Ellington Residential Mortgage REIT

NYSE:EARN

Earnings Call Analysis

Ellington Residential Q4: Strong Returns, Lower Leverage

In Q4, Ellington Residential posted a net income of $0.75 per share, with an economic return of 7.7%. Adjusted distributable earnings rose to $0.27, covering dividends comfortably. Efficient risk management allowed for capital shifting from mortgage-backed securities (MBS) to collateralized loan obligations (CLOs), enhancing the CLO portfolio by $13.6 million. The financial strategy led to a reduced debt-to-equity ratio from 7.3:1 to 5.3:1 and expanded the net interest margin from 1.34% to 2.19%. Agency RMBS holdings declined by 8% to $728 million, while non-Agency and interest-only securities decreased by 13%. Conversely, CLO holdings increased more than fourfold to $17.4 million.

Changing Tides in the Fourth Quarter Reflect an Eventful 2023

2023 was a roller coaster for markets, with the fourth quarter exhibiting the kind of dramatic swings investors have grown accustomed to. October saw a peak in interest rate volatility and widening yield spreads, only for a November-December rally to stabilize conditions, drive capital into fixed income, and tighten spreads. EARN, focusing on Agency and non-Agency RMBS along with corporate CLOs, found conditions favorable in fixed income spreads. Despite interest rates peaking in October, the year concluded almost precisely where it began, amplifying the quarter's calmer end.

EARN's Strategic Moves Yield Positive Financial Results

A pivotal move for EARN this quarter was to sidestep forced sales during the market volatility in October, preserving the equity and earning power to benefit from a subsequent market rally. This strategy translated to EARN's impressive net income of $0.75 per share and a 7.7% nonannualized economic return. A smart relocation of capital from MBS to CLOs underscored the quarter, growing the CLO portfolio by $13.6 million, and fostered EARN's adjusted distributable earnings (ADE) past dividend coverage to $0.27 per share.

Net Interest Margin Expansion and Lower Leverage Ratios

EARN reported an expanded overall net interest margin to 2.19%, a notable increase from the previous quarter's 1.34%. Particular strengths were seen in their agency NIM and credit NIM, the latter including CLOs and non-agency RMBS. Furthermore, the pivot towards CLOs reduced the quarter's leverage ratios, and a more diverse and profitable portfolio emerged, all while net losses were hedged through interest rate swaps and maintaining a strategic net short position as measured by 10-year equivalents.

A Look Ahead: CLO Strategy and Optimism Despite Early 2024 Setbacks

With an eye on the future, the CLO strategy is expected to be less volatile and contribute positively to EARN's returns. Despite an early 2024 setback with a negative 1.9% estimated economic return through February due to the Agency MBS market's underperformance, EARN foresees an upturn. Growth in the CLO portfolio by an additional 70% in early 2024, the potential for rate cuts, ongoing spread widening, and lower volatility present compelling reasons for optimism in the overall outlook of EARN's performance.

Earnings Call Transcript 2023-Q4

[Operator]

Good morning, ladies and gentlemen. Thank you for standing by, and welcome to the Ellington Residential Mortgage REIT 2023 Fourth Quarter Financial Results Conference Call. Today's call is being recorded. [Operator Instructions] It is now my pleasure to turn the conference over to Alaael-Deen Shilleh, Associate General Counsel. Sir, you may begin.

[Alaael-Deen Shilleh, executive]

Thank you. Before we begin, 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 and 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. We strongly encourage you to review the information that we have filed with the SEC, including the earnings release and the Form 10-K for more information regarding these forward-looking statements and any related risks and uncertainties.

Unless otherwise noted, 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. Joining me on the call today are Larry Penn, Chief Executive Officer of Ellington Residential; Mark Tecotzky, our Co-Chief Investment Officer; and Chris Smernoff, our Chief Financial Officer. As described in our earnings press release, our fourth quarter earnings conference call presentation is available on our website, earnreit.com. Our comments this morning will track to the presentation. Please note that any references to figures in this presentation are qualified in their entirety by the notes at the back of the presentation.

With that, I will now turn the call over to Larry.

[Laurence Penn, executive]

Thanks, Alaael-Deen, and good morning, everyone. We appreciate your time and interest in Ellington Residential. As with much of 2023, in the fourth quarter, markets gyrated between a selloff and a rally with tumultuous October giving way to a market rally in November and December. In October, interest rate volatility spiked as U.S. treasury yields rose to 15-year highs and that drove yield spreads sharply wider on most fixed income products. Markets then reversed course in anticipation of the conclusion of the Federal Reserve's hiking cycle, with interest rates and volatility, both declining into year-end. With rates lower, and trading in a more stable range, demand for spread products picked up and capital flowed into fixed income funds. With the notable exception of CMBS, which has its own unique challenges, virtually all fixed income spreads tightened for the fourth quarter, including in the markets where EARN invests, namely Agency and non-Agency RMBS and now corporate CLOs, but we've been investing to an ever-increasing extent after our recent pivot.

Turning to the investor presentation. On Slide 3, you can see that medium and long-term interest rates, despite spiking to multiyear highs in October, actually declined overall for the quarter. And the 30-year Freddie mortgage survey rate despite reaching a 23-year high mid-quarter also finished lower on the quarter. Incredibly, despite all of the fluctuations during the year, both the 10-year treasury yield and the 30-year mortgage survey rate finished 2023 within 1 basis point of where they started the year, as you can see here on this slide as well.

As the backdrop for our mortgage-backed securities portfolio, you can also see on Slide 3 that option adjusted yield spreads tightened across agency coupons during the fourth quarter, and that the most pronounced price increases were on lower and intermediate coupons. Dollar prices on Fannie $2.5s through $4.5 were up more than 5 points sequentially. The outperformance of those coupons benefited EARN's Agency MBS portfolio specifically because coming into the quarter, roughly 2/3 of our Agency MBS had coupons of 4.5% or less. Meanwhile, as the backdrop for our CLO portfolio, corporate credit spreads followed a similar pattern, first widening in October and then tightening in November and December and tightening overall for the quarter as an economic soft landing narrative [indiscernible] the market. You can see on the bottom of Slide 3 that credit spreads on both high yield and investment grade tightened significantly over the quarter while prices on the Morningstar LSTA Leveraged Loan Index rose.

Turning now to EARN's results. In the fourth quarter, we generated net income of $0.75 per share and a nonannualized economic return of 7.7%, while our adjusted distributable earnings grew to $0.27 per share and more than covered our dividend. As with other market disruptions we've seen before, the key in the fourth quarter was to avoid force-selling when the market sold off in October in order to preserve equity and earnings power and be in a position to participate in the subsequent market recovery. In the fourth quarter, we again relied on EARN's risk management and strong liquidity position to accomplish this. That said, we did sell pools in the fourth quarter to free up capital from MBS to CLOs and the majority of our sales took place in November as yield spreads were tightening. We ended up increasing the size of our CLO portfolio by $13.6 million during the quarter.

On Slide 12 of the earnings presentation, you can see some of the underlying characteristics of our CLO portfolio as of year-end. The corporate loans underlying our CLO investments span a diverse array of industries and the overwhelming majority are floating rate first lien senior secured loans. Our rotation into CLOs has continued into the new year with our agency portfolio now incrementally smaller and the size of our CLO portfolio now up an additional 70% from year-end to approximately $30 million. Even after the recent credit spread tightening in the sector, we still see returns on equity for new CLO investments in the high teens to low 20s.

Besides contributing to and diversifying EARN's GAAP results, our high-yielding CLO investments have also helped drive the substantial growth of our net interest margin and thereby have supported our ADE as well. In addition, because we employ less leverage on our CLOs compared to agency, the portfolio rotation has also driven down our leverage ratios. At year-end, our debt-to-equity ratio adjusted for unsettled trades declined to 5.3:1, down from 7.3:1 at September 30.

I'll now pass it over to Chris to review our financial results for the fourth quarter in more detail.

[Christopher Smernoff, executive]

Thank you, Larry, and good morning, everyone. Please turn to Slide 5 for a summary of Ellington Residential's fourth quarter financial results. For the quarter ended December 31, we reported net income of $0.75 per share and adjusted distributable earnings of $0.27 per share. AD excludes the catch-up amortization adjustment which was positive $566,000 in the fourth quarter. During the quarter, positive net interest income and net gains on our Agency MBS significantly exceeded net losses on our hedges, driving strong performance from our agency portfolio. Our CLO portfolio also generated strong returns, driven by net interest income and net gains as did our non-Agency RMBS and interest-only portfolios.

On Slide 5, you can see that our overall net interest margin expanded to 2.19% from 1.34% quarter-over-quarter, which drove the increase in ADE. Broken out by product, our agency NIM increased to 2.02% from 1.26%, driven by higher asset yields and a lower cost of funds. Meanwhile, our credit NIM, which includes CLOs and non-agency RMBS, increased to 6.28% from 4.55%, boosted by high asset yields on our larger CLO portfolio.

Please turn now to our balance sheet on Slide 6. Book value per share was $7.32 at year-end compared to $7.02 per share at September 30. Including the $0.24 per share in dividend in the quarter, our economic return for the quarter was 7.7%. We ended the quarter with $61 million in cash plus unencumbered assets, which was approximately 45% of total equity. Next, please turn to Slide 7 for a summary of our portfolio holdings. Our Agency RMBS holdings decreased by 8% sequentially to $728 million as of December 31 as net sales and paydowns exceeded net gains. Our Agency MBS portfolio turnover was 25% for the quarter. Our aggregate holdings of non-Agency RMBS and interest-only securities also shrunk in size by 13% quarter-over-quarter. Over the same period, we increased our CLO holdings more than fourfold to $17.4 million as of December 31 compared to $3.8 million as of September 30.

At year-end, our deployed equity was allocated 89% to mortgage-related securities and 11% to CLOs. Our debt-to-equity ratio adjusted for unsettled trades decreased to 5.3x as of December 31 as compared to 7.3x as of September 30. The decline was primarily due to an increase in shareholders' equity and a significantly lower leverage on the CLO portfolio relative to our agency holdings. Similarly, our net mortgage asset-to-equity ratio decreased to 6.5x from 7.2x over the same period despite our holding a net long TBA position at December 31 as compared to a net short TBA position at September 30.

On Slide 9, you can see the details of our interest rate hedging portfolio. During the quarter, we continue to hedge interest rate risk primarily through the use of interest rate swaps. We ended the fourth quarter with a net long TBA position on a notional basis, but a small net short position as measured by 10-year equivalents. Lastly, on Slide 12, you can see that nearly all of the loans underlying our CLO portfolio, our floating rate and as such, carry minimal interest rate risk.

I will now turn our presentation over to Mark.

[Mark Tecotzky, executive]

Thanks, Chris. The fourth quarter was really a tale of 2 markets. The first part of the quarter was characterized by continued rate sell-off, wider spreads, fund outflows and market uncertainty about how high the Fed would need to hike short rates before achieving a noticeable improvement on inflation. But then starting in late October and early November with economic indicators and comments from Chairman Powell pointing to a possible end to the rate hike cycle, markets started to make a U-turn. Rates dropped, spreads tightened and there was a significant fund in bank buying of Agency MBS and other spread products. In agency, sector outperformance in the second part of the quarter exceeded underperformance in the first part. And overall, for the quarter, Agency MBS significantly outperformed hedging instruments. I'm happy to report that EARN is well positioned to capture this agency outperformance, posting a total economic return of almost 8% for the quarter.

During the market sell-off in the first part of the quarter, we were able to manage the interest rate volatility and keep our Agency MBS portfolio largely intact. We were confident that it was just a matter of when, not if spreads will recover, and maintaining our portfolio allowed us to capitalize on the spread tightening when it did eventually occur. During Q4, the market pivot and Fed expectations was the catalyst that led to lower implied and realized volatility, which lowered actual and expected hedging costs and prompted capital inflows from banks and investment funds. If and when the first rate cut occurs later this year, we think that could be another catalyst for continued outperformance for Agency MBS. We were able to take advantage of the market strength to shrink our Agency MBS portfolio incrementally and redeploy that capital into CLOs. That rotation not only enhanced our diversification but it also took our leverage down significantly, and yet we were still able to grow ADE.

In Q4, our CLO portfolio grew by $13.6 million as we predominantly added seasoned CLO mezzanine tranches, but also longer-duration CLO equity, shorter-duration CLO equity and newer vintage CLO mezz. Seasoned mezzanine investments outperformed throughout Q4 as prepayment speeds accelerated and CLO cash balances grew, driving expectations of a deal deleveraging in January and that strong performance has continued into 2024. CLO credit spreads tightened across the board in Q4 with BBBs generally rallying around 30 to 50 bps and BBs rallying even more, albeit with significant dispersion. However, these sectors lagged the high-yield corporate bond market, whereby some measures, spreads tightened almost 100 basis points for the quarter.

Q3 earnings were better than expected for many high-yield borrowers with JPMorgan estimating that 86% of high-yield companies generated Q3 earnings that were either neutral or positive for their credit profiles. And investors generally grew more comfortable with noninvestment-grade credits in Q4 as fundamentals improved. Improvements in the leveraged loan market drove strength in junior CLO tranches given that they are more levered to credit performance than senior tranches are. This said, the most credit-sensitive CLO profiles that is those with the lowest credit enhancement and/or most of the stressed portfolios continued to lag as investors anticipated further credit losses.

In Q1 of 2024, we anticipate further strengthen our CLO portfolio due to declining credit [indiscernible] and continued pull to par in seasoned CLO mezz. Approximately 40% of the leveraged loan index traded above par at the end of Q4, which has incentivized lots of borrowers to refinance their debt so far in 2024. This is benefiting both seasoned CLO mezz to faster deal paydowns and CLO equity to lower near-term default risk as underlying corporate borrowers raise incremental liquidity. We also expect the technical backdrop for the leveraged loan market to remain attractive as many new CLOs are expected to ramp up portfolios in Q1, driving demand for loans with a forward calendar of loan supply that remains light.

Looking ahead, we see lots of reasons to be optimistic about EARN's future performance. The most aggressive Fed hiking campaign ever is now behind us. SOFR went from 0 to over 5% in 14 months. The Fed balance sheet has shrunk by well over $1 trillion since its peak post-COVID size. Coupled that with large bank failures putting almost $100 billion of Agency MBS and CLOs into the market and you had the recipe for substantial spread widening, which we've seen over the past couple of years. But now the Fed should soon become a tailwind as opposed to a headwind. And looking ahead, in addition to this expected support from the Fed, we see 5 major factors supporting future MBS performance. First, spreads are wide. Not as wide as October but still much wider than historical averages, and they should be the Fed as a seller, not a buyer and banks while buying are a shadow of their former selves. But being wide and staying wide works out just fine for EARN, we have a big levered NIM to capture.

Second, supply is low, and it's especially low relative to the mountain of Treasury supply. So relative performance versus hedging instruments are supported by this technical. Third, prepayment risk for most coupons is benign and the cost of prepayment protection is reasonable. Fourth, flows into mutual funds that buy Agency MBS, both active and passive have been quite strong as have fixed income annuity sales. Banks have also started to buy in Q4. And fifth, volatility is a lot lower, both actual and implied, so delta hedging costs are lower, and that makes option-adjusted spreads wider. We have room to add leverage at EARN. We have tools to further grow ADE and CLOs are helping to deliver a diversified return stream.

Now back to Larry.

[Laurence Penn, executive]

Thanks, Mark. I was pleased with how we navigated the market gyrations throughout 2023 and finished the year on a high note. Now with yield spreads still wide on a historical basis with markets expecting cuts instead of hikes and with volatility normalizing, Agency MBS are attracting incremental demand from investors, albeit tempered by uncertainty around the timing of cuts. When those rate cuts eventually come, and we ultimately see a steep yield curve again, that should be a further tailwind to the sector.

I'm excited about our growing corporate CLO portfolio. EARN's small size and liquid portfolio has been an advantage year as we've been able to ramp up quickly in a terrific strategy where we see a big opportunity for EARN. While Ellington has long-standing and deep experience managing CLO portfolios, EARN began investing in that product just this past September, and that pivot is already contributing nicely to earnings.

Including investments through today, CLO investments are now a full 17% allocation of EARN's total equity. I expect them to be a significant driver of earnings and ADE moving forward. The CLO market has proven its ability to generate attractive returns over market cycles and over a long-term horizon.

In the short term, and as Mark mentioned, credit spread tightening in some segments of the CLO market has continued to lag a larger rally in corporate bond credit spreads, and we expect that money manager inflows into high-yield and leveraged loans will help narrow that gap. As Mark also mentioned, we expect faster leveraged loan prepayment speeds to drive faster prepayments on many of our seasoned CLO mezzanine positions, which we hold at significant discounts to par.

We've grown EARN CLO portfolio by another 70% so far in 2024, and these discounted seasoned CLO mezzanine positions have continued to be a focus of ours, given their total return potential in an environment with this potential for higher loan prepayment speeds. The CLO market suits EARN extremely well. It not only offers high current interest income, but it has always been a fertile ground for both relative value and absolute value opportunities as well as for trading opportunities given the dispersion in collateral credit performance from deal to deal. Going forward, Ellington's extensive expertise and track record in the CLO market should be a big benefit for EARN.

Our CLO portfolio has continued to contribute nicely to our results so far in 2024. But net losses in Agency MBS have led to an overall economic return for EARN that we currently estimate at negative 1.9% year-to-date through February. The Agency MBS market has underperformed many other fixed income sectors so far in 2024, driven by higher rates and uncertainty around the timing of Federal Reserve rate cuts. This, of course, follows a strong fourth quarter for Agency MBS, which led to the positive 7.7% nonannualized economic return that we generated last quarter.

For the past few years, it's undeniable that Agency MBS has been a volatile strategy for all the agency mortgage REITs, including EARN. But I firmly believe that our CLO strategy will prove to be a less volatile strategy and thereby, stabilize and enhance EARN's returns over time.

And with that, we'll now open the call to questions. Operator, please go ahead.

[Operator]

[Operator Instructions] Our first question will come from Doug Harter with UBS.

[Douglas Harter, analyst]

Can you talk a little bit more about how you see the equity allocation to CLOs playing out, kind of how much of equity could this be?

[Laurence Penn, executive]

Yes. So I guess it's -- I'm not going to answer that question directly. I'm just going to say that I love the strategy. We have a great team here and a great track record. The more the better, as far as I'm concerned. We -- as you know, we have constraints that we operate under as a REIT, the REIT [indiscernible] mostly more sort of capital allocation-based, but it also depends on your sort of financing strategy, your subsidiary structure. So we continue to grow, and I don't want to try to forecast where it's going.

[Douglas Harter, analyst]

Okay. And then I guess just as you think about risk-adjusted returns kind of comparing agency and CLOs. How do you think about kind of the upside, downside in each kind of the base case and kind of how that factors into kind of where you want to be?

[Laurence Penn, executive]

Sure. Well, first of all, from a NIM perspective, you can see that I think Chris mentioned our -- the NIM on our Agency portfolio was probably in the low 200s, something like that and the credit portfolio, including CLOs, I think 600s. So even with a small bit of leverage, the CLO portfolio generates greater leverage NIM dollar-for-dollar. Spreads have tightened, you can sort of see -- if you go back to Slide 3, you can sort of see it differs based on coupon, but if you look at 4.5s, for example [indiscernible] looking at year-end, something around the 9,100 level, even if you leverage that a fair bit, you're still not quite there versus where you are on the CLO portfolio on a leverage term basis.

So now the CLO portfolio obviously has kind of tail credit risk in a -- especially a deep recession or something like that. So that's a factor for sure. It also has much fewer delta hedging costs, right? I mean the thing that's been really tough for agency REITs, including us for the last few years is these big moves in the market, big gyrations in rates cause -- and for us, we like to stay hedged on interest rates, and that's, frankly, over a long period of time, that's a key to our success. But it really cuts into book value over time, those delta hedging costs.

And ultimately, that's going to cut into your dividend as well, right? So -- so yes, so the other thing though I would say is from a trading perspective, given the liquidity of agency, the opportunity for trading profits, including dialing up and down to what extent we hedge with TBAs is probably that opportunity is probably greater in agencies because if you can make a point -- just this is hypothetical, if you can make a point on your assets, by being timely in terms of when you add, when you take off, how you hedge, that's leveraging that 7x or 8x, I mean that's a massive amount boost to earnings.

So there's a lot of pros and cons for each sector. And there are opportunities for trading gains as well in CLOs, but it's not -- I wouldn't say that the assets that we're buying there are super liquid. So it's -- they're different. And like I said, I love the strategy, and I'd love to continue to see this grow as much as possible in CLOs.

[Operator]

Our next question will come from Mikhail Goberman with JMP.

[Mikhail Goberman, analyst]

First question, I guess, is, did I hear correctly that you said book value was down about 1.9% thus far this year? Was that total economic return?

[Laurence Penn, executive]

Total economic return.

[Mark Tecotzky, executive]

Yes, I think that was through the end of February.

[Laurence Penn, executive]

Right.

[Mikhail Goberman, analyst]

So including the 2 dividends. What -- are you guys targeting a sort of leverage ratio range going forward, given the sort of dip -- the pretty dramatic dip that the leverage ratio took in the fourth quarter, is there an area that you're sort of trying to get to?

[Laurence Penn, executive]

It's just a blend of really the appropriate leverage for each asset class, right? So for agencies, we've taken it up into the 9s before in terms of leverage. And in non-agencies and now in CLOs, we've had almost no leverage. We -- but we'll start to add leverage to CLO portfolio at some point. And as I -- well, maybe I didn't mention it exactly in answering the first question, but you can comfortably have [indiscernible] leverage on CLOs. So it's just going to be a blend of where the capital allocation is. And of course, in agencies, we also -- on the opportunity that kind of upper limit of something in the 9:1 leverage area, we can always go lower than that. And as I said a few seconds ago, sometimes we do that because we do want to take advantage of the spread volatility to sort of buy low and sell high, if you will, throughout the year.

[Mikhail Goberman, analyst]

Got you. And one more, if I could. This is sort of a longer-term question. What are your guys' thoughts on what kind of interest rate environment it would take for -- to see a meaningful pickup in prepay speeds sort of back towards, I guess, historical levels of a few years ago?

[Laurence Penn, executive]

Mark?

[Mark Tecotzky, executive]

Yes. So it's sort of interesting. We got -- we get the prepayment reports monthly, and we got one last night. And what I would say is that while prepayment speeds sort of in aggregate have been benign, technological improvements including AI is definitely something that is being embraced by some of the big nonbank lenders. And so there are pockets of the market. If you look at sort of -- there are Ginnie pools out there now that pay over 80 CPR from these real high coupon ones with a lot of VA.

What we were getting at -- or what I was getting at in the prepared remarks is that if you -- there's the vast majority of what's out there is still a couple of hundred basis points away from being refinanceable. There's -- you have a lot of runway in rates for many of the coupons that we own that [indiscernible] prepayments.

And the other thing I would say is that it's not real expensive to buy prepayment protection. So it's not as though prepayments don't exist and mortgage bankers aren't going to be aggressive about trying to solicit refis when they can, they certainly will be. It's just the vast majority of the market and the vast majority of what we own, you still have a lot of -- it would take still a significant move in rates to get it refinanceable. And so -- but still rates are unpredictable, implied volatility is high. So even the forward curve predict rates are coming down. There's a lot of dispersion around it.

So still for most coupons, in some of the higher coupons, we're still choosing to buy pools with prepayment protection. We think it's relatively affordable. And one thing that was a big part of last year was looking for pools. I think we mentioned it a few calls ago, looking for specified pools that have faster than market projected prepayment speeds, and that certainly helped us to get into things like 2.5s and 3s and 3.5s, we have big discounts, even finding pools, you are 1 or 2 CPR faster makes a huge difference in yield.

So prepayment modeling, prepayment speeds, prepayment risk, it's still out there. It's just right now given the current distribution of coupons in the market, the real scary prepayments are only affecting a very small subset about what's out there, and it's a sector that we have avoided, but we -- I'm not surprised to see some of these very fast prepayment speeds because there's still a lot of capacity within the mortgage banking community and the technology has gotten better. And so we look for them to be aggressive whenever they have opportunities to be aggressive on refis.

[Operator]

And our next question will come from Matthew Erdner with Jones Trading.

[Matthew Erdner, analyst]

Kind of following up on that CPR, you guys took off some of the lower coupons during the quarter. Could you kind of talk to the thoughts there?

[Mark Tecotzky, executive]

Yes, we just saw opportunities during the quarter to rotate up in coupon. I think portfolio went up about 10 basis points or so in coupon. The relative performance of lower coupons versus current coupons is really impacted by flows into the big bond [indiscernible] so if you look at mortgage index, if you look at the distribution of mortgage coupons and something like the Barclays agg, it's heavily weighted to 2s and 2.5s. And so the relative performance of those coupons is very much impacted by flows into the agg.

And there's a couple of real big $100 billion-plus ETFs that track the agg. So we saw opportunities where a lot of flows came into agg type portfolios where lower coupons outperformed. We saw an opportunity to go up in coupon, we thought it would add -- it certainly adds ADE. We thought it would also add total return. So we'll continue to be opportunistic about that.

[Matthew Erdner, analyst]

That's helpful there. And then allocating capital to CLOs, continuing that strategy. Are there any other places where you feel like you guys could opportunistically deploy additional capital for total return?

[Laurence Penn, executive]

That's where we're focusing [indiscernible].

[Operator]

And our next question will come from Eric Hagen with BTIG.

[Eric Hagen, analyst]

One more on just kind of the market dynamics. I mean what do you feel like would support more dollar roll specialness in the market? Do you feel like it's reasonable to expect that, that could come back if there's any changes to Fed policy?

[Mark Tecotzky, executive]

I would say that for some of these higher coupons, I think the dollar rolls have predictably gotten weaker because you've sort of built up now an array of kind of faster-paying pools. If you look at things like Fannie [indiscernible] now that they've been production for a few months, you have some pools that have come up the seasoning ramp. So I don't look for specialists to be there. In the lower coupons, you can get specialness if you get a lot of [indiscernible] about into these agg indices because when money comes into an agg index and whoever is managing it and wants to buy a bunch of Fannie 2s and 2.5s to get there to minimize tracking error, you really need to get that from other investors, right?

They initially will buy from the primary dealers, but the primary dealers are then going to want to buy that from other portfolios. And a lot of those bonds are locked up in the Fed and locked up in banks. There's obviously news on Truist a week or so ago. And when that news came out, it caused lower coupons to underperform.

So I think in the lower coupon stuff, you can see some more volatility in the roles. But still there, we tend to find -- if you're thoughtful about fast-paying pools, you can find pools that pick up [indiscernible] so I don't see a lot of specialists going forward for the higher-end production coupons where you're starting to see a float that's faster. And I think for the lower coupon stuff, you can see it, but it's kind of month-to-month, like there was a month ago. I think the Ginnie [indiscernible] was special.

So you get sort of these one-off things, but where most of the production is centered I don't think you're going to have a lot of specialists, which is really a big difference from the 2021 period where production was Fannie 2s and 2.5s and they're routinely special, and everyone spoke about that, and they were special because the Fed and banks are buying so much and they weren't rollers. That -- we don't have that dynamic right now. So the way the rolls work out, I think it favors in aggregate having more specified pools and less TBA. And I think broadly, that's sort of how the industry has moved, the REIT industry.

[Eric Hagen, analyst]

Yes. That's a good perspective. As far as running the portfolio, I mean, what do you guys feel like as a minimum level of liquidity you feel comfortable with having it at these spread levels? Like how much do you feel comfortable with ahead of the Fed cut versus like a cut actually taking place? Do you feel like that would be a catalyst to carry more leverage? Or is it potentially conditional on other factors?

[Mark Tecotzky, executive]

We mentioned in the prepared remarks that if you look at work on Bloomberg or whatever else, people are predicting cuts this year. I think if and when they actually happen, I do think that will be a catalyst for some incremental buying, so I do think that will be supportive. In terms of how we manage our cash, we don't.

That is sort of sacrosanct in that we're going to manage our cash according to market stresses and according to repo roll calendar, and we won't change sort of the guardrails, we have around minimum levels of cash as a function of what the Fed is going to do that. We just -- that's how we run -- that's how we run EARN, that's how we run EFC, that's how we run private funds that there is a certain amount of cash we're going to keep on hand. And the way we determine that is it's sort of -- look at the portfolio and look at the leverage and look at where things can go in a shock.

And then with the agency stuff, you have to look at there's a calendar to it. So you get new factors when the prepayments come out. So we got new factors [indiscernible] repo lenders and margin [indiscernible] versus a lower balance, but you don't actually get that cash to the 25th. So there's a calendar component to it. But -- we have room to add leverage. We mentioned that in the prepared remarks, but it's not because we're going to change the way we manage the cash. But I'd say sort of the way we manage the cash is something that we've been doing for a long time, and I think it served us very well, certainly served us extremely well in stresses like during COVID. But away from that, EARN has plenty of liquidity too. It's been putting more capital in the CLOs we've spoken about, but it still has more room to leverage just the general pool strategy as well.

[Operator]

And that was our final question for today. Thank you for participating in the Ellington Residential Mortgage REIT Fourth Quarter 2023 Earnings Conference Call. You may disconnect your line at this time, and have a wonderful day.