Two Harbors Investment Corp. Q1 FY2025 earnings call

• Focus at RoundPoint this year: bring direct-to-consumer originations platform to scale, increase second liens offerings, evaluate mortgage finance landscape changes, grow third-party subservicing presence, generate cost efficiencies in servicing via tech/AI.
• Portfolio performed well in Q1 with RMBS and MSR contributing, risk exposures kept low. Interest rates ended Q1 lower. Fed held rates unchanged, revised GDP and core PCE expectations. Market expectations for cuts changed. U.S. policy instability raised dollar reserve currency questions.
• Book value increased, comprehensive income generated. MBS funding markets stable, repo spreads normalized. MSR market well supported, transfer volume normalized. Portfolio allocation and return potential discussed.

Total economic return was 4.4% for Q1. Book value increased to $14.66 per share at March 31 from $14.47 on Dec 31. Comprehensive income was $64.9 million or $0.62 per weighted average common share. Net interest and servicing income was higher due to portfolio shifts, offset by lower float and servicing fee income. Mark to market gains/losses were $38.4 million difference QoQ. MBS funding markets stable, repo spreads normalized. MSR market well supported, transfer volume normalized. Portfolio had $1.7 billion UPB of MSR committed to purchase. Return potential: ~65% capital in servicing with static return 12%-14%, remaining in securities with 10%-15%, static return estimate for portfolio 8.7%-12.3% before leverage, potential on common equity 9.1%-14.7% or $0.33-$0.54 per share quarterly.

• Spreads have widened in Q2, variable day-to-day. Actively managed portfolio and risk to take advantage of dislocations. Return potential higher today due to wider spreads. Portfolio leverage and risk kept at muted levels until economic clarity. Still see attractive levered returns on Agency RMBS, MSR portfolio generates stable cash flows across interest rate scenarios.

• Macroeconomic environment uncertainty due to U.S. policy changes (tariffs, trade policy, Fed composition). Instability raised questions about dollar as world reserve currency. Volatility in equity and fixed income markets. Uncertainty around Fed reaction to economic conditions (slowdown or inflation pickup). Convexity costs may be higher with realized volatility.

Q: Just given the volatility, can you give us an update on book value through April? And then also any changes you might have made in the portfolio to sort of adjust for the current environment?

A: Yes, sure. As you know, it's been quite a volatile April with all the activity that's happened. Bill said through last Friday, we were down about 3.5%. Nick said we have modulated the portfolio since April. We took our risk down by end of Q1 and even further into early April, then raised it a bit from where we were, ended quarter around 6.2% debt to equity. Saw developments in D.C. and markets, spreads widened, and still a lot of unknowns ahead.

Q: Can you guys talk a little bit about the big merger or acquisition between Rocket and Mr. Cooper and how you think that sort of impacts the competitive landscape in the servicing market and potentially the bulk MSR purchase market as well?

A: Yes, sure. Certainly, an exciting development in market. In terms of its real life impacts though, I think they may be more muted than the headlines might appear. Both those guys were active buyers of MSR. I think that as a combined entity, demand will stay the same as being equal to the sum of the individual demands. I think the bid is probably a little bit better than it was because now you have the maximum of the two individual bids, right. So, I think it will be a little bit more competitive than it was. On the margins though, I don't think it's a wholesale change. The securities market has reacted somewhat with regards to prepayment speeds in terms of pools that were serviced by Cooper being modulated to forward prepayment speeds being in line with more rocket speeds. So, we've seen an impact there. So, I think the impacts generally are non-zero, but it's not market change. The dynamics in the servicing market are largely the same as what it was. There's still, as we said in the prepared remarks, the supply of MSR is back to pre-COVID levels, and that's not going to change that.

Q: Can you guys maybe provide a little color around the change you had to the previously announced CFO transition?

A: Yes. Well, as we announced, William Dellal has been appointed Chief Financial Officer instead of Interim Chief Financial Officer. We're thrilled to have him here. William has added so much value and he's been terrific. And we're happy to have him. And we continue to benefit from all of William's experience and we're happy to have him. So that's all I'm really going to say about that.

Q: A quick follow-up to Doug's question just on the impact on spread. So, you said up $0.03 Is that both the high end and the low end or just volatility do anything just in terms of what happens to that range?

A: Hey, Bose, it's Nick. Yes, it's very high end and low end. And it's a little bit of a rough estimate, but yes, it's $0.03 on both sides.

Q: We're just looking at the book value sensitivity for the change in a tightening in spreads and I realize that in April maybe it's changed a little bit. But for a 25 basis point move, we may be expected the sensitivity to be a little bit higher. Is there a way to flush out kind of like what's going on with the sensitivity in the MBS portfolio versus the MSRs?

A: I'm sorry, you think that this so spreads up 25 basis points in the deck, we showed minus 4.9%? Bill said the tightening of 4.1%, you think that should be higher? Bill said as you know, if you go back down to Slide 14, right, we say that 65% of our capital is allocated to the hedge servicing strategy part of the portfolio and only 35 is allocated to hedge securities. So, when MBS spreads do something, the MSR part of the portfolio doesn't react really at all because that is incorporated quickly and immediately into the MSR values, right. That's where the MSR market prices itself and works. And so, by definition, right, all else being equal, you would think that our portfolio would have 35% of the spread sensitivity exposure to a portfolio without MSR. Now there's somewhat differences between capital structures and so forth. So there's slight differences, but as a base case, that's where I would start. Let's say whatever spreads do, we would be 35% of what portfolios without MSR would be. And that's true when spreads tighten and that's true when spreads widen. Nick said yes, Eric. So just I would point you to the fact that if you look at our net mortgage exposure as in my comments and if you look at it, quarter-over-quarter, our net mortgage exposure, which is on Page 17 of the deck, I think was down to about $5.5 billion, when you take out the effect of the MSR, which was down significantly. And we talked about this before, the amount of duration the amount of spread duration you have in a mortgage with a mortgage portfolio, it does depend on where you are on the coupon stack too, right? If we had all of that $5 billion in 2.5s or 3s as opposed to having the majority of it more up in the stack, you would have more spread exposure there. It's the notional amount and where it's located on the coupon stack that determine that sensitivity. And I think that if you distill down that the net mortgage exposure is lower, number one, and two, we did have a little bit more of a shift up in coupon. Both of those things will decline the amount of spread sensitivity. But again, that was something that we very intentionally did by the end of the quarter because we were sort of hunkering down getting into second quarter with all of these changes flowing through the macroeconomic world.

Q: So, it sounds like you think that volatility at least in the near term will continue to remain elevated, especially on the realized side perhaps? And given maybe the more deeper negative convexity in the portfolio, has there sort of been more hedging activity that could impact the static return estimates that we should be thinking about?

A: Hi, Harsh, how are you doing? Yes, I mean, look, we know that a lot of realized volatility is never good for a hedge mortgage portfolio. So, we clearly ourselves and I'm sure other REITs have been experiencing more convexity costs than they would under other circumstances. But we have been, as you know, and we've been keeping our risk pretty tight through this whole time period. I think this is a time period where you have to be very cautious about taking on any sort of excess risk in the portfolio. But there is compensation for it, of course, and the fact that spreads have widened, right? So, I mean, there are pluses and minuses to it. Yes, convexity costs are going to be higher when you have more realized volatility, but thankfully spreads have widened in concert with that, which should mitigate some of that convexity cost.

Q: Maybe as you've been stepping back into getting more spread exposure, maybe towards the April, is it fair to assume that the up in coupon bias still remains or where you see relative value across the coupon stack today?

A: Yes, that is fair to assume. We really do, I'm what the issue with lower coupons for me is I think they trade in such a technical way that the value across the curve is on an OAS basis at least is fairly evenly distributed, but like everything else, there are trade-offs. But the lower part of the stack continues to be extraordinarily technical in the way it trades, and I think can trade in ways that are not necessarily intuitive. So, at the moment, we do like being more up in coupon than down in coupon.

Q: Your static return estimate on Slide 14 appears to have widened a bit. Can you talk about what drove that and how has varying your leverage impacted this measure compared to last quarter?

A: Yes, sure. Thanks for the question. As Nick said in his prepared remarks, we added this other dimension of changing the portfolio leverage a little bit to the lower and upper end of those ranges. And really that was a reflection of when we're dynamically and actively managing the portfolio, especially as we have been this quarter, it's important -- we felt it was important to show the range of expected static returns, projected static returns that are available over the medium term in a sort of equilibrium style leverage portfolio. As Nick said, we closed the quarter with leverage at 6.2%. We've got down as low as low 5% s and we're back up towards the high five is right around six now. So, it seems like it would be not as informative to show these static return estimates moving around so much as we actively manage the portfolio in order to take advantage of dislocations and opportunities in the market. And so, we've made the higher and lower end of the range. There also include some leverage ranges, basically at the low end of the range, something like six at the high end of the range, something like seven, right. In addition to the changes in the funding spreads and prepayment rates that we've been doing for some time. And so that's the source of why that's widened out a little bit, but we think it gives a more reflective view as to what the portfolio really is.

Q: I just have three questions or three things I'd like to hear from you about. One is liquidity. You say you maintained high liquidity, but you could just put some rough numbers around that, that would help.

A: Good morning, Merrill. It's we maintained our liquidity at very high levels from the end of the year, and we're maintaining a lot of cash and capacity in our borrowing.

Q: And on the recaptured, you referred to the -- whatever $179 million of loans that were slow and recaptured, I assume. What's the nature of that? How much of it was recapture versus flow? And how did that meet your expectations relative to the market runoff, because the portfolio is ever so slightly larger. I'm just wondering if you're holding your own against the runoff with the recapture effort.

A: Yes. Thanks for that question, Merrill. The organic recapture coming from our direct-to-consumer channel is still very low. It's still in its nascent stages. As Nick said in his remarks, I could put a finer point on it, only about 0.5% of our portfolio is re-financeable from a rate and term perspective right now, right? And so, there's very little ability to recapture some of those loans. However, we are seeing -- within that 0.5%, we are seeing some increased activity in our direct-to-consumer channel there. So, I'm very optimistic about what we'll be able to do when we bring the platform fully to scale, right. We are and I think we'll be able to achieve some really good numbers there once we get there. We have been active in the flow market in small size, right, and able to replenish the portfolio somewhat, as well as being participating in the flow market. And so, we're confident that we're going to be able to maintain our servicing portfolio or even grow it over time. And we think we're going to be using all of the tools that are available to us in both the bulk and flow markets and eventually the DTC once that gets more up to scale and fully formed.

Q: And the last one is quite small, but you threw out a mention to being interested in the Ginnie Mae market. Do you see price dislocations there? Or is that just sort of a safe harbor for return generation when the GSEs are being recapped and released?

A: I think it's many factors. I mean, we're taking the beginning stages in order to get involved in that market. But the rationales are number one, just to have the ability to participate in the broader aspect of the servicing market rather than just conventional, the ability to be able to service and subservice for Ginnie Mae's. It is slightly cheaper. It's a little bit more opportunity there. And I think being able to be a more full-service mortgage originator slash servicer sort of requires us to be involved in GMA market. So, we're taking the beginning steps in order to do that.