Two Harbors Investment Corp. Q1 FY2020 Earnings Call

Good day and welcome to the Two Harbors Investment Corp First Quarter 2020 Financial Results Conference Call. Today's conference is being recorded. At this time, I would like to turn the conference over to Ms. Margaret Karr. Please go ahead.

Thank you and good morning everyone. Thank you for joining our call to discuss Two Harbors' first quarter 2020 financial results. With me on the call this morning are Tom Siering, our President and CEO; Mary Riskey, our CFO; and Matt Koeppen and Bill Greenberg, our Co-CIOs. On our call today given potential conflicts of interest, any comments about internalization will be delivered by Mary, Bill and Matt. The press release and financial tables associated with today's call were filed yesterday with the SEC. If you do not have a copy, you may find them on our website or on the SEC's website. In our earnings release and slides, we have provided a reconciliation of GAAP to non-GAAP financial measures. We urge you to review this information in conjunction with today's call. I would also like to mention that this call is being webcast and may be accessed in the Investor Relations section of our website. I would like to remind you that remarks made by management during this conference call and the supporting slides may include forward-looking statements. Forward-looking statements are based on the current beliefs and expectations of management and actual results may be materially different because of a variety of risks and other factors. We caution investors not to rely unduly on forward-looking statements except as may be required by law. Two Harbors does not update forward-looking statements and expressly disclaims any obligation to do so. I will now turn the call over to Tom.

Thank you, Maggie, and good morning everyone. We hope that you had a chance to review our earnings press release and presentation that we issued last night. Coming off the yields of a very strong period, the company, we could not have imagined the societal economic conditions that would result from the global coronavirus pandemic. This has been a challenging time for our nation and the world; our thoughts are with those most affected by COVID-19, especially those who lost loved ones. Additionally, we want to express our gratitude and admiration for the heroes who are on the front lines. Like everyone else, we are hoping for a return to normalcy. First and foremost, we have been concerned about our most valuable asset—our people. We have implemented mandatory work-from-home measures across our three offices and are utilizing technology to stay well connected. We are all adapting to the new reality, and we were able to continue our operations with very minimal disruption. The first quarter was not spared from the volatility of the markets. Unlike the 2018 financial crisis, the speed with which the market dislocation appeared and the large spreads that were shocking on a daily basis were unprecedented. Every asset class was impacted. Spreads, especially on liquid assets, widened dramatically and margin calls intensified. I am proud of how our team managed the volatility by actively derisking risk in our portfolio. In order to establish and maintain a very strong defense of liquidity position, we satisfied all margin calls and did not enter into any forbearance arrangements with our lenders. In the midst of the volatility, we also made the difficult decision to suspend our first quarter preferred stock dividends to ensure we maintain sufficient excess liquidity and preserve stockholder value for the long term. Decisive actions allowed us to weather the extreme volatility in March and end the quarter with $1.2 billion in unrestricted cash. Given the derisking of our portfolio and increased confidence in our liquidity position, we announced an interim common stock dividend of $0.05 per share, as well as full payment of our first quarter preferred stock dividends. While we are not able to provide guidance on future dividends, together with the Board, we will be evaluating our quarterly dividends going forward based upon the composition of our portfolio and the decisions made. Additionally, post-quarter end, we announced that the independent Board of Directors elected not to review the Company's management agreement with PRCM Advisers. Mary will comment on this in more detail shortly. We believe that how we are regarded once this crisis passes will depend upon our actions now. We strive to be best-in-class, and in this time of great uncertainty. We are committed to all of our stakeholders: our employees, stockholders, business partners, and our communities. We are weathering this time together. While we can't predict how this global pandemic will unfold and what lasting effects it will have, we are making every effort to best position our company against factors outside of our control. Despite all the uncertainty, we believe that we can withstand future volatility and ultimately, on the other side of this crisis, drive long-term stockholder value. With that I will now turn the call over to Mary to review our financial results.

Thank you, Tom. Turning to Slide 4, let's review our financial results for the first quarter. Our book value at March 31 was $6.96 per share, representing a decline of 52% from $14.54 at December 31. On the right-hand side of this slide, we have provided a more detailed attribution of our book value decline in the quarter. As expected, the sale of substantially all of our non-agency securities was the largest contributor, resulting in over 70% of the decline. Our rate strategy also contributed to the decline, driven by fair value markdowns in MSR, specified pool underperformance, as well as specified pool sales as a result of delivering our portfolio in March. Matt and Bill will detail this portfolio activity shortly. Moving to Slide 5, let's discuss our core earnings results. Core earnings were $0.25 per share in the first quarter, consistent with the previous quarter. Core earnings were favorably driven by higher net interest income due to the purchase of higher coupon agency RMBS early in the quarter and lower amortization. This was offset by increased interest spread cost and swap positions due to the LIBOR reset and lower TBA roll income. Turning to Slide 6, our portfolio yield in the quarter was 3.52% and our net yield decreased to 1.13% from 1.19%. Aside from improvements in repo costs, this was more than offset by higher swap costs as 3-month LIBOR rates declined in the quarter. As bond rates come down post-quarter-end, we expect to continue to see improvement in repo costs. On Slide 7, we have summarized our financing profile as of March 31. As the turmoil in the market unfolded in March, our financing and investment teams worked closely together to preserve liquidity and deliver the portfolio. As a result, we made all of our margin calls, our liquidity position remains strong, and we are confident in our ability to continue to meet margin calls and servicing advance requirements on our MSR. Our economic debt to equity at quarter-end was 7.0 times compared to 7.5 times at December 31, and our quarterly average economic debt to equity was 7.4 times. It is important to note that we sold our lower levered non-agencies during the quarter. Thus, while the change was nominal in our quarter-over-period debt to equity leverage on agency RMBS and MSR decreased meaningfully. We did not experience any issues with access to the agency repo market and were active in rolling every repo position. As of March 31, we had 22 active agency repo counterparties with a weighted average maturity of 52 days. In aggregate, we have been able to roll our agency RMBS positions without much change in haircut. As has been true in many markets, repo levels were stressed in March. However, as we progressed past quarter-end, we have seen more clarity and consistency in rates, and the term markets are developing again. Repo levels for one month to three-month terms are indicated today to be in the OAS plus 25 to 35 basis point range. As of March 31, we had $252.1 million outstanding across our MSR bilateral facilities and $400 million outstanding in MSR terminal. Our total committed capacity across our MSR financing alternatives is $450 million. Importantly, we are in advanced discussions with two major banks regarding servicing advance facilities. This will provide us with additional liquidity to continue to make servicing advances in the event of increased forbearance or default. For more information on our financing profile, please see Appendix Slide 28. Please turn to Slide 8. On April 13, we announced our election to not renew the management agreement with PRCM Advisers. As a result, the agreement will terminate on September 19, 2020, and we will become a self-managed company. The decision was the result of a diligent, thorough, and extensive month-long process led by the independent directors of our board, with the advice of independent legal and financial advisers. The board-driven process commenced long before the COVID-19 pandemic and the timing of the announcement was predicated in part on the non-renewal provisions of the management agreement. We believe that our stockholders will benefit from significant annual cost savings, as well as from further alignment of interests between management and stockholders, enhanced returns, and the potential for attracting new institutional investors. Under the terms of the agreement, we are required to pay a one-time cash termination fee, which we estimate will be approximately $144 million. We expect to realize an annual cost savings of approximately $42 million, or $0.15 per share, before giving effect to any additional cost savings from the elimination of the management fee on future capital growth. Therefore, the payback period on the termination payment is relatively short at just over three years, and the return on investment is approximately 29% per annum without accounting for future capital growth. The manner in which the termination payment is calculated will use a 24-month look-back period, especially since the lower management fees paid in earlier quarters are taken into consideration. The payment of the estimated termination fee will result in the most favorable economic benefit to stockholders. In conclusion, we are confident that this is the right time to make this change given the maturity of our business model, our well-established infrastructure, and the material economic benefits for stockholders, and we believe strongly that the transition to being self-managed is a very positive step for the future of our company. With that, I will now turn the call over to Bill and Matt for our market overview and portfolio update.

Thank you, Mary, and good morning everyone. We'd like to start this morning by spending a few minutes discussing the effects of the COVID-19 pandemic on the residential mortgage market. Then we will discuss our portfolio activity for the quarter. Finally, we will talk about our outlook for servicing advances in our portfolio liquidity. Please turn to Slide 9. In March, the COVID-19 pandemic had a swift and dramatic effect on valuations, leading to extraordinary spreads widening across virtually all asset classes. The left-hand chart shows the 10-year swap rate and the S&P 500 in the first quarter. Although the tenure swap rate trended lower early in the quarter, it moved sharply in March, finishing the quarter roughly 120 basis points lower. The S&P 500 performance really emphasizes how quickly events unfolded, as the all-time high on the index was realized on February 19th before falling 35% just a month later. The right-hand chart shows the spread widening impacts across various fixed-income asset classes from pre-crisis levels in February, indicated by the low to the high point of the vertical line, and then ending mid-April indicated by the gray circles. No asset classes escaped this volatility, and this affected us and other mortgage REITs significantly. Most spreads remain significantly wider than they were in February, with the exception of agency RMBS, which have retraced all of their widening due to the direct impact of QE4 purchases by the Federal Reserve. Moving to Slide 10, the left-hand chart compares RMBS Treasury spread index in 2008 versus March 2020. While the magnitude of the widening is notable and similar in both cases, the real difference was the speed at which it unfolded. The round-trip was completed in three weeks in March compared to six months in 2018. The right-hand chart shows the large spread changes in the Fannie coupon during the most volatile week in March. As you can see in this chart, each day is delineated by vertical lines and starts at zero. It was not uncommon during that week to see 50 basis point moves, both widening and tightening, sometimes even in the same day. This high level of volatility was further exacerbated by extremely low levels of liquidity in the MBS market. At times, the bid-offer spread in the most actively traded securities were as much as 100 times normal levels. At the end of April, the agency market has substantially healed and bid-offer spreads were largely back to normal. Please turn to Slide 11. As the stress on the RMBS market grew, the Federal Reserve took a number of actions to stabilize the markets. At the outset, the Fed cut interest rates by 50 basis points on March 3rd in an attempt to offset increasing liquidity fears. The market was not soothed, and by March 15th, it was clear that the system needed additional action. The Fed responded by cutting rates to zero and announcing its commitment to purchase $500 billion in treasuries and $200 billion in RMBS. The market was still underwhelmed, and volatility in the treasury and RMBS markets actually worsened. In response, on March 23rd, the Fed committed to unlimited purchases of both, in what has become known as QE4. Following the recommendations of various market participants, the Fed also began to buy bonds on a short settle basis, waiting until regular mid-month settlements. Fed RMBS purchases were also expanded to include higher coupons and not just the lower dollar price production coupons, which was very helpful for many market participants, including mortgage REITs, who generally had large holdings in higher coupon bonds. The left-hand chart shows deal Fed purchases of MBS, which have now exceeded $500 billion. The right-hand chart shows the pace of QE4 compared to QE1 in 2008 and QE3 in 2012, demonstrating again the speed and size at which events are unfolding today. With that, I will hand it over to Matt.

Thank you, Bill, and good morning everyone. Please turn to Slide 12. With the foregoing discussion as a backdrop, let's now discuss the actions we took in March in response to the liquidity crisis caused by the pandemic. As volatility increased and spreads widened on agency RMBS, we reduced risk to raise excess cash. We liquidated around $18 billion specified pools and TBA representing about 50% of our agency portfolio. When liquidity was poor in March, we initially sold lower coupons where there were some sponsorship, but eventually reduced exposure to higher coupons once the Fed adjusted its programs with a focus on purchases in that part of the coupon stack. Timing of most of these portfolio sales, occurring as they did before unlimited QE was announced, was such that we did not fully benefit from the spread tightening driven by outside spread purchases. Regarding our legacy non-agency portfolio, we became increasingly concerned about levered portfolio liquidations occurring across the market, with an acceleration in size and frequency of margin calls arising from widening spreads, increasing haircuts, and uncertainties around the ongoing funding in the repo market. As a result, we decided to liquidate substantially all of our non-agency portfolios to eliminate these risks. The chart on the bottom of the slide shows our portfolio composition on December 31, where you can see the effect of the portfolio sales we just described. Please turn to Slide 13. Another large driver of performance during the quarter was the collapse in specified pool pay-ups. The chart on the upper right-hand side of this slide shows the 3.5 coupon higher loan balance specified pool pay-up levels. This chart is indicative of the price action that impacted all of our assets as the month unfolded. With the risk appetite and balance sheet capacity low, specified pools pay-ups—indicated by the blue bars and measured on the left axis—dropped significantly, while the ratio of the pay-up to one measure of its theoretical value, shown by the gray line and measured on the right axis, also indicated a drastic decline. You can see that the premiums fell from around 4.0 to around 1.0 at the end of the quarter. There were even trades that took place in the market at negative pay-ups, indicating that the specified pools traded below settlement date-adjusted and cost of funds-adjusted PPA levels. Specified pools have significantly improved in April. The lower charts show our specified pool breakdown on December 31 and on March 31. You can see that we liquidated essentially all of our lower pay-up stories during March to minimize losses while delevering. Specifically, we sold a net of $13.4 billion in pools during the quarter across 3% to 4.5% coupons, predominantly in high LTV pools. Moving to Slide 14, the graph on the right-hand side of this slide shows coupon performance for the first quarter. Despite the massive volatility, agency RMBS ultimately performed well after the Fed stepped in with support. Coupons in the middle of the stack, including 3%, 3.5%, and 4%, all outperformed by a point or more. Our effective coupon positioning is shown in the bottom chart. Our implied short positioning from the MSR asset moved to the 2% coupon in March from 2.5% and 3% at the end of December as a result of falling interest rates. Our long holdings include 2.5% to 5%. Please turn to Slide 15. This slide shows our interest rate and mortgage spread exposure. Although we recognize that these representations of our exposures are too simplistic to accurately depict portfolio performance during the crisis, as volatility subsides and market movements return to normal, these kinds of risk measures regain their usefulness. These exposures remain low, consistent with our historical positioning. In the top-right chart, you can see our exposure to instantaneous changes in mortgage spreads. As of March 31, a 25 basis points spread widening would decrease book value by 1.4%. The chart on the bottom of the page shows our book value exposure to instantaneous parallel changes in all interest rates, which shows that as of March 31, an instantaneous parallel shift in interest rates upward of 50 basis points would negatively impact book value by only 1.3%.

Please turn to Slide 16. We would like to dedicate time today to one of the biggest challenges we are currently facing: increasing mortgage loan forbearances and servicing advances. While the COVID-19 pandemic began as a liquidity crisis, as these things often do, it has transformed into a credit crisis. Shutting down large parts of the United States has had far-reaching economic impacts. Unemployment has skyrocketed and is expected to continue to climb. With so much of the economy shuttered and people out of work and staying home, the ability of all types of borrowers, renters, and homeowners to pay their obligations is being called into question. At the end of March, Congress enacted the CARES Act, which, besides including relief checks and other support measures, has important impacts directly on our industry. One provision in the Act provides for up to 180 days of forbearance relief from mortgage loan payments, with the right to expand for an additional 180 days for borrowers with federally-backed mortgages who experienced financial hardship related to the pandemic. The Act also prohibits foreclosures for 60 days. Through one of our subsidiaries, we own the MSR for over 8 loans guaranteed by Fannie Mae and Freddie Mac, and as a result, we are responsible for making advances for certain payments in the event that they are not made by a borrower. In normal times, this is not burdensome; pre-COVID, our 60-day delinquency rate was about 30 basis points, and we were able to manage the liquidity needs related to this in the ordinary course of our business. However, with the forbearance programs now in place as a result of the CARES Act, the situation is different. During the forbearance period, we are responsible for remitting monthly scheduled principal and interest for loans backed by Fannie Mae and the monthly scheduled interest for loans backed by Freddie Mac. Additionally, we are responsible for making interim taxes and insurance payments to local authorities and insurance companies. On this slide, we show our MSR portfolio forbearance rates as we started collecting data on March 23. This data is shown by the blue bars at the bottom of the chart and references the left axis. After April 28, our experience is that 5.7% of our loans by count have entered forbearance. The gray line shows the daily percentage changes in the number of loans in forbearance and is measured by the right axis. This number has been steady in the low single digits for some time. It's worth pointing out that both the Mortgage Bankers Association and Black Knight, a loan servicing technology company, publish data regularly regarding forbearance rates for GSE loans, and our data has so far been very consistent with both of those sources.

Please turn the page to Slide 17. On this slide, we consider how our servicing obligations relate to our liquidity position. We have taken a scenario analysis approach and created 50 scenarios, including a moderate stress case and a severe stress case. Depending on the speed of the relaxation of social distancing measures and the reopening of the economy, we believe that a 15% forbearance rate is a reasonable estimate for a base case. Of course, it is impossible to predict the ultimate take-up rate with any certainty, but as you will see, we believe our liquidity position is strong enough to withstand stresses to this outcome. Scenarios differ primarily by varying the maximum forbearance take-up rates; the base case assumes 15%, the moderate stress case assumes 20%, and the severe stress case assumes 25%. Additionally, as we move across these three scenarios, we vary pre-payments from 25 CPR in the base case to 20 CPR in the moderate stress case and 15 CPR in the severe stress case. The pre-payment rate is an important driver forecasting advancing obligations. If custodial accounts for principal and interest can be used to offset obligations, these funds must be repaid the next month. The next variable is the existence of a servicing advance facility. We are in the process of negotiating documents for such facilities with large Wall Street banks. We expect that any facility would accommodate both Fannie and Freddie's advances. Most subjects are customary closing conditions and GSE approvals, and we have included the existence of such a facility in our scenarios. The last variable is the valuation of our existing MSR portfolio. We have two outstanding facilities where we borrow against our MSR assets. Our distinct MSR financing facilities are meant to finance advances, and the borrowing base depends on the market value of MSR. Therefore, in each scenario, we stress the market price to simulate potential market calls. This price distress starts from our March 31 valuations and declines from a 3.0 multiple down to a 2.0 multiple in the severe stress scenario. All three scenarios explicitly include the payment of the estimated non-renewal fee due in September. The chart at the bottom left of this slide shows our projection for our advancing obligations. In all three scenarios, you can see that the servicing obligation peaks around December 2020. In all of this, the curves rise quickly due to the accumulation of P&I and P&I advances and then decline as we are able to recoup P&I advances from interim claims processing. Even in the base case, the maximum advancing obligation is around $100 million, while in the moderate stress and severe stress scenarios, the obligation increases to $250 million and $450 million, respectively. To determine whether these are significant numbers or small ones, we need to overlay those with our liquidity. The chart on the bottom right of the slide lays out a path of our liquidity over the next 18 months. Starting with a steady-state cash balance of around $1.2 billion, the reason we are holding such a large cash balance today is precisely in consideration of this advancing obligation. As you can see, we expect our excess cash balances to decline as we pass through the advanced stage after which they will return to a more normal level. In all three cases, we maintain an excess liquidity buffer of between $400 million and $600 million. For a portfolio of our size, with the big tolerances we expect to be running, we believe full transparency is essential regarding what we're seeing around servicing advances and how we are thinking about it in the context of our MSR portfolio. We hope this explanation helps summarize our outlook based on the information we have been discussing today; we feel quite confident about our liquidity position. Please turn to Slide 18. The reality is that this pandemic has reordered everything, and we have a lot of work to do before we are in a position to return to business as usual. The advancing obligations are significant, and while we feel confident in our ability to manage them, it will take time for events to unfold before we anticipate returning to a steady-state business model. There are multiple obstacles that will certainly drag down earnings in the near term. These include higher than normal cash balances, increased servicing costs from our sub-servicers relating to the loans, costs to set up and maintain the servicing advance facilities, and uncertainty in MSR pricing. Having said that, we believe that the opportunity set in our target assets is very attractive. We discussed earlier, with such support, agency TBAs recovered all of the March widenings and then some; thus, returns in that part of the market are similar to what they were pre-crisis. Specified pool pay-ups have recovered much of what they lost in March but are still somewhat lower than they were when the crisis began. As a result, we are seeing returns on rate-hedged agency RMBS in the low to mid-teens. In the MSR asset space, we see the most interesting opportunities. While it's true that there is still much to work through regarding forbearances, advances, and higher service costs, we still estimate that the forward-looking returns on our existing book of MSR, when paired with MBS, are in the mid-teens based on our Q1 valuation. Earlier we discussed some potential scenarios about the future path of MSR prices. Should MSR multiples decline to 2.5 times on our existing book from three times, we would estimate that the paired return would be in the very high teens. The MSR bulk market remains shut down, so it's not possible at the moment to acquire new assets there. The MSR flow market is slowly coming back to life, and indeed we are buying a trickle of MSR assets in that channel. The multiples in the flow market are usually lower than in the bulk market. While visibility remains somewhat unclear today, we think that flow products can be acquired below two multiples. On a forward basis, we think that translates to yields above 25% when paired with agency RMBS. We don't yet know the amount of assets that can be acquired at those levels, but it is clearly very interesting at those prices. The barriers to entry in the servicing business are many. Every asset is time-consuming and complex. Only market participants who already have the infrastructure, processes, and relationships in place will be able to access the kind of returns just mentioned. Here at Two Harbors, we are uniquely positioned to take advantage of that attractive opportunity on an ongoing basis. With each passing day, the outlook becomes a little clearer. As Tom said at the outset, we aim to be as transparent as possible, and we will continue to update you with new developments over time. With that, I will turn it back to the operator.

We will take our first question today from Doug Harter. Please go ahead.

First, thanks for the additional disclosure regarding the liquidity just Matt, then on your commentary about the attractiveness of new returns, can you just talk about where you are in the flow program? Is that something that you guys are acquiring loans or MSRs through the flow today, or what would it take from a liquidity position to feel comfortable acquiring loans through the flow?

Yes, this is Bill; I'll take that. So as Matt said, we are acquiring what we call a trickle during the depths of the liquidity crisis. Like many of our competitors, we suspended our acquisitions during that period. As I said, that market is slowly coming back to life. We are back in the market on a limited basis, really in a way that does not increase our forbearance and advance risk. There are different ways to view that by targeting certain types of collateral and so forth. We can discuss that later if you'd like, but we are doing it in a way that minimizes that liquidity burden, and we're using the opportunity to see how much of those assets can be acquired at those levels. And obviously, at those levels, it's very interesting. If we're able to do that, we will figure out a way to make that work.

Great, Tom, and then if you could just talk about the capital structure after the book value decline in the quarter. The mix of preferred is obviously much higher now, and how do you think about that mix and what the plan would be to get that back in line with goals you had previously or targets you had mentioned?

Well, thank you, good morning. Obviously, it's something that we're acutely aware of and are reviewing. Yes, you're right that it had an effect on the capital structure, and it's something that we have under review. Some people might be curious as to whether we would raise common equity, and obviously that would depend on the opportunity. But it's something that we're very much aware of, and we will attempt to deal with that the best we can.

Okay. Thank you, Tom.

Our next question today comes from Mark DeVries. Please go ahead. Your line is open.

Yes, thanks. I had a question about the liquidity projection on Slide 17. Is that liquidity number, is that just based on the cash we have on hand now, or does it also assume draws on these facilities during the process of negotiating?

This is Bill; I'll take that. It's a combination of all of that—it's a pretty comprehensive set of assumptions and models that incorporate to the best of our knowledge all of the potential drags and sources that we will have at our disposal. I mean, we think where we can, we tried to be fair but conservative and try to stress the models; as you see, there are three scenarios, so it includes all of those things.

Okay, got it. And just given the high-quality nature of the servicing assets and all the risks around it, are you generally sensing somewhat unlimited capacity to finance that? So God forbid, you had a scenario that was even more stressful than your severe stress; you would just be able to expand the financing available to you?

Yes, I mean we are certainly— and I think you can see this in our liquidity projections. We are sizing our needs based on stresses to the base case, to the severe stress, and even a little bit more than that. You can see that in the liquidity projection because we only went to this stress. But it would still be true it would hold up for something much bigger than that. Now, it is true if we see forbearance rates increasing a lot higher than 25%, which, right, I mean, I don't see; I think we're already starting to see, as staggeringly big as the unemployment statistics are, that we're seeing in the market. We are starting to see them leveling off a bit. So we feel pretty good about the projections. They won't rise above 20% or 25%, but I am obviously, if that does happen, the negotiations that we have with the banks could allow potentially to renegotiate and upsize, but we're sizing it to be larger than 25% at the moment.

Okay, great. And could you give us a sense of how your blended returns might be impacted under the different scenarios if you're forced to reallocate capital due to funding advances and other business scenarios?

Well, it might be a little bit more, but I don't know if I have those numbers at my fingertips. I do know what we said was, including the higher forbearance and increased costs of service and reduced cash flow from loans in forbearance on the servicing asset, that in our base case, the paired levered return is still in the mid-teens. I'd have to check about the other scenarios. I don't have them here; I can get back to you.

Our next question today comes from Bose George. Please go ahead. Your line is now open.

Good morning, hope everyone is staying safe. Actually, first, can you give us an update on book value since quarter-end?

Good morning, Bose. This is—I’ll take that first and then hand off to Bill and Matt for further comments. So firstly, I always have to compromise the caveat that five weeks to model quarter make. But as Bill and Matt alluded to in their script, in April and May, we experienced a notable and favorable repricing of specified pools, which generated a significant amount of P&L. However, there are a couple of things to note, as Bill and Matt discussed. There is some uncertainty with respect to the effects that forbearance and extended servicing advances will have on future MSR remarks, and then additionally we plan to record the internalization fees to Pine River in Q2. So there are a couple of substantial variables; obviously, the forbearance effect.

Yes, I'll take that one. Thanks, Tom, and good morning, Bose. To put a little more color around those moving parts, the specified performance has been strong in April and followed on in May. We've seen a multiple point recovery after the big fall-off in March—basically things back to closer to where we saw them in February. So that component for Q2 is driving an attribution of around 14%. Tom also mentioned servicing, so we always are getting our servicing valuations; they are valued by three independent brokers. However, as time passes, right, and we had this in some of our liquidity projections. As you can imagine, as forbearance flows through and prepayment fees possibly pick up, we might see some pressure on multiples. So if that sort of pricing were to come through, we could see that sort of in the base case, with the 2.5 multiple leading to a book value change of around $150 million or $175 million. That could be around an 8% or 9% impact. And the third thing Matt said, Tom noted the internalization payment, although it is due to be paid in September. We’re going to record it in Q2, so that's about another 8% offset. So I'd like to give you a more straightforward answer. But there are a lot of moving parts here, and it will depend on how those things play out.

Okay. Actually, thanks. That is very helpful. So actually, just leaving aside the charge for the management fee, the spec pool up 14%, the MSR maybe down 8 or 9. So like a 5% up for the quarter-to-date is probably a reasonable ballpark?

For a second, so the MSR mark that we're guessing about that—we don't know. We have end of circle marks that don't reflect that kind of movement at the moment, given what we think is happening with forbearance rates increasing and how we think the world will react. That is a possibility that could happen, but we don't know; we'll have to wait and see.

Okay. Now, that's really helpful. Obviously, that makes sense for in terms of conservatism on that estimate. So yes, thanks very much, and then just going back to that stress scenario slide, you know, that you assume the reimbursement on that, the P&I is after the 15 months. There's been some discussion and chatter about the GSEs might come out with something to reimburse servicers sooner than that. Do you have any thoughts on how that might play out?

I don't, unfortunately. I'm hearing the same thing as you are. There's lots of proposals and possibilities in flux. These projections are made with the best information and assumptions that we have at the time; we update them every day as we learn new things and incorporate new information. So as of the moment, that's what these things include, and should we be able to reimburse it differently or hopefully better, then we'll include those, and we can update you guys on those when they happen.

Okay, great, thanks very much.

Our next question comes from Trevor Cranston. Please go ahead.

Hey, thanks. One more question on the liquidity projections in the forbearance scenarios. First, I just wanted to clarify that, that does include the buyout of the management contract. And then two, I was curious if there's any assumption in that projection around the dividend level changing from what was paid for the first quarter? Thanks.

Hi, Trevor, thanks for the question. So the answer to your first question is yes, it explicitly includes that payment. Also, point out by the way that what's not included here is any liquidity or cash raising from any portfolio adjustments that we could make if necessary. All right, this is all keeping the portfolio cast of what it is; it does not include any dividend payments, but it doesn't, it doesn't in a way, but the idea here is that the portfolio is going to earn what it's going to earn, right, and that's going to be paid out, right. So we're not building into these liquidity projections a growth in the liquidity from not paying the dividend, but it's assumed to be paid out to the dividend.

Okay. Got you.

It sounds good, Trevor.

Yes, thank you for that. And then in areas you were discussing for MSR valuations. I was curious if there's any way for you to sort of provide some context around the rough approximate sort of multiple on the MSR book would be in the scenario with if your stress case in terms of forbearance combined with a significant compression in the primary, secondary, mortgage credit spread?

Look, it's very—thanks for the question. That's a very hard question. It's very hard to know; you're asking about what I think the future prices of something might be as the world unfolds. That said—and we've talked about this a little bit earlier in the quarter—the MSR market is hard to value, especially today because there are not very many trades taking place. All right! And so the brokers are in a situation where they say, 'Well, how do I put numbers on something when I don't see any observable trades in an illiquid and stressed market without willing buyers among sellers?' In the absence of that, the brokers have taken the approach to keep the moment's risk premiums constant; we don't know whether that's true or not. The speed with which the markets have been evolving and moving has been astonishing. As Matt just told you, the specified pool pay-ups went from four points down to one point or below, back up to four points. They are all back to where they were. There will be increased costs from servicing delinquent assets from the forbearances, and there will be reduced cash flows as the non-paying borrowers pay their service fees. So there is what I would call a technical cash flow effect from the servicing cash flows, but depending on the curing scenario that you think might exist, whether it's a deferral option or something—how long that persists for is unknown; maybe it's even short. Maybe after one year, people start paying the mortgages again if some of these deferral options come to fruition, and there's not much of an interruption. It's hard to know. I think we do think generally that the low 2 multiple is very hard to achieve on any long-term basis for these assets with willing buyers and willing sellers, just the nature of the cash flows in the timing and what they're worth and the interest rate and convexity characteristics and so forth. As I said in the previous comment, I think it could go to 2.5 million depending on some of these factors, depending on lots of inputs. I don't see it going much below that below 2. I guess if we had some very high-stress scenario, 25% with no one expecting including us, I guess, it's very hard to know. As things normalize, as trades start to occur and the market starts to heal, we'll get more visibility and we'll be able to speak with more clarity on what we think those scenarios are.

Trevor, I would add that in our liquidity projections, we did include scenarios that do take us down to a 2 multiple even though we think that's maybe severe. We'll have to see what happens, but we do include those numbers in our projection.

Okay, got you, that's helpful. And then on the advance facilities you guys are working on, I know they're not closed yet. But are you able to say anything about kind of what the cost of those facilities are likely to be when they're utilized and assuming that you closed?

Not really; I'd say, we're still working on it. I mean, they're not closed, and we're still negotiating. I mean, they're very market-based, I would say.

Thank you, guys.

Hey guys, thanks for taking my questions this morning, and I hope everybody is doing well. I wanted to touch on servicing advance facilities briefly; historically, and I think Mark pointed this out, servicing advances are AAA-type assets and get very favorable financing. I am curious if pricing remains as attractive as it has historically, and does access to servicing advance borrowing represent another barrier to entry in terms of that business?

Sure, thanks for the question. So servicing advances as an asset class do typically enjoy favorable financing due to their perceived quality. That would be an asset, though I think what you're thinking of is, if we were to create the advancing securities and then sell them to the marketplace, what it would look like—AA, AAA securities for the investors that would buy those notes, right? That's on the end-user side of what we're making here for those who would buy that.

No, actually I misspoke and you’re over-complicating my question— I apologize. I just meant from a collateral perspective, the servicing advance is considered to be AAA collateral, so it gets— that it gets very favorable borrowing costs and very high advance rates.

Yes, that's true because the counterparty risk is generally considered to be manageable compared to the servicer risk. Yes, the advance rates are typically, but I think how you rate market advance rates on advanced facilities depends on P&Is and P&Is corporate advances typically range in the 85% to 95% range, right? Those are just market terms for those kinds of things.

Yes.

And you’re reflective of what you just said.

Yes, and are you seeing any change in pricing given the current market conditions related to those facilities?

Yes, I'd say it's probably a little bit wider than what it was and what it would have been in February, but it's still very low compared to other things in the world at the moment.

Got it, okay. Yes, you made the classic mistake of assuming my question was smarter than it really was. I wanted to, on a very simple level, make sure I understood that. Second thing, look, you’ve exited the non-agency business; we understand why you did that given market conditions. I am curious if you think that that is a permanent exit or that will be a business that you will revisit when markets normalize?

Well, obviously...

I will take it.

Good morning, Rick, it's Tom. Obviously, it’s something that we have deep expertise in, but we'll just have to see how things unfold.

Great thanks, Tom.

Our next question comes from Stephen Laws.

Hi, good morning. Effectively no credit risk now, so I want to focus one question on prepayments into our servicing if I can. First on repayments and refinance activity; really more at the higher level, how do you see the recent environment impacting processing times and repayment speeds? A lot of informing applications—borrowers that would have maybe done a clean rate-driven refi will now have an employment gap, unemployment, possibly other things that would slow that approval pipeline. But how do you guys see that impacting refinance activity? I know the Mortgage Bankers are pretty optimistic that the mortgage treasury spread tightens pretty quickly back to pre-COVID levels, but I would love to get your thoughts. And then how to think about that from a lag on the refi index? Is it going to extend where that sort of 6 weeks becomes 8 or 10 or 12 or more? Comments around that would be great.

Sure, I'll start with that one. Stephen, that’s a great and interesting question. These days, there is definitely a lot of uncertainty and speculation going on out there. I think one interesting thing that we've observed very recently—just yesterday actually—speeds were released for the April period. Broadly, I think they surprised most of the market on the high side. So the April closings would have reflected the very high MBA refi rate and the rate environment back in late February and early March before we really came under any stress. I think the speeds that came through in April showed that there probably were pretty high pull-through rates. They actually increased by about 25%, which I think is what people might have thought they would have increased in typical times given the prevailing interest rates and conditions. So I think that was only one month, and it was a bit surprising, and we’ll have to see. This is unlike anything anyone's ever really observed. So we're monitoring that, and we're obviously going to watch the data as it comes in to see where that goes.

Yes, I'll also add a few words to that. I think there are forces on both sides obviously, rates are at low levels, right? So that, in ordinary times, should create very high refinance volumes. 47 states now permit some sort of remote modernization, right? By our count, I think that we read two-thirds of recording offices allow electronic filings. Verification of employment requirements have been relaxed across the country and so forth. But the world has made adjustments to allow the refinancing machines to continue, and I think this month speeds have been a collection of that, which was a surprise to many people. That said, I think various Wall Street analysts are projecting speeds to slow down more than 15% or maybe even more next month. We are seeing some signs in the refinance index and so forth, but I think the speeds will probably remain more elevated than people might think, given a naive thought about what the pandemic is doing and the social distancing measures.

Great, appreciate the comments. And I guess before asking to predict the future again, I'll ask about the servicing portfolio. And just how the advance obligations are structured, but can you provide any kind of mix of your obligations? How much of it is scheduled versus actual for both interest and principal? I know it's different across Fannie and Freddie and certainly Ginnie; I believe is scheduled, but I don't think you guys have any exposure there. Can you give us any breakdown there and maybe quantify the difference? How much is that in average payment principal if its actual principal advances instead of scheduled—how much is that a savings on the total advance obligation to you guys, 5% or 15% or what is that number on the delinquent payments?

Okay, so the short answer to your last question is I do not know; I don't have those numbers handy. On the breakdown of the portfolio, 20% of our portfolio is Fannie Mae actual/actual; I want to say around 30% is Freddie Mac, and therefore scheduled/actual, right? And the balance is Fannie Mae scheduled/scheduled. Even for the actual/actual, which has no principal and interest advancing obligation, we do have D&I advancing obligations.

Yes.

In most cases, the way the model works in the world, is that the bigger part of your obligations anyway.

Okay. But still, if I heard correctly 20% is scheduled and is actual/actual, which means you're not advancing any scheduled and unreceived interest for principal in that 20%?

That's correct.

Great, I appreciate you clarifying that. Back to predicting the future—events, outlook is certainly concerned. People are talking about one aspect of the 120-day rule; if you get one good payment and they go into forbearance, again the clock resets. So I don't want to ask you to run a scenario for every possible situation, but in that event where you may be forced to cover seven of the next ten mortgage payments, how does that look from a liquidity standpoint? I would imagine the government would have to step in in that type of situation, because it's certainly not a Two Harbors-specific problem; it would be everyone. But do you have any thoughts around that, with the clock resetting if you get one payment in?

Not very deep ones. I would say that the advanced facilities are built to cover that, right? So I think these things would be scalable in order to accommodate that. But the size is large enough. As you know, especially with the P&I advancing, the principal and interest custodial accounts can be used to offset that. And so, look, while low rates and fast prepayments for premium mortgage portfolios are generally not great, in this instance, it would go a long way towards covering those sorts of obligations. As I said, the P&I part of the advancing is generally—we look at it as being easier to accommodate because of that. And so I think all of those thoughts would be true if that were to take place.

I would add too, like we said earlier, if a scenario like that were to happen, we would obviously have time. That allows us to be reactive and work on either additional facilities or upsizing facilities, which I think we would be able to do if we needed to in that scenario.

That's a great point because it's not a margin call one-day issue; it's something you'll see playing out over months, and you'll see that as you call your delinquency numbers. So I appreciate the comments today. And thank you very much.

Thank you.

Thanks, Stephen.

Our next question comes from Kenneth Lee. Please go ahead.

Hi, thanks for taking my question. Just wondering on a broader level, how would you characterize your current appetite for making investments in the near term, just given the potentially attractive opportunities you see weighed against the uncertainty in the market environment? Thanks.

I’m sorry, could you repeat that question? I don't think I caught it.

Yes, certainly. Just on a broader level, just wondering, how would you characterize your current appetite for making investments in the near term, given the potentially attractive investment opportunities you’re seeing weighed against the uncertainty you're seeing around environments? And relatedly, how leverage in the portfolio could evolve over the near term?

Yes, I'll start. This is Matt. I mean, in one word, we’re quite cautious here. I think we have to get a little bit of time passage and a little bit more visibility into what our advancing obligations are going to be and see how forbearance unfolds and sort of look at its impact on all mortgage assets. We're looking forward to doing that. I think we're not quite there. We do need some passage of time. But like I said earlier, there are indications that there are interesting opportunities out there. Once we feel like we're in a comfortable position to take advantage of things, but we're still pretty cautious today.

And I would add one thing that I think is sort of self-evident in the way, like one signal for that would be when the forbearance upgrades start turning over like seen in other contexts; once you see that, then you can sort of project really what the future is going to look like and so forth. And so I think that's one of the main things that we're looking for.

Okay, very helpful. Thanks again, and hope everyone stays safe.

Likewise, thank you very much.

Likewise, yes.

Our final question comes from Matthew Howlett. Please go ahead.

Hey guys, thanks for taking my questions. Just a few quick ones. First, how are you monitoring counterparty risk with the non-bank sub-servicers or the guys you buy from them? Just curious on what's it like out there?

I'm sorry, from people who we're buying from, or?

Well, kind of both. I mean mainly sub-servicers or some of the capital rules coming out, what could come out of liquidity also issues? How does that sort of—not being monitored—I guess the question is, if you are there, could there be a need to take servicing at some point?

Right, so the answer is no, I mean, we have three sub-servicers, as you know, and we've been public about who they are: it's Flagstar Bank, the Bank; there’s Bill Renewals, which is a pure sub-service; they don't own any servicing; and they don’t have particularly any capital stress here. We have ongoing diligence reviews with them, and we're checking with them about these sorts of things regularly and periodically. So we don't have concerns at the moment about that.

Okay. And so there's no need to move back up servicing or anything of that nature if you have any one of them run into liquidity issues?

We do not have a hot backup in place; no, we don't have that. I mean, one of the advantages, I think, of our model is the diversified nature of the thing that we have, it's spread out among different sub-servicers. As you know, our portfolio is roughly split 40:30:30. To the extent that we would be able to potentially see through our regular reviews any stress, we could, of course, initiate the process to move servicing from one to the other; we do that not regularly, but we do it. We're very experienced in that, and it could be accommodated in some number of some amount of time, obviously requiring coordination and approval from the GSEs. So we could do that, but we do not have a hot backup in place.

Yes, okay. Thanks, and then just from a modeling question that sort of net interest spread, did you guys get that 1.13%? I mean, there's been a lot of moving parts obviously with the sale of the non-agency book, repo costs coming down. How do we think about trending that? How should we think about modeling that going forward?

So I can take that question; this is Mary. So I think we expect on the asset side for the yields to be in the low to mid-threes in the near term. I would note that as we de-lever the portfolio, we correspondingly needed to reduce our net swap book at a time when three-month LIBOR was extremely elevated and stressed while long-term rates were low. We do expect that this will impact us in Q2, but as swaps and repos reset, we expect the net yield to return to more recently observed levels.

Got it, and then there will be an impact from lower yields because the non-agency, higher yielding non-agency book is out?

Yes, that will have a slight impact.

Got it. Great, thank you.

Ladies and gentlemen, that concludes today's question-and-answer session. I would like to turn the call back over to Margaret Karr for any additional or closing remarks.

Thank you, Claire. And thank you all for joining our conference call today. We look forward to speaking with you again soon. Have a wonderful day.

Ladies and gentlemen, that will conclude today's conference call. Thank you for your participation.