Two Harbors Investment Corp. (TWO) Earnings Call Transcript & Summary

Okay. Good morning, and thank you for joining us. I'll kick off the 18th Annual Barclays Global Financial Services Conference. I'm Barclays Consumer Connect Analyst, Mark DeVries, and I'm pleased to be joined by Two Harbors' CEO, Bill Greenberg; and CIO, Matt Koeppen. We'll be conducting a fireside chat, but we'll break it up with some polling of the audience. And we'll also leave time for any questions that come in from the audience during this session. [Operator Instructions]

Before my first question for management, I'd like to lead off with a question for the audience. [Operator Instructions] Turning to the first question for the audience, what do you view as the biggest driver of total returns for Two Harbors over the next year, higher leverage, lower funding cost, lower spread volatility, stable interest rate environment or other? So as you respond, we're going to move on to some questions. First question for management. Can you provide an update on book value quarter-to-date, if there are any worth mentioning?

Yes, I'll take that one. Good morning, and thank you for hosting. We are estimating total return of about up 2% for the quarter through the end of August. So far, it hasn't been a particularly volatile period with the summer months and mortgage spreads have generally been on a slow grind tighter.

Okay. Great. Can you talk through how you think about the dividend level in this environment?

Yes, I'll take that one. Again, thanks for having us very much, Mark. It's a pleasure to be here. The way we think about it in this environment isn't particularly different from the way we think about it in any other environment. We always try to align our dividend with the earnings -- with the economic earnings power of the portfolio. We try to set a dividend with more than 1/4 in mines and its potential sustainability going forward. Due to the COVID virus this year, we gave a little bit more guidance on a forward-looking basis than we have in the past indicating that we think $0.14 dividend is achievable here, and that's still true. One thing I would note is that for the remainder of 2020, we do expect our dividends to be characterized as a return of capital based on the activity that occurred during the first and second quarters.

Okay. Great. Moving to the next question for the audience, if you could, but what is the biggest risk to the shares that you see as investors, low asset yields, rising funding costs, spread volatility or interest rate volatility? Please register your answer, if you can. Moving back to questions for management. Can you provide an update on forbearance and delinquencies in your portfolio? How has your liquidity position changed quarter-to-date? And what level of defaults could you support with the current capital?

Sure. I'll take that. I think one of the biggest surprises in 2020 in our portfolio and maybe many others has been the level of forbearances and delinquencies for these assets. As of the end of August in our servicing portfolio of $160 billion or so, we were experiencing at the end of August, we had 5.1% of our portfolio was in active forbearance. But of those, 27% had made their August payments. So that's a net rates, those that are both in the forbearance and delinquent of about 3.7%. And we've seen the gross forbearance rate decline even further in the last -- in the early days of December, going from 5.1% down to 4.8%. So we feel very good about what that means for our liquidity, for our advancing obligations. They're very manageable. We are working on some facilities, as we've talked about in the past. We have -- and as we'll talk about in a little bit, I think we have increased risk a little bit, so that's drawn a little bit into our excess capital. But overall, we still think these numbers are very, very manageable. And we're very confident with our liquidity position.

Okay. And could you remind us how high you think those delinquency rates could go and still be manageable?

Well, so number one, that's an interesting question. Obviously, there are many factors that could go into that, whether there's a second wave, whether some of the things that we saw with like the PPP expiring in July, whether that puts additional stress on borrowers. So it's hard to say how high it could go. We feel that we're very well protected and very well able to manage if those numbers were to quadruple. That seems like a very, very unlikely scenario. When we started this thing in March and April, people were talking about scenarios that could have been 30% or 40% delinquent. And obviously, it's 1/10th of that. So our philosophy has always been to, hope for the best and prepare for the worst. We have scaled down our expectations a little bit from what they were at the beginning, but we still feel confident that we could handle many multiples of where we are.

Okay. Great. Do you have an update you can provide on the servicing advanced facilities that you mentioned in the 2Q earnings call?

I'll take that one. We have actually made good progress on that front. And while it's not closed, we expect to close our first servicing advanced facility this month. It does take a high degree of customization. So it does take some time, for example, the -- any facility needs to have connectivity built specifically to a specific sub-servicer, but everything has been moving according to plan. And as Bill was just describing, our forbearance experience has been pretty favorable. So the timing for all this is working out just fine. It will be up and running in plenty of time for any advances to flow through for us.

Okay. Got it. Moving to another audience question. What do you think is Two Harbors' best use of capital at this time, maintain dry powder, invest into Agency MBS, invest into MSRS, increase the dividend, buy back stock or other? Turning back to management. What's your current outlook for returns on agency mortgages in this environment?

Sure. We see agency mortgages hedged with swaps, generally to be, I would call it, in the high single-digit area. Maybe into the double digits given the spread tightening that we've seen in the last couple of quarters. We do still see agency mortgages paired with servicing to be in the sort of low-mid double-digit gross return area, which we still see as attractive. And as a note, as we sort of indicated after the second quarter, we expected to be adding some risk in the RMBS market, and we did -- we have managed to add about $3 billion in current coupon mortgages this quarter so far.

Okay. What coupons do you currently find attractive for your Agency MBS? And are you looking at particular spec pools to help manage prepays in this low rate environment?

Sure. We still continue to like the Fed's targeted coupons serving the current coupon part of thing. So that ends up being 30-year TBA 2 and 2.5 coupons. The continued Fed purchases, they've been buying, call it, 1/3 of the new production, which is -- production has been high, but the takeout has been very high, too. That's led to an attractive dynamic where those coupons are supported in spread terms, and there's also a significant role advantage in funding. In terms of specified pools, I think you're aware that all of our specified pools are high-quality call protected stories. So in the higher coupon part of the sector, we do still continue to favor specified pools.

Got it. How are you thinking about leverage here?

So I'll take that one, Mark. For us, leverage means a little bit something different than it does for many people, mostly due to the presence of MSR in our portfolio, right? So we like to think more in terms of what we call a drawdown risk measure. I like to say that when people ask about leverage, what they're really asking about is how much money can be losed due to spread movements, right? And because the presence of MSR in our portfolio. And when mortgages widen or tightened, there is a corresponding offsetting impact in the MSR performance. We are able to run a higher leverage with lower spread risk -- mortgage spread risk than a portfolio that does not contain MSR, right? And so that's the way we look at it. We focus more on risk than on a nominal leverage number, although we are respectful of that. And with that, we can translate our leverage numbers to that risk measure. And so we think that we're comfortable running, as we indicated, I think, in our second quarter call, running a leverage number that translates into an 8 or 9x number, but the way we think about it is really in terms of the spread risk part of the portfolio.

Okay. And -- but having said that, how should kind of appetite for leverage today versus what it might be, what it might have been 6 months ago or kind of what you anticipate on a go-forward basis?

So I think if you look back 6 months ago, I think it's roughly similar to where we were then. As you know, our portfolio has undergone some changes. We've simplified our portfolio with not having any non-agencies in the portfolio and being an agency-only REIT at the moment with Agency RMBS and Agency MSR. And so that will result in, again, a nominal leverage that looks higher than it did 6 months ago because non-Agency were generally more lowly levered. But if we think about the agency part of the portfolio only, I'd say that's about the same as where we imagined it before.

Okay. Got it. One last question for the audience. Over the next year, would you expect your position in Two Harbors to increase, decrease or remain the same? And thanks for participating on. Turning back to now on the remainder of the questions for management. Can you comment on any additional MSR portfolio activity outside of the $4.5 billion of MSR flow commitments in July that you called out on your 2Q call?

Yes, sure. Sure, I'll take this one, Mark. So as I think everyone knows, mortgage origination volumes have been at record highs. As a result, we've been able to achieve, what for us, is record high servicing flow production in our portfolio. For both July and August, we were able to lock commitments for more than $5 billion each month, more than $10 million so far for the quarter. And September is on a similar pace to that. So we're very happy with that. On the bulk side, we have seen the bulk market coming to life a little bit. We've unfortunately not won any of those packages yet. We have bid on a bunch. We've been high bid on several that unfortunately have not traded. But that market is slowly coming back to life. We expect there to be more volume in the future as participants ultimately try to sell in the future.

Okay. Is that bulk markets still impacted by too wide a spread between what buyers are willing to pay and what sellers are reluctant to or willing to part with it?

Yes. I think that's 1 way to say it. Another way to say it is that, to my eye, the origination gains are so large, the spread between primary and secondary spreads are so large that there's not as much of a need, a cash need for originators to sell servicing in order to monetize the asset in order to help run their business. And so I think many people when they see Power Malts being 3 today instead of 4, which there's a reason for that, mainly because the primary, secondary spread is so wide that they think that it's better for them just to wait and hope for mean reversion and prices to go up rather than what the market prices are today. So yes, I think that's one reason why the bulk market is and the flow market, too. Is not being able to take out as much origination as there is it actually being produced.

Okay. Great. How has prepaid speeds performed in the quarter? And how does the -- how do the prepaid speeds in your portfolio compared to the broader TBA universe?

Yes, I can take that one. Well, the sort of cheapest to deliver collateral out there being the most financeable major -- sorry, major pools have been pretty fast. We've seen that the refi machine has been efficient and effective. I think that surprised people early on as we got into this environment, but people have adapted. We've heard that mortgage originators have actually been adding capacity if they've been hiring in that sector, and refi capacity has been increasing. So the most refinanceable collateral has been paying pretty fast. Now in specified pools, like I said, there's -- we -- that's where most of our specified holdings are. And those speeds, I would say, have been more -- they've been in line with expectations. Those stories are holding up well and not really showing us any particular surprises.

Okay. One of the things, Bill, you just alluded to is the kind of the unusually wide primary-secondary spread. Kind of wondering what your thoughts are on how much that can compress and how much more prepay speeds could accelerate even if the kind of underlying interest rates don't move here? And how do you think about that in terms of what you're willing to pay for servicing?

Yes. So I think as always happens, most always happens in refinancing environments, that the primary-secondary spread usually widens out until the origination machine can work through the near-term capacity. And then assuming rates haven't moved up since then, then it can tighten and get lower rates to borrowers and then the thing starts all over again. And so I think it's unlikely that prepayment speeds increase a lot from here. Capacity is already running at what it's -- at its maximum. And so I don't anticipate that. What I do anticipate is fast speeds will stay fast for longer here. If we're making $300 billion a month in origination and that's $3.5 trillion a year. The size of the mortgage universe is more than $5 trillion. That's more than a year's worth of refinancing, assuming every mortgage in the universe is refinanceable, which it basically is, it's more than a year's worth of activity that we can sustain here at these levels. So I think that is the more likely path here, assuming rates don't ultimately back up.

Okay. What's your general estimate for the percentage of loans out there that are in the money at this point to refinance?

So it's almost all of them. It's 80%, 85%, something like that. I'm sorry, you also asked a question about servicing, which is something which I alluded to in one of my previous answers, which I can expand that more here, which is, all else equal when we think about -- when you value servicing for pricing or hedging, we look at what we think the long-term speeds might be on such a thing. And so as a result, we have to project forward what we think is going to happen in, not just for scenarios where interest rates change, but where interest rates stay the same. And when interest rates stay the same, as you pointed out, the primary-secondary spread will compress over some period of time, right. And that is the reason why Power Malts today are more like 3 as opposed to what they were 6 months ago when they were more like 4. And so to someone who isn't looking at that carefully or building that in, it might seem like the nominal price is lower today, 3 versus 4. But for us, it's really the same. And we look at the values and the projected returns of the servicing assets being roughly the same as it was 6 months ago, even though the nominal price is 3x as opposed to it was 4x because we're projecting primary-secondary spread to come in, in some way.