Ellington Financial Inc (EFC) 2021 Q1 法說會逐字稿

Thank you for standing by, and welcome to the Ellington Financial First Quarter 2021 Earnings Conference Call. Today's call is being recorded. (Operator Instructions)

It is now my pleasure to turn the call over to Jason Frank, Deputy General Counsel and Secretary. Sir, 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 filed on March 16, 2021, 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, please turn to Slide 3, and I will now turn the call over to Larry.

Thanks, Jay, and good morning, everyone. As always, thank you for your time and interest today. Ellington Financial continued its strong performance through the first quarter of 2021, significantly increasing book value and core earnings as we built on our profitable 2020 and the strong momentum we had coming into 2021.

Turning to Slide 3, you can see that for the quarter, we generated net income of $0.86 per share, good for an annualized economic return of 21%. We also generated core earnings per share of $0.43, which was up 16% sequentially and covered our new dividend run rate, which we increased by 40% last month. The continued growth of our high-yielding loan portfolio drove the sequential increase in core earnings. We also had significant gains in our CMBS and CLOs, sectors that have steadily recovered following the COVID-related stress of early 2020. We also generated significant earnings in our non-QM loan business, where we completed our first securitization deal of the new year. I'm delighted to say that we executed that securitization at the tightest yields of any of our deals so far. Finally, our consumer loan, residential transition loan and non-Agency RMBS portfolios also had excellent performance this quarter.

Besides growing our proprietary loan portfolios during the quarter, which we did to the tune of about 21% across small balance commercial mortgage, residential transition and consumer loans, we also used our strong balance sheet to rotate capital opportunistically to where we see the best relative value. We monetized gains in some of the more liquid parts of the credit portfolio. In fact, several of our CMBS sales were at prices that were actually higher than pre-COVID levels, and we redeployed that capital to acquire more loans from our pipelines and also added some attractively priced Agency-specified pools as wider Agency yield spreads provided an attractive entry point.

While those Agency pool purchases did cause our recourse debt-to-equity ratio to increase to 2:1 from 1.6:1 quarter-over-quarter, that leverage ratio is still well below the 2.7:1 that we averaged during 2018 and 2019. As it stands today, we have significant additional capacity in many of our financing facilities, and this should enable us to continue to be opportunistic with our investment approach and to grow earnings further. We're very comfortable taking leverage up from here, especially to fund more loan growth.

With that, I'll pass it to JR to discuss our first quarter financial results in more detail.

Thanks, Larry, and good morning, everyone. I'll continue on Slide 3 of the presentation. For the quarter ended March 31, Ellington Financial reported net income of $0.86 per share and core earnings of $0.43 per share. These results compare to net income of $1.44 per share and core earnings of $0.37 per share for the prior quarter. On April 5, we announced the increase of our monthly dividend to $0.14 per share from $0.10.

Next, please turn to Slide 5 for the attribution of earnings between our credit and Agency strategies. During the first quarter, the credit strategy generated a total gross profit of $1.14 per share, while the Agency strategy was roughly breakeven. These results compare to $1.69 per share in the credit strategy and $0.13 per share in the Agency strategy in the prior quarter.

The 2 primary drivers of the excellent results in our credit portfolio during the first quarter were: number one, higher net interest income quarter-over-quarter, which was a result of larger small balance commercial mortgage, residential transition, non-QM and consumer loan portfolios, combined with lower financing costs; and number two, significant net realized and unrealized gains, which were mainly in our CMBS, CLO and non-QM strategies as well as our equity investments in mortgage originators.

While our U.K. nonconforming RMBS portfolio generated gains for the quarter, our euro-denominated RMBS portfolio generated losses. Finally, our credit hedges detracted from results as credit yield spreads tightened during the quarter. Our Agency strategy was roughly breakeven for the quarter in the face of a challenging market marked by sharply higher long-term interest rates, increased volatility and a steepening yield curve. Agency RMBS durations extended and yield spreads widened and most Agency RMBS prices declined sharply, particularly for lower coupon RMBS.

The increase in long-term interest rates also reduced the demand for prepayment protection, which caused our prepayment-protected specified pools to further underperform. Meanwhile, the rise in long-term interest rates drove net gains on our interest rate hedges and Agency interest-only securities, which along with net carry from the portfolio more than offset the net losses on our long Agency RMBS holdings.

Turning next to Slide 6. Our total long credit portfolio decreased by approximately 9% to $1.3 billion in the first quarter. The quarter-over-quarter decrease was due to opportunistic sales of CLOs, CMBS and European RMBS as well as the completion of a non-QM loan securitization in February. Other portions of the portfolio did grow sequentially, however, including our small balance commercial mortgage, residential transition and consumer loan portfolios, which grew by a combined 21%.

Moving now to Slide 7. As Larry mentioned, we capitalized on Agency yield spread widening during the quarter to add some attractively priced specified pools, which increased our Agency portfolio by 55% to nearly $1.5 billion. In conjunction with these purchases, we also increased the size of our short TBA position significantly.

Flipping back to Slide 4 now, you can see the impact of these portfolio changes on our asset yields. Not surprisingly, you can see here that the market yield on our Agency portfolio at 2% is significantly lower than that on our credit portfolio, which was 9.8% at March 31. A 2% yield on our Agency portfolio is actually a 40 basis point increase from year-end, but because we grew our Agency portfolio significantly this quarter, we are now allocating 21% of our equity to Agency versus 15% last quarter. This larger Agency allocation brought down the blended yield on the overall portfolio. I would also note that the average market yield on our credit investments actually increased to 9.8% from 8.8% at December 31 as a result of our rotating capital from lower-yielding CLOs, CMBS and European RMBS into higher-yielding loan investments.

Turning ahead to Slide 8, you can see that our recourse debt-to-equity ratio, adjusted for unsettled purchases and sales, increased during the quarter to 2:1 from 1.6:1, driven by the growth of the Agency portfolio, while our overall debt-to-equity ratio increased to 3.1x from 2.6x. Our average cost of funds decreased in the first quarter as well to 1.64% as compared to 2.03% in the prior quarter, mainly due to a higher proportion of borrowings on Agency securities. Also during the quarter, we continued making progress extending and improving our sources of financing and leverage. In addition to the non-QM loan securitization, we also added a new loan financing facility and extended the term and improved the economics of 2 others.

For the first quarter, total G&A expenses per share were up $0.02 per share to $0.17, while other investment-related expenses decreased $0.01 quarter-over-quarter to $0.11 per share. We accrued an income tax expense of $2 million for the quarter, primarily due to an increase in deferred tax liabilities related to realized and unrealized gains on investments held in a domestic TRS. Finally, our book value per common share at March 31 was $18.16, up 3.2% from $17.59 per share at year-end.

Now over to Mark.

Thank you, JR. While the vaccine news was great for the economy [and in] Q1 as evidenced by a recently reported GDP growth of 6.4%, the reopening of the economy, which is still gathering steam, was a strong tailwind for EFC and its diversified origination businesses. Across the board, in residential, commercial and consumer strategies, credit performance was good, and we are generally seeing healthier loan growth as the economy reopens.

First, let me discuss housing. In non-QM and residential transition loans, we picked up the pace of origination volumes, and we completed another successful non-QM securitization in the quarter. The technicals for the housing market are phenomenal, and we are seeing some housing statistics that are absolutely eye popping. First, the supply-demand imbalance is quite acute in many regions, resulting in some truly bizarre statistics. The inventory of homes for sale is the lowest it's been in 40 years, but that data series only goes back 40 years. So it might be the lowest in 50 years, and that ignores population growth.

According to Redfin, the average time on the market before a home sells is a mere 25 days. And incredibly, 45% of homebuyers in the past year made an offer on a property site unseen, up from 28% a year ago. And of course, home price appreciation has been off the charts. None of this is indicative of a normal housing market. It may stay this way for a while, but it is certainly at odds with historical norms. And there are clear headwinds. Many of the commodity inputs to a new home are way up in price. Look at lumber, it's tripled in price, plus labor shortages abound. So affordability is going to be challenged and the cost to build a new home has risen significantly.

In our portfolio, we have to resist just looking at last year's housing statistics and extrapolating them into the future because last year's statistics were greatly impacted by COVID. And everybody loves the technicals for housing, so capital has poured into housing-related investments. The price of non-QM loans is up materially from the start of the year. So some of the outsized securitization economics we saw in Q1 are a thing of the past and margins are back to more normal levels. In addition to increasing non-QM volumes, in the last 12 months, we have seen a nice pickup in our residential transition loan volume. These loans are shorter term, typically 1 year, and they're typically made to builders that are acquiring and renovating a home. The expectation is that within a year, they generally complete their work and sell the property.

With the median age of a U.S. home nearing 40 years and large parcels of land either difficult to acquire or difficult to get permitted in much of the country, we believe that you can increase the value of many U.S. homes with renovations that focus on addressing deferred maintenance. And the evolving way in which homes are being used with COVID lockdowns, such as more home offices, makes this need for renovation even more compelling. The challenge here for these operators is that so few homes are for sale.

We have already been in this business for several years. The performance of our loans has been excellent, and we have a very experienced team in RTL lending running the operation. We are pursuing potential equity investments in RTL originators to give us additional control over underwriting and secure for the company a pipeline of new originations, and we expect to close on one of these this quarter. This is a sector that we believe over time will need capital, and we clearly have the expertise. So I look forward to continuing to grow the RTL portfolio from here.

Moving next to commercial real estate. 1 year after the first COVID lockdowns, commercial real estate has performed much better than market projections, our own included. A year ago, we were bullish on residential housing, but had many areas of concern about commercial real estate. Well, while we are still cautious on some commercial sectors, performance has been strong and the pace of activity we are seeing in the commercial space is a good sign. Capital is flowing as evidenced by the increasing amount of new transactions. This quarter, we were able to grow our commercial bridge loan portfolio significantly, which is a great driver of core earnings.

Turning next to consumer lending. Consumers are sitting on a mountain of savings now, which has increased from $1.5 trillion pre-COVID to an estimated $6 trillion now. And while a portion of that increase is obviously from the rising stock market, a lot comes from stimulus checks and lower spending during COVID lockdowns. Also, many have cut down their monthly mortgage payments by refinancing an all-time low in mortgage rates. For us, this dynamic has been a mixed blessing. The good news is that performance in our portfolio has been very good. Last year, we saw many consumer loans enter deferment, and we have seen borrowers leave deferment and continue making their payments. This is how deferment is supposed to work when it's well designed, helping borrowers manage through a temporary loss of income without permanent damage to their lifestyle or their credit history.

On the flip side for us, with consumers less active in 2020, there was less demand for consumer loans. So in addition to managing our current portfolio and our origination partnerships, we have continued to actively look to expand our consumer loan flow arrangements. We believe very strongly that our analytics and data science give us a significant advantage in underwriting many types of borrowers. I'm very happy we were able to grow that portion of the portfolio this quarter as loan growth has started to pick up. The most important story in the Agency MBS space this quarter was the big yield -- was the big increase in yields and a much steeper yield curve. The magnitude of both these changes were similar to the taper tantrum.

The performance of MBS was much better this time as Fed support was consistent and Fed messaging was clear that their support will be with us for a while longer. And when that support is eventually reduced, we expect that the Fed will taper gradually, meaning that they will continue buying, they will just be buying less. We manage the interest rate move and yield curve move by dynamically hedging, but negative convexity and hedging costs were substantial, essentially netting out our positive carry. We had a very slight positive gain for the quarter in this strategy. Agency MBS origination has been strong and our reverse mortgage portfolio company, Longbridge, has continued to grow its volume, market share and profits.

Let's look at how the portfolio evolved during the quarter. As you can see from Slide 6, the credit portfolio might look as though it shrunk quarter-over-quarter, but that's just the result of our non-QM securitization, where we retained a good portion of the economics. We had not done a -- had we not done a securitization, the portfolio would have grown by about $100 million. This was certainly a risk on quarter and credit spreads tightened, but that doesn't mean we were sitting on our hands. We aggressively sold down our CLO portfolio, which has had phenomenal performance since the start of the year. We also rotated out of CMBS into commercial real estate loans. Throughout the wild year of 2020, EFC did a good job of allocating capital to the best opportunities.

Last summer, there was distress in securities prices, but loan origination volumes were still pretty small, so it made sense for EFC to take advantage and deploy capital in securities. While when legacy non-agencies got cheap, we added aggressively, and we held off trimming our CLOs or CMBS positions when prices were distressed. Now CLOs and CMBS have recovered to a large degree, so it makes sense to recycle that capital back into loan opportunities. We also sold some of our nondollar holdings. This is one of the biggest benefits of being part of a larger manager with a broad platform and wide-ranging expertise. We are able to look across sectors, so commercial versus residential; and within sectors, namely loans versus securities, to find the best opportunities.

The reach for yield is really strong right now, and we took advantage of that in this past quarter to sell some securities. And bullishness about housing strategies is very high. The challenge moving forward is staying disciplined and working closely with our origination partners to secure high-quality loans at prices that will allow us to continue to grow core earnings and support our higher dividend.

Now back to Larry.

Thanks, Mark. The year is off to a great start for Ellington Financial. We have continued our strong performance following the COVID liquidity crunch of last spring. And the Board just approved a big dividend increase last month, our third in the past year.

Given the magnitude of the most recent dividend increase, I wanted to share a few thoughts on our dividend. When determining our dividend recommendation levels, besides first making sure that we are fulfilling our REIT distribution requirements, we usually start with core earnings. That's because core earnings helps isolate the most recurring components of our earnings. And so we and many other market participants believe that this metric provides valuable insight into our GAAP earnings potential, and thus, our GAAP earnings to dividend coverage potential. Our core earnings has been growing nicely lately, fueled largely by loan growth. And this quarter, our core earnings covered even our new higher dividend run rate.

That said, our portfolio has also been generating substantial noncore earnings in recent quarters. This has included not only the typical noncore earnings items, such as price appreciation on loans and securities and opportunistic trading gains, but importantly, we've also had significant additional noncore earnings generated by our equity stakes in originators. And these originators are not only experiencing very strong earnings lately, but just as importantly, they are experiencing rapid earnings growth, and we think that growth is sustainable. Longbridge has increased its market share in the reverse mortgage business by over 50% compared to pre-COVID periods, while LendSure recorded its first $100 million-plus origination month in March.

At the same time, both companies are seeing excellent margins, and we think that's sustainable, too. Remember, these 2 companies operate in noncommoditized markets with significant barriers to entry, namely reverse mortgage origination and non-QM origination. The bottom line is that each of these companies is on pace in 2021 for over 50% year-over-year earnings growth. And while none of these tremendous results are technically captured in our core earnings and while operating businesses always present risks, we do see these strategic investments of ours as providing recurring, sustainable earnings for us. And so we view these as helping support earnings, earnings growth, and yes, even dividend growth in the future.

Meanwhile, we are in active negotiations on multiple additional small but strategic equity investments in loan originators, which we hope will not only generate earnings for us directly, but also indirectly thanks to loan flow agreements, which would help us expand our asset base and loan pipeline volume underwritten to our standards. We're hopeful that we'll close on 1 or 2 of these strategic investments later this quarter. We believe that our equity investments represent underappreciated franchise value for Ellington Financial.

As the economy continues to reopen, but market volatility returns, our focus continues to be on our dual mandate of growing the portfolio and earnings while also staying disciplined on risk and liquidity management to preserve book value across market cycles as we did last year. We will continue to seek to grow our high-yielding loan portfolios. We will continue to be opportunistic about where and how to allocate our capital, and we will continue to hold appropriate levels of leverage and liquidity. In this way, we can continue building a powerful and consistent earnings stream for shareholders while also protecting downside.

With that, we'll now open the call to questions. Operator?

(Operator Instructions) The first question you have is from Doug Harter with Credit Suisse.

I was hoping you could just talk a little bit about there are some differential that you're seeing loans versus securities today. And how much of that difference would be kind of on the asset yield versus the financing structures that are available for loans versus securities?

Doug, it's Mark. So it's a good question. I think -- I guess what I would say is that we've seen term financing via securitizations as very attractive, but we've also seen repo rates come down too. So it's not so much the financing differential. I guess part of it is that the price of securities now sort of -- right now, the price of the securities embed pretty good optimism about the health of the economy, which is a lot different than where they were last summer than we added them. So you're not seeing big discounts to par, and you're not seeing a lot of scenarios where we see price appreciation significantly above where securities currently trade. So that's sort of what tilted us back to loans.

Loans have -- while they've done well, they haven't performed as well as securities. And so when we think about total return, which essentially the levered carry plus price appreciation, given the strong price appreciation we've seen in securities and given that many securities now are priced to very optimistic economic projections, which may well be true, we just don't see as much upside in the securities right now.

Your next question is from Bose George with KBW.

This is actually Mike Smyth on for Bose. Quick question on leverage. So it increased to 3.1x from 2.6, and you mentioned you're comfortable taking it up a little bit more. So I was just wondering if you could provide a target leverage range.

The first thing to point out is that overall leverage 2.6 to 3.1 increase has a lot to do with the consolidated non-QM securitizations. So we really focus on the recourse, which also increased 1.6 to 2x. I think we made the point in our prepared remarks that during 2018 and 2019, so kind of pre-COVID years, that recourse leverage was more in the 2.5 to 3x area. So certainly above where it is today. I think it's a function of what investment opportunities are we seeing in financing. And as we continue to grow loan portfolios, which we've continued to do in Q2, that's a good candidate to increase leverage.

And we can do that in a few ways. We can -- there's unutilized financing capacity on some assets. We have very large shares of unencumbered assets, which we disclosed is in the kind of $400-plus million area, so that are unfinanced assets that can be financed. There's also situations where we're not just closing the minimum amount of haircut capital. So yes, I think there's -- we haven't provided guidance on where that leverage will be and could be. It's really a function of what the composition mix is. But suffice to say, we can see it going up from 2x for the right opportunities.

And if I could just add to that, JR. If you turn to Page 24 in the deck, you can see there the 2 charts on the top, I think, are pretty useful for this discussion. The one is, of course, just our capital usage by strategy, credit versus Agency and unemployed. And you can see that we increased the Agency allocation, as we said in the remarks, to about 21%. So that was a substantial increase. And then if you look at the chart to the right, on the top right of this slide, you can see basically the breakdown for each strategy of how much leverage we employ in each strategy. So you can see that our Agency strategy, we increased leverage to 7.2:1. And in the credit strategy, it remained fairly level and now at 2.1:1.

So as we allocate capital, which we do opportunistically between Agency and non-Agency -- sorry, between credit and the Agency, that will obviously affect the overall mix, which, as you can see on this slide, is at 3.1. And I think we've said before that we're comfortable, especially given the -- our ability and our practice of using TBAs to hedge our agencies, which really creates a much lower volatility and lower-risk strategy in agencies, we're comfortable bringing that Agency leverage probably up into the 9s at certain times from 7.2. And of course, the credit strategy is going to be very asset-specific in terms of how much leverage we employ there. But I think 2:1 is a very comfortable level, and we have room to increase that as well.

Your next question is from Crispin Love with Piper Sandler.

Larry, you hit on my first question in some of your final comments, but I'm just looking for a little more detail or clarification. So the new dividend is $0.42 per quarter. And in the release, you talked about even having potential increases to that going forward. So how should we be thinking about the core earnings power of the company in the near term? Do you expect to continue to be covering the dividend with core earnings following that 40% bump to the dividend? Or could the core earnings probably be even higher than that $0.42 run rate?

Right. So yes, we do expect to be able to continue to cover that dividend with core. As I mentioned, there are other components of our earnings that we think are core-like if not core. I'll mention, for example, that Longbridge, most of the tangible net worth in Longbridge is in the form of MSRs. Now if those MSRs were directly on our balance sheet, right, that would flow right through the core, the -- at least the sort of the core yield, if you will, on those MSRs. But since they're trapped, if you will, in Longbridge, so just by virtue of our structure, if you will, those earnings don't flow through through our core but rather flow through -- they appreciate the book value of Longbridge. As Longbridge earns -- generates earnings through MSRs and origination, and of course, as Longbridge's book value grows through earnings and otherwise, that's going to increase value for us.

So -- but in any case, yes, so I think we absolutely see the new dividend as being covered by core going forward. Otherwise, I don't think we would have raised it to that level. But going forward, I think that it's -- I think we'll be in a position to raise the dividend, not necessarily if we're covering it at that moment, but just as long as we see the visibility and have the portfolio and the -- especially with the -- some of these pipelines we have in flow agreements. If we can see the core in the near term as covering the dividend, then I think that would be another reason to be able to cover it. So as we mentioned on the call, we're working on some other strategic equity investments. Should those be consummated and should we get flow agreements to where we have good visibility in terms of what the flow will be from those, that could be another reason to raise the dividend.

Okay. That's all really helpful. And then just one more from me. So with the recent news of the QM patch potentially expiring in July and then some complexity in the revised final rule, how would you expect that to impact EFC and non-QM originations coming out of LendSure? And just, I guess, overall -- just your overall thoughts on that non-QM market currently.

Sure. It's Mark. I can take that. It's -- that's a good question. So what's interesting is that the shift we've seen from the GSEs this year has been to pull back a little bit and let some market share go to the private label market. And where you've seen that most significantly in the last 2 months has been in loans that Fannie and Freddie had originated for investors, right? So it's a property that someone owns in their name and they rent out, right? And so what Fannie and Freddie recently did was they limited that to 7% of an originator's deliveries, right? And there are some originators that were significantly above that 7% limit. And so what you're seeing now is a bit of a transition for some of that volume now going either a securitization route via private label or just winding up in loan portfolios.

So the language of the non-QM patch, as you mentioned, has changed. And we're sort of waiting to see what the final -- what the real final details are. But I guess the overall trend we're seeing is a growing private label market. And that backdrop, I think, is beneficial for Ellington Financial because I think there's going to be a greater percentage of loans that are originated that are going to not go through Fannie, Freddie guarantee and are either going to go into loan portfolios or through private label securitizations.

Your next question is from Mikhail Goberman with JMP Securities.

I just wanted to jump on and quickly ask if you guys are maybe seeing more opportunities to make strategic investments in any originator partners?

Mikhail, it's Larry. Yes, we are. We're focusing on, as we did with the -- we have a few strategic investments, as you know, and those all started out small. And that's kind of our -- been our preference is to help grow something more organically with the capital to buy the product that's originated and the capital [I think to finish] to buy the product that's originated, but to not buy into, let's just say, an operation that's huge, that's going to just really put us at maybe more risk from an operational standpoint. So we're looking at a number of smaller investments right now, in many cases, $5 million and under, but in a variety of originators that we think either have some good capabilities today to provide flow or with the help of our capital, can step up and expand their presence in a variety of markets.

Where -- we have one that we're looking at in the RTL space, which I think we mentioned. We have one that we're looking at in actually the conforming space. We have one that we're looking at in the non-QM space. So there's a whole bunch of ones that we're sort of actively pursuing. And I think, as I said, I wouldn't be surprised to see 1 or 2 of those close this very quarter.

Your next question is from Eric Hagen with BTIG.

I've got a couple. First, on non-QM, I think the prepayment speeds have been pretty elevated there across the market. Some originators, I think, are starting to include prepayment penalties on those loans. I imagine most of what you guys are acquiring is at a premium. So can you share how you structure around prepayment risk as you continue building out that portfolio? And then the second one is, can you just talk about the overall approach to credit hedging right now, just given the way you see the world and the capital markets? And then can you remind us which corporate credit indices you're are long and which ones you're short?

Sure, Eric. This is Mark. I can talk about the non-QM speeds. So yes, you have seen the note rates offered to non-QM borrowers come down in the last 6 months. When the non-QM market sort of started to reopen last May or June, initially, the difference in note rates between non-QM and the Agency loans was wide. And that has certainly come in. And as the rates are lower and also as you have borrowers sitting on a lot of equity, you've seen faster speeds. But you're seeing faster speeds relative to sort of the slow speeds we saw in 2020. I mean, ever since its inception, you've seen non-QM speeds pretty fast, right? And there are certain things we know about non-QM speeds as a function of loan attributes that we think about that you tend to see self-employed borrowers a little bit slower than borrowers that are W-2 employees.

And so in terms of penalties, you can put them on the investor loans. It's not a huge part of what we do. And if you look at the premiums in the non-QM market, there's still a big discount to where a lot of the pools in the Agency space trade. So the way we approach is that there are certain things we can't control, right? We can't control HPA. We can't control how aggressive originators are going to be on non-QM rates. So it's a risk that we manage through and we hedge through, and it's something that we've dealt with really since 2015. And it will be with us, I think, as long as we're in the origination space.

Because essentially, originators are selling loans at a premium because they're advancing par to the [bar] with the closing table, and then they have fixed expenses. So it ebbs and flows. I think now you've seen this little bit of a backup in Agency origination rates, so maybe that will feed back into non-QM rates. You haven't seen that yet, but it's something we watch. I think we have really good analytic tools to stay on top of it, but it's sort of a risk inherent in that business.

And Mark, I'll take the part about the corporate credit hedges.

Sure.

Yes. So -- and the other hedges. So if you turn to Slide 17, we show each quarter what our credit hedging portfolio looks like. It's fairly light right now -- or as of quarter end, I should say. The 2 -- I mean you can see there are 3 bars that have any significant height to them. The European sovereign debt is -- that's really almost exclusively a currency hedge. So I wouldn't focus on that. We do have some European RMBS. That's very straightforward. It's really just to hedge the currency risk. The CMBX is certainly something that we use a lot against our CMBS portfolio, specifically. And that market is a terrific market to be a trader and an active trader, which we are. Spreads can move around a lot. Deals, new deals, new issues can get hung up. And there's often a pretty good disconnect between the CMBX that can use the hedge and the cash markets. So -- and of course, our outlook can change there, too, in terms of how much we want to be hedged as a result. So that's the CMBX portion.

Now the corporate portion of the portfolio we have pretty much used almost exclusively against our CLOS, which is where our corporate credit is. And as we've talked about, that portfolio is shrinking in favor of more of the mortgage and consumer loan portfolios that we have flow agreements on [and] otherwise. So -- and I would expect it to continue to shrink. Now of course, we're opportunistic, so if any great opportunities arise there, then you could see that increase. And then you would see the CDX that we use, the corporate hedges we use, to increase as well. You can see that as of March 31, our -- you can see the height of that bar is $20 million in terms of our hedge. We measure that because we use different instruments. We can use ETFs, we can use CDX. But here, we sort of equate things to what we calculate as a CDX equivalent. So our portfolio is the equivalent of just a bit over $20 million of long protection on high-yield CDX. So that's a relatively small hedge right now.

We view a lot of the risk that we have on the credit side of the portfolio as idiosyncratic, as you can imagine, not related really to corporate credit directly. On the resi side, being in first mortgages, there would have to be a significant downturn in housing prices really for us to take a hit there. And same -- even on the commercial mortgage side as well, where it's even more idiosyncratic our loans. So we don't view really having a corporate credit hedge as anything more than potentially an out of the money put, like if that's something that from time to time, we've considered doing just to kind of hedge your tail risk. But we don't have currently any kind of direct or in-the-money hedges in corporate credit against our portfolio, other than the CLO portion would be the -- really the substantial thing that we'd be hedging there.

That was our final question for today. Thank you for your participation in Ellington Financial First Quarter 2021 Earnings Conference Call. You may disconnect your line at this time, and have a wonderful day.