and current pricing on pass-throughs, or spread levels, is now narrower than it was through the bulk of 2025. Perhaps these capital regulations will shift that. I simply can't say. It's definitely conceivable that they might hold onto a greater number of loans, right? Some of that has already been happening, particularly with adjustable rate loans, <<inaudible>> loans. Even so, bank backing for the mortgage sector during 2025 was, frankly, surprisingly weak. I'm aware that certain of these regulations are designed to push banks toward deeper participation in the servicing business, right? That strikes me as a plausible move for them. Yet when it comes to servicing's major participants and the sizable deals, the sizable sell-offs and sizable purchasers, the nonbank segment has dominated for quite some time. So we'll need to wait and observe.
So I suppose once we hit the beginning of '26, it would be helpful if you might share a few of the top objectives, or what management has front of mind, and how you plan to get there — recognizing that bringing the mortgage servicer into the fold leads to more long-term financing. Though perhaps inside the portfolio, be it through allocation or in the capital stack, by deploying additional cash to repurchase preferred or along those lines; it would simply be helpful to hear a few items you're all aiming to get done in '26 that sit near the top of your minds.
Mark, I'll take the capital structure portion, and after that you can discuss what we're considering from the standpoint of portfolio allocation, perhaps. So on — yeah, regarding the capital structure, well, we recently redeemed that preferred; we hold a second preferred that will start floating at some stage and turns callable at that stage. Naturally, there's a possibility we might call that one too; its spread is somewhat narrower than the previous one. Still, we possess plenty of optionality once that occurs. Provided we believe our marginal deployment of that capital beats the coupon on the preferred, there's no rationale — there isn't any genuine urgency to call it; yet it's definitely something we'd weigh at that moment. And as I said, we'll also keep watching the preferred market. The prints we observed from certain peers regarding where they issue preferred didn't sit well with us. It didn't sit well with us in the sense that — we felt issuing there wasn't suitable for us. Yet if an opportunity emerges, we could certainly seek to swap out the preferred we redeemed with a preferred of probably comparable size. I believe that if you examine our capital structure at present — and there isn't much rigorous methodology behind this — I think the majority of companies would likely view a marginally larger share of the equity base held in preferred as more standard for this sector; so I think that's something we'll keep an eye on across the year. And then certainly, I think if we require the capital — and I noted that our unsecured notes, the ones rated by Moody's and Fitch that we issued last — during the fourth quarter, early in the fourth quarter, have tightened. Naturally, we'd be pleased if they kept tightening and we might come to market with another offering for sure later in the year, we'll see.
Absolutely. Mark, let me add something here for a second — this is JR. Tim, I appreciate the question, and it's a solid one. Larry, I think, walked through the 5 buckets of what we got done in 2025. Those 5 categories, I'd argue, tie in very directly to what you're asking about for 2026. First, we cover the dividend with ADE while keeping earnings steady and robust. Second, we reinforce our liability structure — Larry spent a fair amount of time on those efforts. Third, we back our originator affiliates, since gaining more market share is truly central to how we perform and grow. Fourth, we work through delinquencies. As we mentioned, delinquencies came down quarter-over-quarter. We're making substantial headway on cleanup, so performers stay focused on that same theme. And fifth, we keep expanding. Looking at the figures, our portfolio holdings stood at nearly $5 billion at year-end. That figure was $2.5 billion a bit over 2 years back, and leverage actually fell across that same stretch, going from 2.3 to 1.9 — so we achieved that growth without leverage rising. So as we look ahead to 2026, I don't want to simply call it more of the same, but rather keep building on each of those themes while adding more strategic relationships with originators, keep investing on the technology side, and just lift the overall earnings quality, so to speak, that we hand to shareholders.
Right. And incidentally, consider a few of the companies we compete with — or rather, other mortgage REITs that have run into serious trouble, and whose ability to obtain funding, particularly without collateral, has taken a tremendous hit. So our aim is to sustain that business in terms of consistent profits, consistent book value, and covering the dividend, while also remaining, in my view, the top destination for lenders to put their capital within our sector.
To add to what I said about doubling, that really applies to Credit and Longbridge specifically. We've brought Agency down, which has brought leverage down as well. So when I cite that figure, my attention is genuinely on the Credit and Longbridge portfolios.
One final point I'd like to leave everyone with. What we discussed during the earnings call, and what shows up in the earnings presentation, reflects what EFC is doing at this moment, correct? That's really front of mind — how we produced returns during 2025, and the main subject of today's call, Q4 2025. Yet this company has existed for almost 20-odd years now, first as a private entity and afterward as a public one. Across that span, we've delivered returns in numerous areas. I believe that demonstrates how broad the capabilities of Ellington Management Group truly are, correct? So you've watched CRT serve as a driver at various points, along with legacy non-agencies. Our capabilities in CLOs are tremendous, and so are our capabilities when it comes to purchasing distressed commercial loans. You've watched us get involved in mobile home lending, correct? And beyond that, unsecured consumer, auto, aircraft — there are just so many capabilities, skilled PMs, experienced researchers across all these areas, that I completely expect the opportunity set for Ellington Financial to shift as time goes on. It's not — we won't forever be doing what we're doing at present. But here's what I think matters for shareholders to understand: tremendous capabilities exist across nearly every structured product inside Ellington; and when we gather and consider how to structure — how to allocate Ellington Financial's capital, we enjoy the luxury of possessing so many different sorts of arrows in our quiver, correct? For instance, you might spot an opportunity in auto. You might spot an opportunity in unsecured consumer. Lately those have made up small portions of the portfolio. But over time they can become interesting and exciting and attractively priced. So I included that bit about policy risk now, because it's accurate, correct? And we're considering it; we're attempting to position for it; we can forecast what's probable, yet we lack a crystal ball to forecast precisely what will occur. Still, the resources and capabilities Ellington Financial can access through its shared services agreement with Ellington Management Group, I believe, hand us a tremendous opportunity set.
There's one area I'd like to call out, Mark — small balance commercial. We've picked up some excellent assets from banks. Obviously, it's common knowledge that certain parts of the commercial mortgage space have faced significant pressure. I believe our handling of the portfolio has been strong, with very few problems, and that leaves us well positioned. Sellers are bringing us auctions, after all. This market is extremely fragmented. At times it's quite locally focused, so we aren't up against — definitely not up against the large banks when it comes to those bridge loans. Yes, spreads have narrowed across the board, but the spreads on our financing have narrowed by a similar amount. So lately that's been a source of growth for us. My expectation is that it keeps being one. The technicals are — well, they favor buyers and hurt sellers. So I'd say that's certainly a space where stress and opportunity will keep showing up.
Great, that's really helpful. And then, for my second question, if I may — on the topic of book value as of today, I may have missed this earlier, but would you be able to provide an update, either in terms of the actual dollar amount or simply the direction?
Sure. Apologies — please go ahead, JR.
Correct — $13.16 was where we finished the year. January's month-end figures aren't out yet; those should come early next week. We said economic return came in at roughly 2% for January. That would suggest book value rose about 1% after accounting for the dividend.
Correct. At this moment those figures are approximate, but within a handful of days we'll be releasing them once more.
I was just reflecting on what you said in the prepared remarks—you mentioned a broader set of opportunities, partly because the seller network has grown. With your financing capacity getting bigger and more flexible, is it reasonable to look for a wider band on intra-quarter recourse leverage, and for securitization activity to speed up more going forward?
Sorry, can you say that again?
Yeah, so considering that your financing platform's flexibility and reach have grown a lot, would it make sense to anticipate a broader leverage band that shifts from one quarter to the next, plus more rapid growth in securitization transactions?
Yeah. For sure, within a quarter, for instance, if we reported month-end recourse debt to equity, it moves around. What I mean is, 2 deals closed in early February that were still open when January ended. So moving those ahead out of January has brought leverage down, yet they remained on the balance sheet and were completed earlier in February; thus you definitely get noise inside a quarter. Still, I believe thematically, we'll see expansion as long as we're able to complete additional unsecured notes offerings. And we've gotten off to a solid start. What I mean is, we're past 6, 7 weeks of 2026, we're running ahead of 2025's pace, and 2025 was roughly more than 3x quicker than 2024. So that speeding up persists, at least up to now.
So my view is that, you know, we've been moving pretty fast on securitizations, correct? And so, if a situation arises in which we think securitization spreads — suppose they widen out and we're not happy with them — well, then I'd say there's a good chance we'd be holding more loans in warehouse when the quarter closes, along with leverage that's a bit higher, though that would only be for a while.
Understood. Now regarding that fresh RTL securitization you priced — could you elaborate a bit further on how it's structured? In particular, I was curious about the window for the reinvestment period.
Of course. As I noted earlier, this is a revolver, and my understanding is that the reinvestment period spans 2 years. So that's right. And like I said, on a monthly basis, we're able to swap out essentially the loans that get repaid for new ones.
Yeah. That matters since, clearly, those loans tend to last far shorter than 2 years on average, which makes the financing much more efficient.
Okay, operator, I believe that wraps things up. Listen, I'm sorry about how long this took. Thank you for staying on the line with us. We'll be sure the phone bill gets paid on schedule going forward, and we appreciate you bearing with us. And listen, it was a fantastic quarter. We're excited about a fantastic year ahead.
Thanks so much. We appreciate everyone joining us for the Ellington Financial Fourth Quarter 2025 Earnings Conference Call. At this point, you're welcome to hang up, and we hope the rest of your day is wonderful.
Read the full transcript via the API
You're viewing the first half of this call. Get the complete Ellington Financial Inc. transcript - plus 132,000+ transcripts from 7,000+ companies, speaker segments and full-text search - through the EarningsAPI REST API or hosted MCP server.
Get an API key View API docs →For developers and AI pipelines
Programmatic access to Ellington Financial Inc. earnings transcripts and 132,000+ others is available through the
EarningsAPI REST API and the hosted MCP server.
Quarterly plans from $162 - full transcripts, speaker segments, full-text search,
and the /api/v1/transcripts/recent polling endpoint for ETL pipelines.