Earnings Call
Ellington Financial Inc. (EFC)
Earnings Call Transcript - EFC Q3 2022
Operator, Operator
Please standby. Your program is about to begin. Good morning, ladies and gentlemen. Thank you for waiting. Welcome to the Ellington Financial Third Quarter 2022 Earnings Conference Call. Today’s call is being recorded. It is now my pleasure to turn the call over to Jason Frank, Deputy General Counsel and Secretary. Sir, you may begin.
Jason Frank, Deputy General Counsel and Secretary
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, 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 J.R. Herlihy, Chief Financial Officer of EFC. As described in our earnings press release, our third 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 will now turn the call over to Larry.
Larry Penn, CEO
Thanks, Jay, and good morning, everyone. As always, thank you for your time and interest in Ellington Financial. After a challenging first half of the year, July started the third quarter on a constructive note, with volatility, interest rates, and most yield spreads reversing much of their second quarter increases. The rally proved short-lived, however. Over the course of August and September, continued elevated levels of inflation and hawkish messaging from the Fed drove interest rates sharply higher. Volatility shot up to record levels, fears of a recession intensified, and the yield curve inverted, stressing equity and fixed income markets alike. Market sentiment steadily weakened, and we saw widespread selling across sectors, particularly towards the end of the quarter. In some cases, this included forced selling by asset managers to meet margin calls or redemptions. Liquidity declined and yield spreads widened in virtually every fixed income sector, with many sectors reaching their widest levels of the year. Against this difficult backdrop, Ellington Financial generated a net economic loss for the quarter of 3.4%, driven by losses on non-QM, Agency RMBS, and originator stakes. Nevertheless, our diversified portfolio, stable sources of financing, and dynamic hedging significantly limited the magnitude of that loss. During the quarter, we had strong performance in our residential transition loan, small balance commercial mortgage loan, CLO, and CMBS portfolios. And we benefited from significant net gains on our interest rate hedges and non-QM interest-only securities. We also completed our third non-QM securitization of the year during the quarter. Turning now to the investor presentation. On slide three, you will see that we are reporting adjustable distributable earnings of $0.44 per share, which is a $0.03 sequential increase and which nearly covered our dividend. The increase resulted from our continued rotation of capital into higher reinvestment yields and from a larger credit portfolio, as we got more invested. With interest rates rising so dramatically this year, especially in the short end of the yield curve, we have been playing a bit of catch-up with our adjustable distributable earnings. The purchase yields on many of our assets, especially our agency pools, still reflect the much lower interest rate environment that we had earlier this year. So as we continue to rotate out of that lower yielding portfolio, we should get a boost to our ADE. In addition, keep in mind that a lot of our credit portfolios are quite short in duration. So those will rotate more or less by themselves, naturally. The bottom line is that we are still constructive on where our ADE is heading. Meanwhile, as you can see from our cash and unencumbered asset figures, we have continued to maintain a strong liquidity position, even as we have grown the credit portfolio. Finally, I’d like to move to the Longbridge transaction. With all required regulatory approvals finally obtained, we closed on the acquisition of the other half of our affiliate reverse mortgage originator, Longbridge Financial, shortly after the quarter end. The final purchase price of $38.9 million was substantially lower than the initial estimated price of $75 million that we announced in February and reflected a discount to Longbridge’s book value rather than a premium as originally estimated, along with a lower book value. With the closing of this transaction, Ellington Financial now holds a controlling stake in Longbridge, and so we will fully consolidate Longbridge onto our financials beginning with our fourth quarter financials. J.R. will elaborate on that later. From a business perspective, we believe that Longbridge’s future earnings prospects are strong, even with the challenging market conditions we have seen so far this year. Given the substantially lower final purchase price that we paid, we believe that the stage is set for an excellent return on equity on our investment going forward. Well, it’s been a really tough market for all mortgage originators, Longbridge actually managed to turn a profit in the third quarter, and with much of its competition hobbled, Longbridge also became the second largest issuer of new issue HECM, HMBS, with a 20% market share. That said, the surge in interest rates has driven HECM volumes lower so far in Q4, and the Ginnie Mae HMBS outlet for Longbridge’s HECM production is still trading at wide levels. Those are going to be headwinds for the business in the near term. But make no mistake, the situation in the reverse mortgage market is quite different from that of the forward mortgage market. The reverse mortgage market is still largely untapped. Longbridge’s market share has been rising and the demographic trends are extremely favorable. So we are definitely constructive on Longbridge’s long-term prospects. Furthermore, the income stream that we expect to see from the Longbridge acquisition should also enhance the diversification and quality of EFC’s earnings stream. The reverse mortgage origination business can flourish in an economic downturn, for example, because reverse mortgages provide liquidity to borrowers without the requirement to make monthly principal and interest payments. In fact, the last peak in HECM originations was in the wake of the global financial crisis in 2009, when home prices were falling rapidly. And Longbridge’s origination profits soared in the second quarter of 2020 during the depths of COVID, when other fixed income businesses were teetering. There is definitely a counter-cyclical component to the reverse mortgage business. With that, I will pass it over to J.R. to discuss our third quarter financial results in more detail.
J.R. Herlihy, CFO
Thanks, Larry, and good morning, everyone. I will continue on slide three of the presentation. For the quarter ended September 30th, Ellington Financial reported a net loss of $0.55 per share on a fully mark-to-market basis and adjusted distributable earnings of $0.44 per share. These results compared to a net loss of $1.8 per share and ADE of $0.41 per share for the prior quarter. On slide four, you can see that we further increased our capital allocation to credit investments during the quarter to 88%, up 1 percentage point from June 30th. Based on the opportunities that we are currently seeing in credit, including the recently closed Longbridge transaction, I expect this allocation to continue growing slowly relative to agency. You can also see on this slide that average market yields are up on both our credit and agency portfolios considerably as compared to last quarter. As Larry mentioned, the purchase yields on many of our assets still reflect a much lower interest rate environment that we had earlier this year. As we continue to turnover our assets, we expect that the gap between our purchase yields and market yields will narrow and that should be supportive of our net interest margin. On slide five, you can see the attribution of earnings between our credit and agency strategies. During the third quarter, the credit strategy generated a gross loss of $0.19 per share, while the agency strategy generated a gross loss of $0.17 per share. These results compare to gross loss of $0.80 per share in the credit strategy and a gross loss of $0.20 per share in the agency strategy in the prior quarter. Net interest income on our credit portfolio increased significantly quarter-over-quarter, driven by the larger portfolio, while we also had strong performance from our CLO and CMBS strategies, and significant net gains on interest rate hedges and retained non-QM tranches, driven by the appreciation of our non-QM interest-only securities. On the other hand, rapidly rising interest rates, widening yield spreads, and weak securitization economics generated losses on our unsecuritized non-QM loan portfolio and continued to pressure gain on sale margins and origination volumes for our loan originator affiliates. Both LendSure and Longbridge were profitable in the third quarter, but the valuation for each declined significantly. For Longbridge, that was due to lower earnings compared to prior periods. And for LendSure, the valuation decline reflected the reduction in our final purchase price for the other half of the company. As a result of these valuation declines, EFC booked a significant mark-to-market loss on its investments in these loan originators for the quarter. Agency RMBS continue to face fierce headwinds in the third quarter as durations extended in response to higher interest rates and as elevated volatility contributed to yield spread widening. In our portfolio, net losses on Agency RMBS exceeded net interest income and net gains on interest rate hedges, while we also incurred double the hedging costs stemming from the volatility. As a result, we had a significant net loss for the quarter in our agency strategy. Turning now to slide six. During the third quarter, our total loan credit portfolio grew by 3% sequentially to $2.74 billion at September 30th. The increase was driven primarily by a larger RTL portfolio, partially offset by opportunistic sales, paydowns, and mark-to-market losses elsewhere in the credit portfolio. Larry alluded to the short duration of many of our loan portfolios, and these portfolios continued to return significant capital during the quarter. We received principal paydowns of $205 million on our RTL, SBC, and consumer loan portfolios, which represented 14% of the combined fair value of those portfolios coming into the quarter. On slide seven, you can see that our total long Agency RMBS portfolio decreased by 15% to $1.14 billion, resulting from net sales, paydowns, and price declines. Please turn next to slide eight for a summary of our borrowings. Our weighted average borrowing rate increased by 115 basis points sequentially to 3.76% at quarter end, driven by higher short-term rates and a greater proportion of our borrowings secured by our loan portfolios, which carry higher borrowing rates than agency assets. For both our credit and agency strategies, our cost of funds increased sharply during the quarter, driven by higher short-term interest rates. Book asset yields for both strategies also increased over the same period, thanks to portfolio turnover to buy a lesser amount. As a result, the NIM on our credit and agency portfolios declined quarter-over-quarter to 2.34% and 1% from 2.75% and 1.76% in the prior quarter, respectively. Despite this NIM contraction, we were able to increase our ADE by $0.03 per share quarter-over-quarter. Our recourse debt to equity ratio adjusted for unsettled purchases and sales was unchanged at 2.6 to 1, as lower recourse borrowings on our smaller Agency RMBS portfolio and the maturity of our $86 million of old senior notes was roughly offset by higher recourse borrowings on our credit portfolio and a decrease in total equity. I will note that the availability of secured financing has continued to hold up well amid the market volatility, but we have seen haircuts increase and our financing spreads widen on several of our credit loan facilities. In fact, we were able to further expand our loan facilities during the quarter; we added a new facility for residential loans and we extended the term of one loan facility by 24 months. On a technical point concerning our liabilities, you will see on our balance sheet that we marked down our 5.78% senior note liability by a fair amount, but since we are hedging that liability with SOFR swaps, which were also marked down, those two markdowns largely offset each other this quarter. Total G&A expenses for the third quarter were $0.14 per share, unchanged from the prior quarter, while other investment-related expenses increased by $0.02 per share to $0.10 per share. During the quarter, we were opportunistic with our capital management strategy. In August, we issued approximately 517,000 common shares under our ATM program at an average price of $15.55 per share, and later in the quarter, we repurchased 40,000 shares at an average price of $12.38 per share. As of September 30th, our book value per common share was $15.22, down 6.2% from $16.22 per share at June 30th. Including the $0.45 per share of common dividends that we declared during the quarter, our total economic return for the third quarter was negative 3.4%. Finally, as Larry mentioned, we will consolidate Longbridge beginning in Q4 and we are planning to report Longbridge as a separate operating segment in our GAAP and non-GAAP financial reporting. Because Longbridge does not achieve true sale treatment on its HMBS securitizations, it consolidates all of those non-recourse securitizations for GAAP reporting purposes. As a result, the gross size of EFC’s balance sheet will more than double next quarter, even though Longbridge’s equity is quite small relative to EFC’s. The $38.9 million purchase price for the other half of Longbridge was less than 3.5% of EFC’s total equity at September 30th. I will also point out here that Longbridge’s recourse debt to equity ratio was 2.3 to 1 at quarter end, which was marginally lower than EFC’s. We are excited to assimilate the reverse mortgage business into Ellington Financial, and we believe that the investment will be accretive to both our GAAP earnings and ADE over time. Here, I’d like to make an important note about how consolidation will impact EFC’s earnings going forward. Until now, Ellington Financial only owns a minority stake in Longbridge. Like virtually all of our other investments, EFC’s fair valued that minority equity stake in Longbridge each quarter and the change in that fair value flowed directly through EFC’s net income through our income statement in earnings from investments and unconsolidated entities. Of course, Longbridge’s own earnings were a primary driver of that fair value change. So Longbridge’s own earnings did impact EFC’s earnings, but other factors came into play on the fair value determination, including earnings multiples and book value multiples prevailing in the M&A and stock market. Starting next quarter, fair value determinations of our investment in Longbridge will no longer be factored into our financial results, and instead, Longbridge’s GAAP earnings will simply flow directly into EFC’s earnings. I will also note that similar to EFC, historically, Longbridge has also fair valued virtually all of its assets. So the book value that we acquired was effectively based on fair value, not historical costs. In addition, the calculation of our adjusted distributable earnings will change upon consolidation. In the past, when we calculated our own ADE, we did not incorporate any adjusted distributable earnings generated locally at Longbridge, but upon consolidation in Q4, EFC’s ADE will include the ADE generated by Longbridge going forward.
Mark Tecotzky, Co-Chief Investment Officer
Thanks, J.R. This quarter has been remarkable, considering the significant changes in interest rates, the adjustments in the yield curve, and the widening spreads. What makes this period unique is the duration of the volatility. We've seen intense market movements before, such as in March 2020 and during parts of 2008, but during those times, the volatility did not persist day after day for months as it has this quarter. An economic return of down 3.4% for Ellington Financial is quite impressive given the circumstances we navigated. We encountered instances of forced selling in the third quarter, particularly in September due to turmoil in the U.K., where some managers had to sell against their will. Although I don’t believe all market volatility is behind us, there are signs that some sectors are beginning to stabilize. For instance, the Agency MBS market is now functioning more effectively, not just absorbing forced liquidations, and unlike some days in the previous quarter, there is better two-way flow today. Spreads are moving consistently, with liquid credit indices, and balance sheet issues no longer appear to be the primary worry. With spreads very wide, the opportunities look appealing. However, other structured product markets, such as Non-QM, have not yet found that balance. Non-QM was particularly impacted during the third quarter; rising mortgage rates have drastically reduced prepayments in this sector, and AAA bonds issued before Q3 are facing significant extension risk. Changes in call assumptions have led to notable price declines in non-QM AAAs. However, we do not hold these in EFC; instead, we maintain loans and retained tranches. Despite some credit concerns, our retained tranches have seen value appreciation this year due to their large IO component and reduced CPRs. We also have hedges in our unsecuritized loans, including being short MBS in the TBA market, along with credit hedges against those positions, which have helped offset losses in our non-QM strategy during the third quarter. Non-QM has experienced significant widening as the market grapples with repo spreads and AAA spreads in its quest for stability and sustained investor demand for bonds. We didn’t expect the non-QM securitization market to remain dysfunctional for this long, but we anticipated potential challenges, which is why we began extending our repo terms and adding counterparties back in Q1. We've often discussed the advantages of non-mark-to-market term securitization financing over repo, but at times, the pricing dynamics are so extreme that using additional repo makes sense for a quarter or two, allowing us flexibility to delay a securitization if needed. The non-QM market is now showing early signs of recovery, as supply decreases with lower repo activity and many originators pulling back or exiting the market, while new capital, mainly from insurance companies, is entering with added demand. Additionally, yields on new originations are quite attractive. There are other factors we are monitoring. For instance, regarding commercial mortgage bridge loans, as these loans reach maturity, property owners are compelled to refinance at much higher rates, making DSCR the main limiting factor on new loan sizes, shifting from LTV. This scenario can be so critical that for some properties meeting their business plans, the new loans available might not be sufficient to pay off existing loans. This creates an upside-down capital structure, even without property-level performance issues, and can only be addressed with new equity or mezzanine capital. However, it also presents an opportunity for us to provide that necessary capital at attractive yields. The downward pressure on LTVs could also lead to considerable NPL volume, a type of product we've not seen in significant amounts for years. So far, our commercial mortgage bridge loan portfolio has performed well, and valuations of underlying properties remain solid, but we are preparing to navigate potential mortgage delinquencies. Due to the opportunities in other sectors, our lending volume for the quarter was lower. EFC participated in five new originations and resolved four loans, keeping our portfolio size about the same from the previous quarter. Next, regarding fix-and-flip, if mortgage rates remain stable, we anticipate a continued dip in home prices to improve affordability. Many markets are witnessing all-time low housing affordability levels. Additionally, with many borrowers locked into low-cost mortgages below 3.5%, we believe existing home sales will continue to decline, posing a challenge for fix-and-flip operators who need homebuyers to pay off their loans. We are observing clear signs of this trend, as the market time for new home listings is increasing nationwide. In our portfolio, we continue to see a healthy volume of paydowns and strong performance, as our operators generally sell homes above our underwritten values. For new loans, we are proactively lowering LTVs, narrowing the scope of work we lend against, capping the value of the homes we finance, and becoming more selective about lending regions. Unlike last year, it’s a buyers' market now, allowing us to enforce stricter terms, increase lending rates, and still achieve our desired volume, given the limited alternatives for these operators. Overall, we incrementally grew our credit portfolio during the quarter, primarily in RTL, while decreasing our agency portfolio, which now makes up only 12% of our capital, as indicated on slide four. Both strategies present great return potential, but the agency strategy relies significantly on leverage, prompting a continued shift of funds from agency to credit. We mentioned in the press release that we marked down our originator stakes this quarter. For LendSure, this stemmed from lower origination volumes and reduced gain on sale margins, although they remained profitable, reflecting well on their management's quality and discipline. The upcoming quarters will be challenging for loan originators, but the competition has decreased significantly, creating an exciting opportunity. Besides managing EFC’s portfolio of loans and retained tranches, we maintain ongoing communication with our loan origination partners to keep them updated on secondary market conditions, while they inform us about changes in the origination landscape; this dialogue is invaluable. On page seven, you will see that our agency portfolio decreased by $200 million due to a mix of net sales, paydowns, and price declines. We drew from our COVID playbook, reducing the agency portfolio to enhance liquidity during credit market strains. However, Q3 did not resemble the liquidity crisis of COVID. In 2020, the Agency MBS market recovered before credit due to the Fed’s purchasing actions, whereas in Q3, both agency and credit markets were under review, with agency markets remaining liquid throughout. Looking ahead, we need to consider the implications of recent market fluctuations. What opportunities have emerged, and what risks do sharply rising interest rates and a heightened risk of recession present? We recognize the potential impacts on rates and spreads when the Fed increases rates at its fastest pace in 40 years while simultaneously shrinking its balance sheet, along with the potentially severe and prolonged consequences. Our seasoned senior management, experienced PMs, and strong risk management team have navigated numerous shocks and bear markets before. Once the turbulence settles, I am confident we will find ourselves in a favorable environment where yields on assets are significantly high relative to hedging financing costs and bearish pricing assumptions will lead to substantial returns. The future outlook appears as promising as it has in years. Although spread widening has posed challenges for book value recently, we believe the potential for future earnings has improved significantly. With wider spreads and increased yields, even if yields stabilize, many segments in RMBS and CMBS could see multiple points of price appreciation simply from tightening spreads. The agency market is also experiencing considerable widening, yet is demonstrating better stability and balance since the quarter ended. Additionally, competition in origination markets has decreased, allowing for greater pricing power to tighten investment guidelines and implement higher rates. There is still much to concern ourselves with, and we will maintain our discipline; however, new capital is increasingly entering the market to capitalize on yield opportunities. Now, I'll turn it back to Larry.
Larry Penn, CEO
Thanks, Mark. As we move into the final weeks of 2022, I think you can tell that we are excited about the ample investment opportunities in both securities and loans. But we continue to weigh the deployment of additional capital against maintaining an adequate liquidity buffer to guard against a deeper market downturn. We also sense an opportunity to help our loan originator affiliates to continue adding market share just as many of their competitors withdraw from the market. Finally, with recession fears looming, I will note that the credit performance of our loan portfolios, as measured by delinquencies, defaults, and credit losses, continues to be strong. But with the increased risk of an economic slowdown, we are focused on tightening our underwriting guidelines, with a particular emphasis on keeping LTVs low and being even more selective about geography and property type. For our residential transition loan and commercial mortgage bridge loan borrowers, sharply higher interest rates are stressing refinancings and takeouts, so a lower starting LTV point helps insulate against property value declines. We are now seeing clear evidence of home price weakness, given deteriorating housing affordability, and so we are preparing for meaningful price declines in some regions of the country. With that, we will now open the call up to questions. Operator, please go ahead.
Operator, Operator
Our first question comes from Bose George from KBW.
Mike Smith, Analyst
Hey, guys. This is actually Mike Smith on for Bose. My first question, can you just help us get a sense for run rate, core earnings kind of relative to that $0.44, just given your outlook for capital deployment and then the addition of Longbridge?
Larry Penn, CEO
Thank you for the question. Regarding the addition of Longbridge, I mentioned that we will start capturing ADE from Longbridge in Q4. We plan to present this as a separate operating segment and provide details on our calculation methods. The components of ADE at the originator differ from those at EFC, primarily because Longbridge focuses on origination and gain on sale, which we do not include in EFC's realized and unrealized figures. While the components will be different, we will display both separately. Due to the gain on sale aspect at Longbridge, their ADE is typically more volatile than EFC's. However, we anticipate that it will be beneficial over time, and the addition of this segment looks promising concerning our current dividend. As for the run rate, I believe 44, which is close to 12% of our book value, is a reasonable estimate moving forward. There are several factors involved in this calculation, including the overall size of the portfolio. This quarter, part of the sequential increase resulted from the credit portfolio expanding despite declines in Agency business, which was a significant factor. Additionally, as purchase yields improve into higher reinvestment yields, we expect this will contribute to topline growth in NIM. We have seen the cost of funds adjust to market conditions more rapidly, leading to what we anticipate will be a short-term contraction in NIM, along with adjustments related to interest rates and swaps. To summarize, I think 44 is a solid run rate, and we believe we will be able to maintain the dividend of $0.45, which we are currently close to meeting, especially with Longbridge being factored in.
Mike Smith, Analyst
Any expectation there?
Larry Penn, CEO
A few cents per share per quarter at least.
Mike Smith, Analyst
Great. That’s helpful. Thanks. And then maybe just one on book value, do you guys have a sense for how much of the decline over the last few quarters has been realized versus unrealized? Just kind of wondering how much potential for recovery there is in your book?
Larry Penn, CEO
Sure. A significant part of our losses this year has been unrealized, particularly within our agency strategy and our originator stakes, where we've marked them down below book value due to the transaction value of LendSure. The company has a strong cash position, and there is potential for improvement there. Additionally, non-QM has also contributed to losses this year with a notable unrealized component. In our income statement for the nine months ending September 30th, we reported unrealized losses of nearly $2.50 per share, whereas realized losses were around $1.18 per share. This means that approximately two-thirds of the losses are unrealized, and we believe there is a high chance of recovering a substantial portion if spreads tighten, even if yields remain elevated.
Mike Smith, Analyst
Great. That’s really helpful color. Thanks for taking the questions.
Larry Penn, CEO
Thank you.
Operator, Operator
And the next question comes from Doug Harter from Credit Suisse.
Doug Harter, Analyst
Thanks. Can you just talk a little bit about how you are thinking about balancing the deployment of cash versus kind of holding liquidity for volatility in the near term and kind of what you are looking for that might move you one way or the other in that decision?
Larry Penn, CEO
Mark, do you want to take that?
Mark Tecotzky, Co-Chief Investment Officer
Certainly. The balance between leveraging high yields and managing the portfolio for potential uncertainties, such as upcoming Fed announcements or developments from the Bank of England, was a significant focus in our discussions on company management during Q3. Overall, not much has changed. There might be some encouraging signs, as market expectations and comments from the Fed suggest that the pace of rate hikes is expected to slow, which the market may view positively. In my previous comments, I noted that certain markets are starting to demonstrate a better equilibrium of supply and demand. For instance, in the agency market, daily spread movements are now aligning more closely with credit indices, including investment-grade and high-yield indices, which we interpret as a sign of improved balance rather than just reacting to forced sell-offs. However, if we look back at Q3 and into October, there were segments of the securitized products market, such as legacy non-agencies, where we witnessed days when liquid credit indices tightened while bonds weakened. This indicates that supply and demand are still out of sync, with some forced selling from money managers. While we observe some indicators suggesting improvement since the third quarter, there are still areas in the market exhibiting imbalances. We continue to encounter forced sales at prices that do not align with our expectations for those securities. Additionally, we monitor the capital flow into the market. A notable aspect of 2022, distinguishing it from prior years, is the decline in bank purchases of securities, particularly in Agency MBS and AAA CLOs. This situation arises as banks grapple with reduced capital due to significant mark-to-market losses in their available-for-sale portfolios and issues with deposit erosion, resulting in diminished participation in securitized products compared to historical norms. Conversely, we've seen insurance companies emerging as new players with substantial capital, which we view positively. Overall, while we are cautiously optimistic and starting to see some signs of recovery, we still prioritize conservatism regarding cash reserves, feeling that while there are positive developments, they do not warrant a complete sense of reassurance just yet.
Larry Penn, CEO
If I could expand on that, we only increased the credit portfolio by about 3% last quarter. Looking ahead, the focus will be on replacing assets that are maturing with higher-yielding ones. We have various strategies in place, and RTL is currently one of our higher-yielding return on equity strategies, so we expect to see strong inflow there. However, as noted in our prepared remarks, the small balance commercial mortgage sector has seen a significant slowdown, and we are primarily replacing maturing loans rather than growing that portfolio. This allows us to be very selective in how we allocate our capital. We anticipate the agency portfolio will continue to decrease slightly, which we believe is sensible. Typically, we aim to maintain our cash positions between $100 million and $200 million, although this can fluctuate based on our unencumbered asset portfolio. Another point to consider is the non-QM securitization; we completed one in the third quarter, but as Mark mentioned, we've opted to hold more loans on repo due to unfavorable spreads in the non-QM securitization market. This could be a potential catalyst for acquiring more credit assets if we proceed with further securitization. Overall, I wouldn't foresee significant growth in the overall size of the credit portfolio in the short term.
Doug Harter, Analyst
Great. Thank you.
Operator, Operator
Our next question comes from Trevor Cranston from JMP Securities.
Trevor Cranston, Analyst
Hey. Thanks. You guys talked a little bit about seeing some forced selling towards the end of September. Can you talk a little bit about kind of what you guys are seeing in terms of secondary market investment opportunities and how you are evaluating deploying capital there potentially versus the more proprietary loan you guys have come in?
Mark Tecotzky, Co-Chief Investment Officer
Hey, Trevor. It’s Mark. That’s a great question and quite relevant. In 2020, we observed a significant decline in security prices. Often, the fall in securities outpaces the decrease in loan prices, prompting us to add securities to our portfolio. We continuously assess the relative value between securities and loans, and we find securities to be intriguing. In certain situations, particularly with selective investing, they can be much more attractive compared to loans. Another important aspect of securities in the current market is that when originating loans, you're tied to property valuations from November 2022. Conversely, when purchasing securities, you are acquiring seasonal securities. For instance, credit risk transfer is a good example. You can invest in CRT bonds that were issued in 2018 or 2019, where borrowers may have seen a 30% to 40% home price appreciation since taking out their loans, resulting in a lower loan-to-value ratio than at origination. We consider all these factors and have noticed that some securities have dropped in price to a point where they're very compelling in comparison to loans. Therefore, we are actively exploring both sectors. On the loan side, we've been tightening guidelines and have identified several metrics where we are becoming more cautious. Overall, I believe there is a significant opportunity in securities for EFC right now.
Trevor Cranston, Analyst
Got it. Thanks for that. You mentioned that you are using repo financing more for loans since the securitization market is a bit dysfunctional. Can you remind us how much repo capacity you have for loans and if there are any limitations regarding how long loans can be held on those lines? Thanks.
Larry Penn, CEO
Yeah. No, there is no limit on how long loans could be held on those lines. But sort of total capacity is not something that we disclose. But we do have now a wide variety of facilities, and we are not concerned about capacity at this point.
Trevor Cranston, Analyst
Okay. Fair enough. Thank you, guys.
Operator, Operator
Our next question comes from Eric Hagen from BTIG.
Eric Hagen, Analyst
Good morning. I hope you are well. I have three questions to ask. First, can you provide the balance of non-QM retained tranches, including the IO strips on the balance sheet? Second, could you discuss the repo financing for those subordinate tranches and other subordinate non-Agency RMBS, and how stable the haircut is there? Lastly, can you elaborate on how your dividend will be characterized this year, specifically the breakdown between the return of capital and the ordinary dividend you plan to pay? I believe you mentioned plans to reduce the agency portfolio; can you explain the reasoning behind that?
Larry Penn, CEO
We believe that the agency relative value is currently excellent. Spreads are quite wide, and particularly in the case of discounts, they are not really exposed to extension risk at the moment. In fact, we think we can identify and have been working to accumulate specified pools that should have quicker turnover rates. Therefore, the opportunities in agencies appear to be very strong, and we have no issues with agencies.
Eric Hagen, Analyst
Okay.
Larry Penn, CEO
We see a better leveraged return on equity and are maintaining a healthy cash position. As Mark mentioned, we're taking lessons from our COVID experience, which serves as a solid source of liquidity, and we aim to keep our liquidity levels high.
Eric Hagen, Analyst
Okay.
J.R. Herlihy, CFO
The fair value of the retained non-QM tranches was $138.5 million as of September 30th. We provided details on this in our portfolio summary slide, where we distinguish that component from non-QM.
Mark Tecotzky, Co-Chief Investment Officer
Slide four. Yeah.
J.R. Herlihy, CFO
Mark, do you want to comment on how financing for those retained tranches has changed regarding haircuts and availability this year?
Larry Penn, CEO
I think Mark can correct me if I am wrong, but I believe the haircuts are still around the 50% level in terms of what's available in the market, and we still haven't had any issues financing those. So, I haven't seen any material change in the terms we are observing regarding haircuts or availability.
Mark Tecotzky, Co-Chief Investment Officer
The point I wanted to make is that while the haircuts can be stable, they are still reductions based on current market levels. You must manage the mark-to-market volatility, even with stable haircuts and repo spreads, as they are essentially lending you 50% or 60% of fair market value. If fair market value changes, you could receive a margin call if it drops, or become a margin-call lender if it rises. It's interesting that the retained pieces were primarily interest-only heavy going into this year, which we evaluate by looking at the dollar price relative to principal. We noted that this segment of the overall portfolio has performed well this year. However, there has been a significant slowdown in non-QM speeds, and for us, the concern in the non-QM market hasn’t been financing costs but rather significant changes in spreads, particularly a widening in spreads on investment-grade bonds like AAA, AA, A, and BBB. This trend is visible across fixed income and within BSL CLOs and NPL RPL securitizations, where the top of the capital structures have widened considerably this year, far more than the investment-grade indices. Although repo spreads have increased slightly, they have not risen nearly as much as the spreads on securities. Traditionally, money managers and banks support the top of the capital structure, but as I mentioned earlier regarding banks, their market role has diminished this year while money managers have seen many redemptions. Consequently, the top of the capital structures needed to widen to draw in new buyers. While it is functioning, it has changed significantly. This situation is why we are managing our portfolio with more loans on repo than usual, especially as the term financing available through the securitization market has slightly wider spreads than what we anticipate we will achieve in the future.
Larry Penn, CEO
Let me add one more point that we haven't discussed before. You can actually achieve higher advance rates and lower haircuts on retained tranches if you opt for vertical retention. This approach allows you to hold a strip that includes many rated securities instead of just the bottom tier. We are considering this for our next deal, although it's still uncertain. The last time we completed a non-QM securitization was in July, but we might pursue vertical retention in our upcoming securitization. This strategy could also enable us to sell some of the interest-only securities generated from that securitization, which could be appealing, especially in a high-rate environment where there might be strong demand for IOs. The securitization market offers various options, though right now it seems limited due to the current conditions. Ultimately, we could see advantages such as higher advance rates and lower haircuts with vertical retention, as others in the market have successfully done the same.
Eric Hagen, Analyst
Yeah. Great perspective. Thank you guys very much.
Operator, Operator
Our last question comes from Crispin Love from Piper Sandler.
Crispin Love, Analyst
Thanks. Mark, you talked about the opportunity in securities, but I am curious about the potential for distressed loan acquisitions. Is this an area that you would expect to be more active in the next couple of quarters? And I guess have you started to see banks start to shed some CRE and non-core assets, and are there any other areas where there could be an opportunity here for you in early 2023, if not already?
Mark Tecotzky, Co-Chief Investment Officer
No, that's a great question. I would say yes to the potential for distressed loan acquisitions. Initially, our small balance commercial strategy at Ellington Financial focused on acquiring 100% non-performing commercial loans, particularly from the old Washington Mutual and GreenPoint portfolios, which came with significant challenges post-financial crisis. We launched this strategy in 2010, but opportunities dried up as the distress subsided. Currently, debt service coverage ratios on new loans are limiting the size of new loan balances, and sometimes, a new loan might actually have a smaller balance than one that wants to be paid off, even if the property is performing well. I anticipate that we will encounter more of these opportunities soon. On the bank side, I expect to see potential in unsecured consumer loans. Previously, we had a steady program for acquiring these loans, but we paused it over the last several quarters because we were being outpriced by credit unions and banks. However, we are now seeing unsecured consumer loan originators presenting packages of loans, often at significant discounts. There’s potential there. Additionally, we are also looking into packages of auto loans, which we have started to engage with. On the residential front, Ellington Financial has participated in non-performing or re-performing loan sales from the government-sponsored enterprises (GSEs). We're observing a modest decline in home prices which, if it accelerates, could lead to an increase in delinquent loans in the residential sector. However, I don’t anticipate the distressed levels approaching those of 2008 for various reasons. Distressed loan opportunities are indeed exciting for Ellington Financial moving forward. Many of these sectors have historically driven good returns, but from 2018 to 2021, there was a lack of distress, so we did not see much in that area. We are prepared and looking for these opportunities and are ready to allocate capital accordingly.
Crispin Love, Analyst
Thanks, Mark. All very helpful there. And then just one last question from me, so your CMBS hedging position increased from about $15 million or so to over $50 million, if I am looking at the chart on slide 17 correctly. I'm just looking at that. Are there any cracks that you are beginning to see in any CRE sectors, where you have worries, or is it more just general uncertainty right now in the current environment?
Mark Tecotzky, Co-Chief Investment Officer
There has been extensive discussion about office spaces due to the rise of remote work, and retail has faced ongoing challenges. The main concern now is that operators seeking new loans are facing interest rates that could be up to 400 basis points higher than when they initially purchased the property, especially if they had taken out a bridge loan. If they didn't secure a 10-year loan and are instead stuck with a floating rate, they might be dealing with rates like LIBOR or SOFR plus 350 basis points, while our bridge spreads are even higher. These were originated when LIBOR was nearly at zero, but now they might be facing SOFR or LIBOR rates around 4.5% to 5%. This means that debt costs could effectively double. If the income from those properties hasn't kept pace with the rising debt costs, their debt service coverage ratio will decline. If it falls significantly, when the loan matures, lenders may hesitate to provide the same amount as before. This is a challenge the market will need to address, but I see it as an opportunity for us, and we are concentrating on this.
Larry Penn, CEO
You should view the CMBS hedge as a safeguard for our small balance commercial portfolio, which remains substantial and stable in size. Approximately 70% of this portfolio is multifamily, and we believe there continues to be strong support for multifamily values. However, the remaining 30% includes categories like office, which could face challenges. Therefore, this hedge mainly protects that portfolio and highlights Ellington Financial's unique approach to implementing credit hedges in uncertain market conditions.
Crispin Love, Analyst
Thanks, Larry, Mark. I appreciate the comments there.
Mark Tecotzky, Co-Chief Investment Officer
Sure. Thank you.
Operator, Operator
That was our final question for today. We thank you for participating in the Ellington Financial third quarter 2022 earnings conference call. You may disconnect your line at this time and have a wonderful day.