EFC - NYSE

Good morning, ladies and gentlemen. Thank you for standing by. Welcome to the Ellington Financial First Quarter 2024 Earnings Conference Call. Today's call is being recorded. [Operator Instructions]. It is now my pleasure to turn the call over to Alaael-Deen Shilleh. You may begin.

Thank you. Before we start, I would like to remind everyone that certain statements made during this conference call may constitute forward-looking statements within the meaning of the safe harbor provisions of the Private Securities Litigation Reform Act of 1995. Forward-looking statements are not historical in nature as described under item 1A of our Annual Report on Form 10-K and Part 2 Item 1A of our quarterly report on Form 10-Q. Forward-looking statements are subject to a variety of risks and uncertainties that could cause the company's actual results to differ from its beliefs, expectations, estimates and projections. Consequently, you should not rely on these forward-looking statements as predictions of future events. Statements made during this conference call are made as of the date of this call, and the company undertakes no obligation to update or revise any forward-looking statements, whether as a result of new information, future events or otherwise. I am joined on the call today by Larry Penn, Chief Executive Officer of Ellington Financial; Mark Tecotzky, co-Chief Investment Officer of EFC; and JR Herlihy, Chief Financial Officer of EFC. As described in our earnings press release, our first quarter earnings conference call presentation is available on our website, ellingtonfinancial.com. Management's prepared remarks will track the presentation. Please note that any references to figures in this presentation are qualified in their entirety by the end notes at the back of the presentation. With that, I now turn the call over to Larry.

Thanks, Alaael-Deen, and good morning, everyone. As always, thank you for your time and interest in Ellington Financial. I'll begin on Slide 3 of the presentation. For the first quarter, we reported net income of $0.32 per share and adjusted distributable earnings of $0.28 per share. The credit strategy was the primary contributor to our quarterly results, generating 48% per share of net income, led by steady performance from our non-QM and residential transition loan businesses, together with strong returns from our secondary CLO, CMBS and non-agency RMBS portfolios. Our agency strategy contributed a modest $0.03 per share in a quarter where agency MBS lagged the broader rally in credit, with market consensus shifting to a higher for longer expectation. Finally, Longbridge generated $0.10 per share of GAAP net income. Though I'll note that after taking out the mark-to-market gain on our reverse MSRs and certain interest rate hedges, ADE from Longbridge was still slightly negative for the quarter. That slightly negative ADE from Longbridge again weighed down EFC's overall ADE for the quarter. But on a positive note, so far in the second quarter, origination volumes and submissions at Longbridge are pacing well ahead of first quarter volumes and second quarter projections. And so we expect Longridge to contribute positively to our ADE in the second quarter. Meanwhile, in the first quarter, we made progress deploying the uninvested capital we held at year-end following the closing of the Arlington merger. Our credit portfolio expanded, driven by a larger RTL portfolio, and opportunistic corporate CLO purchases. We also grew our commercial mortgage bridge loan portfolio after 5 consecutive quarters of payoffs exceeding new originations in that portfolio. With borrowers finally more realistic about commercial real estate property valuations, we are seeing strong origination flow from our affiliate Sheridan Capital, which has also sourced a couple of NPLs so far for us in 2024. We also achieved some key portfolio objectives during the first quarter that I'd like to highlight. First, we successfully completed our inaugural securitization of proprietary reverse mortgage loans from Longbridge, which converted repo financing into term non-mark-to-market financing at an attractive cost of funds. We expect that this securitization marks the beginning of an ongoing program for our proprietary reverse business, similar to the program we have established in our non-QM businesses. Second, we continued to cull lower-yielding securities from our portfolio, selling agency and lower-yielding non-agency RMBS and CMBS in order to free up capital for higher-yielding opportunities. Since we generally use more leverage in our MBS portfolios, especially in our agency MBS portfolio, than we do in our loan portfolios, these MBS sales drove down our overall leverage ratios in the first quarter, despite the increased capital deployment overall. So far in the second quarter, we further expanded our RTL and commercial mortgage bridge loan portfolios. We've originated more proprietary reverse mortgage loans on the heels of that March securitization, and we've also begun adding investments in closed-end second lien mortgages and home equity lines of credit, or HELOCs. While we haven't focused on closed and second the HELOCs until recently, we are optimistic about the prospects of deploying significant capital in this sector going forward. After several consecutive years now of home price appreciation, homeowners across the country are sitting on an enormous amount of home equity. But with mortgage rates up sharply, it doesn't make sense for most borrowers to refinance their existing low-rate mortgage. Closed end seconds and HELOCs enable these borrowers to tap into that home equity without disturbing their low rate locked in first mortgage. We are seeing a surge of demand for these products from some of the highest credit quality borrowers with pristine pay histories and attractive yield opportunities. One wildcard here is potential competition from Freddie Mac and Fannie Mae. As always, we'll want to focus on products where we're not competing with those agencies. Also in April, we completed our first non-QM securitization in 14 months, taking advantage of the tightest AAA yield spreads we've seen in 2 years and booking a significant gain as a result. In recent months, we had been choosing to sell many of our non-QM loans rather than securitize them to take advantage of strong whole loan bids in the marketplace. But with AAA spreads finally back to early 2022 levels, we were able to achieve some great economics in the securitization market, and retain some high-yielding residual tranches to boot. With that, I'll turn the call over to JR to discuss the first quarter financial results in more detail. JR?

Thanks, Larry, and good morning, everyone. For the first quarter, we reported GAAP net income of $0.32 per share on a fully mark-to-market basis, and adjusted distributable earnings of $0.28 per share. On Slide 5 you can see the attribution of net income among credit, agency, and Longbridge. The credit strategy generated $0.48 per share of GAAP net income in the quarter, driven by strong net interest income, net gains on our non-agency RMBS and equity stake in LendSure, and net gains on interest rate hedges. LendSure has now posted several consecutive quarters of solid performance, and it made a sizable distribution in the first quarter, with EFC's share of that distribution well exceeding our total cost basis in the investment. A portion of the net income in our credit strategy was offset by net losses on credit hedges, negative operating income on certain commercial non-performing mortgage loans, and REO, and net losses on residential REO liquidations. We also had a net loss on the Great Ajax common shares we had purchased in connection with last year's terminated merger, which was partially offset by a net gain on the fixed payer interest rate swap hedges that we hold against those shares. Meanwhile, the Longbridge segment generated GAAP net income of $0.10 per share for the first quarter, driven by positive results from servicing and net gains on interest rate hedges. In originations, improved gain on sale margins in HECM, driven by tighter HECM yield spreads, were mostly offset by a decline in overall origination volumes. Tighter HECM yield spreads also led to net gains on the HMBS MSR Equivalent, as well as improved execution on tail securitizations, which contributed to the positive results from servicing. Partially offsetting these gains were net losses on proprietary loans. Finally, our agency strategy generated GAAP net income of $0.03 per share for the first quarter. Despite lower interest rate volatility during the quarter, agency RMBS lagged a broader rally in credit as market consensus for the timing of the first Federal Reserve rate cut was pushed back. This drove interest rates higher across the yield curve and pressured agency yield spreads, particularly in February, and particularly for lower coupons where many of our holdings were concentrated. While agency yield spreads did recover meaningfully in March, driven by lower volatility and capital inflows, Overall for the quarter, agency RMBS generated a modestly negative excess return to treasuries. Despite that negative excess return, our agency portfolio was modestly profitable for the quarter as net gains on our interest rate hedges exceeded net losses on our pools and negative net interest income. Our net income for the first quarter also reflects a net loss driven by the increase in interest rates on the fixed receiver interest rate swaps that we used to hedge the fixed payments on both our unsecured long-term debt and our preferred equity, partially offset by a net gain on our senior notes, also driven by the increase in interest rates. As a reminder, we have used interest rate swaps to effectively convert these long-term fixed rate obligations into floating rate obligations. Turning to Slide 6, you can see the breakout of ADE by segment. Longbridge, after excluding the mark-to-market gain on the MSRS and certain interest rate hedge gains, generated ADE of negative $0.01 per share. Apart from the Longbridge segment, ADE from the investment portfolio, net of corporate other expenses, totaled $0.29 per share, which is an increase of $0.03 per share sequentially, but which was still weighed down by non-accrual loans and REO expenses. During the first quarter, delinquencies ticked up modestly in our residential loan portfolio, driven by incrementally higher non-QM delinquencies, in line with broader market trends. Total delinquencies in our commercial mortgage loan portfolio also ticked up quarter-over-quarter, but that's only because we bought a non-performing loan during the first quarter. Also in commercial, we continued to work through the 2 non-performing multifamily bridge loans that we highlighted last quarter. Next, please turn to Slide 7. In the first quarter, our total long credit portfolio increased by 2% to $2.8 billion as of March 31, driven by larger residential transition loan and commercial mortgage bridge loan portfolios, and net purchases of corporate CLOs. A portion of the increase was offset by a smaller non-QM loan portfolio where principal paydowns and loan sales exceeded net originations, and net sales of non-agency RMBS and CMBS. For our RTL commercial mortgage bridge and consumer loan portfolios, we received total principal paydowns of $384 million during the first quarter, which represented 23% of the combined fair value of those portfolios coming into the quarter, as those short duration portfolios continued to return capital steadily. On Slide 8, you can see that our total long agency RMBS portfolio declined by 22% sequentially to $663 million, as we continued to shrink the size of that portfolio and rotate capital into higher yielding opportunities. Slide 9 illustrates that our Longbridge portfolio decreased by 20% sequentially to $441 million, driven primarily by the successful completion of a $208 million proprietary reverse mortgage loan securitization in March. Please note that for this presentation, similar to how we present our consolidated non-QM securitizations, we only include the tranches we retained in the prop securitization, even though the entire securitization is technically consolidated on our balance sheet for GAAP reporting purposes. In the first quarter, Longbridge originated $205 million across HECM and prop, which was a 22% decline from the previous quarter. The share of Longbridge's originations made through its retail channel increased to 25% in the first quarter from 18%, with the share from its wholesale and correspondent channels declining to 75% from 82%. As Larry mentioned, Longridge is on track to increase origination volume in Q2. Please turn next to Slide 10 for a summary of our borrowings. On our recourse borrowings, the total weighted average borrowing rate increased by 9 basis points to 6.87% at March 31. We continued to benefit from positive carry on our interest rate swap hedges, where we overall receive a higher floating rate and pay a lower fixed rate. Asset yields increased for both credit and agency during the quarter, which drove NIM expansion in both strategies. Our recourse debt to equity ratio decreased to 1.8:1 at March 31, down from 2:1 as of year end, driven by a decline in borrowings on our smaller but more highly levered agency RMBS portfolio, and a decrease in our recourse borrowings related to our securitization of proprietary reverse mortgage loans in March. As I mentioned, we do consolidate that prop securitization. So the sold tranches, which represent long term non-recourse financing for us, do stay on our balance sheet. So while our overall debt to equity ratio also decreased over the course of the first quarter from 8.4:1 to 8.3:1, that decrease was smaller than the sequential decrease in our recourse debt to equity ratio. I'll note here, too, that we added 2 new loan financing facilities in April. At March 31, our combined cash and unencumbered assets totaled approximately $732 million, up from $645 million at year end. Our book value for common share was $13.69 at quarter end, down from $13.83 at December 31. Our total economic return was positive 2.1% for the first quarter. Now over to Mark.

Thanks, JR. The market backdrop in Q1 was a good one for many levered structured spread products. Rate volatility declined during the quarter as uncertainty about the Fed shifted from wondering about how high they would hike to speculating about when they will cut. That's a much more conducive environment for spread assets. Implied volatility came down and yield spreads across corporates and most structured products came in. EFC did what a REIT is supposed to do in that environment, we captured spread income, we converted some repo financing to tighter spread non-mark-to-market financing through securitization, we selectively sold assets that we believed could reach their full potential, and we sourced some high-yielding assets for our portfolio. For the quarter, we had broad-based contributions across our strategies, including both commercial and residential, both loans and securities. We came into the year with ample cash to deploy from the Arlington merger that closed in mid-December, and we put some of that to work growing our credit portfolio. One of the highlights was the growth of our commercial mortgage bridge portfolio, where both new originations and non-performing loan acquisitions have picked up, even as we've substantially tightened our underwriting criteria. In recent quarters, we mentioned an approaching opportunity for EFC related to commercial real estate dislocation and regional bank de-risking. And that opportunity is here now. Including a purchase in April, this year we bought 2 non-performing loans at substantial discounts to par, which we believe are well secured by the underlying real estate, one of them coming from a regional bank. Our partnership with Sheridan Capital and the deep experience of our in-house team leaves us well positioned to capitalize on these types of situations. Another big highlight for the quarter was our reverse mortgage prop securitization. The underlying loans for the securitization were not FHA-insured HECM reverse mortgage loans, but rather private-labeled loans. So this securitization is the non-agency version for reverses. Proprietary reverse mortgages is a sector with a lot of potential for growth and with a few competitors. The securitization allowed us to both recycle our capital and create some very high-yielding long-duration retained tranches. I'm hoping that this and future prop securitizations will lead to improved gain on sale margins and additional prop origination volume, and thus increase profits for Longbridge. In addition, early in the second quarter we priced our first non-QM securitization in over a year. In 2023, we had slowed down securitizations in the face of relatively wide spreads on the debt tranches we could issue. These spreads have tightened significantly in 2024, however, making securitizations attractive again. Our securitization in April allowed EFC to leverage these tighter spreads and create high-yielding retained tranches for our portfolio. Turning back to Slide 7, let's look at how the portfolio evolved during the first quarter. You can see here that we had modest portfolio growth. We grew commercial bridge and also RTL, but working in the other direction, we took advantage of tighter spreads to harvest some gains in our non-agency RMBS portfolio. That portfolio has had tremendous returns over the past 12 months. Portfolio growth has continued following quarter-end. We scaled up further in commercial bridge and RTL, and we've also started buying some Closed End Seconds and HELOCs. With so many homeowners sitting on both substantial equity in their homes and first mortgages with very low rates, we see equity extraction as a potentially large, high credit quality, high yielding opportunity. Both Closed End Seconds and HELOC can provide us with very high note rates, modest purchase premiums, substantial levered NIMs, and potential securitization outlets to manufacture more high-yielding retained investments. On Slide 8, you can see we continued to shrink our agency portfolio as we are seeing what we believe are more interesting opportunities in some other sectors, and we'll continue to be opportunistic about sales in our agency MBS. Looking ahead, the current market backdrop should continue to generate further investment opportunities for high returns and high ADE. Market expectations for interest rate cuts should both dampen rate volatility and continue to draw more capital into the market, which should support spread products. In addition, the Fed has recently announced a slowdown in their portfolio runoff, which should support bank balance sheets; and by expansion in the assets they buy. We see several new exciting opportunities, and EFC has been pouncing. First, the pace of sales of deeply discounted CMBS and commercial NPLs is increasing. We are well positioned to capitalize on those opportunities, as last year we deliberately allowed our commercial mortgage portfolio to run off organically, and as only a few of our bridge loans are currently in workout. Securitization spreads have tightened significantly from last year, which is tailwind for our non-QM and prop-reverse businesses. We'll be exploring securitization financing in our RTL business as the first rated RTL securitizations recently got done, significantly improving financing economics. Finally, we are ramping up both in Closed End Seconds and HELOCs. Now, back to Larry.

Thanks, Mark. As you can tell, we have a wide variety of high-yielding loan strategies where we're currently putting significant capital to work. For the second quarter so far, we've added meaningfully to our credit and Longbridge portfolios in April, and we've continued to make room for more loan flow by further trimming our agency specified pool portfolio, and by securitizing a significant portion of our non-QM loan portfolio. From a leverage perspective, the net effect of our recent purchases and originations has been outweighing the impact of our agency sales and non-QM securitization. So our recourse debt to equity ratio is on the rise again. Moving forward, I expect leverage to continue to ratchet up. But as was the case in the first quarter, it won't always be a linear progression given the push and pull from different elements of our portfolio management strategy. As always, our aim is to balance the dual objectives of growing earnings in the near term with preserving dry powder to capitalize on opportunistic situations such as what we are currently seeing play out in distressed commercial real estate debt. To that end, even after April's activity, we continued to have ample cash and borrowing capacity with lots of unencumbered assets, plus other lightly leveraged assets such as our forward MSRs. Besides portfolio expansion, we expect ADE growth to come from 2 other principal areas. First, continued progress on the handful of non-performing commercial loans and REOs in our portfolio. So far this year, this progress has been slow but steady. And second, Longbridge returning to profitability in originations. As I mentioned earlier, origination volumes and submissions so far in the second quarter are well ahead of projections, despite higher interest rates, which partially reflects additional sourcing channels that Longbridge has successfully established. As a result, we're expecting Longbridge to contribute positively to ADE in the second quarter. Before wrapping up, I'd like to comment on how EFC is positioned for interest rate changes from here. Over EFC's almost 17 year history, we have always endeavored through active hedging to protect EFC's book value per share against movements in interest rates, rather than speculating on the direction of interest rates. This philosophy was again important in the first quarter, where rising longer-term interest rates drove profits on our hedges, which offset losses on some of our longer-duration investments. Moving forward, if we are indeed in a higher for longer interest rate environment, as it seems, I believe that Ellington Financial is well positioned with our hedging expertise, short duration, high-yielding loan portfolios and mortgage servicing rights portfolios. With that, we'll now open the call up to questions. Operator, please go ahead.

Thank you. The call is now open for your questions. [Operator Instructions] Our first question comes from Bose George with KBW.

Why don't we come back to Bose for the second question.

We'll go next to Trevor Cranston with Citizens JMP.

On the opportunity you guys talked about on Closed End Seconds and HELOCs, can you maybe add a little bit of color in terms of are those things you're able to source from the existing proprietary originator relationships you have, or are those coming from different sources? And maybe if you could also talk about sort of the longer-term financing strategy for those products.

Sure. Trevor. It's Mark. Yes, those have generally been sourced from different entities. The real pool of borrowers where it makes sense for them to take the Closed End Seconds are generally people that have agency first lien mortgages that they took out, say, during '20 or '21. They have note rates around 3%. They've had a lot of HPA. So it's primarily working with originators that have an agency presence. In terms of financing, I think there's 2 options. One is repo financing. So in this way it's not really different than non-QM or RTL just repo financing. And then there's also securitization outlets. So that is an area where we have so far, it hasn't been big yet, but you're able to get pretty high note rates, not real high premiums, and get what looks on the surface like very high credit quality underlying loans.

Got it. Okay. That's helpful. And then across all the different investment opportunities, it sounds like there's a lot of focus on the more proprietary things you guys have been doing. I was curious if you could maybe talk a little bit more about the CLO side, which is relatively small right now, but seems like there are pretty good returns available there. So just kind of how you see the opportunity in deploying capital into that versus some of the other loans you guys have been doing?

It's opportunistic there, secondary CLOs. This is, first of all, it's similar to what we're doing over at Ellington Credit as part of that transformation, right, focusing on secondary CLOs. And so Ellington Financial's have been getting allocations in that strategy as well, alongside Ellington Credit. So it's just opportunistic, I think these are obviously more liquid than, say, residential transition loans or things like that, but we do see good opportunities there. So it's still a very small part of our portfolio. And while the growth was certainly significant enough to mention on this call, I wouldn't necessarily think of that as something sort of core holding for a long period of time that would represent a very significant section of portfolio. But we are opportunistic and we did like the risk-reward there. So Ellington Financial has been participating to some extent there as well.

We'll take our next question from Frank Labetti with KBW.

This is Frankie filling in for Bose. Just wanted to touch on your comfort level on the dividend, when do you see normalized earnings being roughly in line with the dividend? And how comfortable are you at the current level?

Yes, we're comfortable. Yes, I think we set it at a level where we hope to keep it there stably for some time, just like we have with prior dividend resets. So, I would remind everyone that it's actually not that different from the dividend, I think $0.14 that we had a few years back. But yes, we're quite comfortable that this is where the dividend is going to be for a while.

And just a follow-up, when do you see the company getting to normalize earnings? Yes, that's a little harder to project, because as we've mentioned on the call, we've got some volatility and drag. So first of all, Longbridge, we said, hasn't contributed to ADE sort of commensurate with the capital in that segment. We think that that's going to turn around pretty soon. We mentioned the second quarter is looking like it's going to contribute positively, so that trend is in the right direction. And then a big thing as well is that we've got some NPLs and a couple NPL/REO that are also not giving us a yield, right. It's really a yield that contributes yield-bearing assets that contribute to our ADE in the investment part of our portfolio. So, while we're working those out, that can be a drag to our ADE. Of course, when we ultimately resolve those assets, we're certainly hopeful that we're going to catch up a lot from an earnings perspective. But that doesn't always even flow into ADE. Sometimes it does, sometimes it doesn't, depending upon the nature of the resolution. That's just how the sort of ADE accounting works. So I would say that I can't give you a precise date, but I certainly hope that our ADE will sort of approach continue to get closer and closer to our dividend. But in the meantime, we also are seeing strong gains on sale and other things that don't necessarily translate into ADE per se, but translate into GAAP earnings. So I don't want to sort of ever give the impression that we're focusing more on ADE than GAAP earnings. And ultimately, our goal is to earn the dividend GAAP wise, and thereby kind of preserve book value and maintain book value while paying out that dividend.

We'll take our next question from Jason Weaver with JonesTrading.

I wonder if you could give some context behind the sort of reluctance of regional banks to be involved in RTL in commercial transitional space amid what we're hearing for the last few quarters from more competition from non-banks, and how that's affecting available yields?

Yes, this is Mark. It doesn't surprise me that regional banks aren't involved in this sector. This is a sector where it is a heavy underwrite. You need to have a team of people that are knowledgeable about construction cost, construction timelines. It's not necessarily the kind of expertise you have in a lot of these regional banks. And the other thing is, it's a lot of work for the dollars you put out as opposed to, like, lending to a commercial developer, putting a multifamily. Here, we're putting out money $250,000, $300,000 at a time. And there's not only the initial underwrite, but there's also draws you have to supervise. So, we like the fact that this sector doesn't attract regional banks and it doesn't really attract sort of generic, middle of the fairway mortgage originators. Yes, the space is competitive, but it always seems like every space we're in is competitive. I don't view this space as more competitive now than what it was a few years ago. You've had a couple of new entrants, but you had some other people that used to be involved are no longer involved. So we have not had an ability of sourcing loans, and we do not feel like guidelines among the competitors have loosened and has pressured us to relax guidelines at all. If anything, we got actually a fair bit tighter in 2022 in terms of our guidelines and how we underwrite.

And then apologies if I missed it during the prepared remarks, but can you give an updated book value and leverage where you're seeing that currently?

So we haven't given on April yet. Ordinary course, we put out a book value per share estimate for each month end, that will happen later in May. And then in terms of leverage, we haven't given the precise current leverage amount. Larry did mention in his prepared remarks that we've been active deploying in April. And that we also securitize the non-QM loans which come off balance sheet. But net-net leverage has ticked up quarter-to-date. So far we haven't quantified that, but that's a color we did provide.

Our next question will come from Eric Hagen with BTIG.

How are you doing, guys? You talked about being well-hedged for a higher for longer environment. Are those mostly interest rate hedges that you're talking about? And do you think about adding any credit hedges to help mitigate some of the risk in areas like the [ bridge ] portfolio and the CRE portfolio? Or is just the underwriting really kind of the hedge, if you will?

Eric, it's Mark. So that being well hedged for higher for longer, I mean, I think I would just say, I'd take a step back in that, we're always well hedged. We've always tried to mitigate interest rate risk at the company level by a series of rate hedges. So, I just think Larry's comments in the prepared remarks is that, if you are in this higher for longer environment, you think our portfolio construction and our sourcing channels can do really well there, commercial bridge, RTL, short spread duration, the commercial bridge, all indexes, SOFR. So that's the kind of stuff where you get a big yield, a big coupon if the Fed keeps short rates higher. In terms of credit hedges, we've had those. Like, we've had credit hedges against non-QM for a while. And they function 2 ways. One is just really sort of like overall macroeconomic credit hedge, but the other way they function is hedging spread risk for securitization. In particular like, some of the IG indices now have done really well on the year. Like the current IG index is right around 50. So that to us seems like a hedge where owning protection on that, we have kind of limited downside and it can certainly mitigate NAV volatility of spreads wide. And it was kind of interesting, we priced our non-QM deal right when you had that sort of little like credit cheapening in April, and that IG index widened out 5 or 6 basis points in time of our pricing. And when we priced, we were able to get good execution. But now that we had sold some of that risk into the market, we bought back some of that hedge. So, I think, the credit hedges, for a long time we had sort of CMBX-specific credit hedges on CMBS B pieces. That's less of what we're doing now. CMBX has gotten a little bit less liquid, but the credit hedges you see from us now are more referencing corporate indices. And they're in part for sort of overall macroeconomic risk, but they're also to just sort of manage on spread volatility.

That's a great detail. I appreciate that you mentioned the non-QM execution, let's talk about that. When you do a securitization at these rates and spreads, can you share kind of what you think is the all-in return, including the ROE on the retained piece of the credit from the transactions, and including any loss estimates you sort of expect that's embedded in that return profile?

Yes. JR, do you want to take that? Wasn't sure how much disclosure we normally give.

Let me think about that. We don't give precise ROEs. I think we can speak broadly on, we can start with where we price the AAAs, and we can compare that and the total capital stack. I think mid-teens is where we pencil is kind of the punchline. As we build up to the components of that, let me think about that, Eric, and how we can kind of provide some more color on how we get from A to B. But big picture, mid-teens with, I think, conservative underwriting assumptions going forward is where we're penciling always on those.

Yes. Let's see what we can provide here. So, first of all, when we do a securitization, right, like sometimes those are non-QM securitizations, sometimes they're consolidated and sometimes they're not, right. And that's an accounting distinction, obviously. But we really don't think of them differently, right. We really are thinking of them always as, okay, we did a securitization and now we've transformed our loans into the retained tranches. And we put a value on those retained tranches, right? So if you sort of think of this as a sale of the loans and a purchase of the retained tranches, which it's not always accounted for that way, but let's just say we kind of try to think of it that way, then you're going to recognize a gain or a loss kind of on that securitization, right. It's the difference between what the loans were worth that you put into the securitization, what your securitization expenses were, right, because there are expenses in doing securitization, legal, underwriting, et cetera. And then what the retained trashes are worth that you are taking back. And so that sort of gives you, let's say, a gain on sale, even if it's not necessarily, accounting-wise, a gain on sale. And for that, that's going to be very variable. So, to make a long story short, that sort of securitization gain is going to be very variable depending upon everything from how well we hedge the loans going into the securitization, as Mark mentioned sometimes we use IG hedges now, in addition to interest rate hedges, we use TBAs sometimes, or interest rate swaps depending upon what we think of the basis. But whatever, we'll have a gain on sale or a loss on sale, hopefully a gain on sale when we do that. So that's more of a one-time event, right? And then we're going to be retaining the tranches. I don't think it'd be giving away too much to say that if you look at the tranches that we retain, what are they? They're subordinated tranches that have principal, maybe the single B rated or non-rated portion of the deal. And then there's an IO, right. And so we hold those knowing that in some cases we're subject to risk retention and are not able to sell those, and we're going to own those basically forever if and until we call the deal. So we're going to value those where we think the market values them, and we try to be conservative on that. But if you look at the spreads and where those deals price, you can sort of extrapolate. These are quite wide and we can get financing on them as well. So as JR mentioned, we're well into the mid-teens. In terms of the IOs, do we want to sort of, and call rights are another thing that we retain as well, and we value those as well, because those often have value. Now, our call rights on the super old deals that we did with really low coupons, those don't have much value now. So as we revalue those every quarter, those have gone down in value from where they were a few years ago. But when we priced one of these deals, we're using what kind of an OAS on the IOs, generally speaking, Mark, do you? If you don't know off the top of your head, we don't need to, I think, answer that. But it's extremely wide, right? I mean, this is something where even before leverage we're in the teens, I believe.

Yes, you're low teens unlevered. And the other thing I would add, Eric, is, if you look back, we have called a lot of old securitizations. Now, those calls were executed before the big sell-off in 2022. But those call options can prove very valuable because as spreads tighten as the loan season, the rating agencies look very favorably on loss expectations of seasoned loans. So we've called our 2017, '18, most of our 2019 deals, and we, generally speaking, recycled the loans, put them into new securitizations, and it was very helpful economics for us. So I think there's ancillary benefit to doing these securitizations that you don't realize at the time when you do it, but it can pay dividends in the future.