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

Good afternoon. Thank you all for joining us for this fireside chat with Two Harbor's CEO, Bill Greenberg; and CFO, Mary Riskey. We have a number of prepared questions we'll be going through them, but if anyone in the audience would like to ask a question, click on the Ask-A-Question button in the upper right-hand corner of your screen and follow the prompts to submit your question, and we'll do our best to address those in the time that we have today. We've also prepared a number of audience polling questions that we would encourage you to answer during the presentation, and we'll be publishing the results in our report summarizing the takeaways from the conference for those joining us.

First of all, just dispense with the book value question. Any updates you can provide us on workbook value per share quarter-to-date?

Sure. Our quarter-to-date through August 31, we estimate that our total return on book value is around 2.5%.

Okay. Great. Next, what's the current outlook for returns on agency mortgages in the environment with the kind of moves we've seen in spreads since you last reported?

Yes. Thanks for that question, Mark, and thanks for having us here today. It's a pleasure to be here. We see returns on agency mortgages to be in the mid-single to high single digits in returns depending on coupon and story. Of course, a common theme this year has been the attractive role specialness in the local bonds, I mean. So their returns can be higher if you assume that role specialness last forever, which, of course, it won't, but we think if you do assume it that on 2s is probably right around 10% and then 2.5 is probably in the mid-teens, but if you assume that's going to end in some period of time, then that's probably not that different than where we see specified pools, as I said, in the mid- to high single digits.

Okay. That's helpful.

But if I could add one thing there, Mark, is that, as I think you know, we focus on what we'd like to call the paired construction with Agency MSR, right? And the paired construction, depending on what coupon story can certainly be in the low teens with where we're seeing servicing today.

Okay. Given some of the comments you made about TBAs and the returns come depending on your view of how long specialists persist. How attractive did you view those returns relative to particularly pools with the paired construction?

Well, not sure I understand the question. I mean the role specialist has been attractive and powerful for a long time. And it's a little bit of a statement about how long the thing is going to persist and when it ceases to be attractive or start to be less attractive, how is the coupon going to behave as people -- and the demand for that coupon becomes less because the role specialists is less, because surely on a spread basis, the coupon is not as attractive as it was compared to historical returns. I think everyone knows that Agency RMBS on a spread basis, say, is 10 basis points or so tighter, richer than previous periods where the Fed has been involved, and it's probably something like 30 basis points richer when compared with periods that the Fed has not been involved. And so the role specialists has been an important part of contributing to returns in this period of tight spreads.

Okay. And how long would you expect role specialist to kind of persist at or near current levels?

Well, I think, it's intimately -- well, if I knew that, then I would be happier than I am. But I think it's intimately intertwined with the Fed taper and the removal of that accommodation. And of course, it's not a surprise. We're not the only ones who know that the Fed is going to taper. And so lots of market participants are underweight mortgages here. And so it's an interplay between the Fed removing support and the desire for those who are underway to be buying if and when mortgage spends widen somewhat.

Okay. Got it. What coupons do you currently find attractive for your Agent CMBS portfolio?

Yes. That's a good question. We see spreads across the coupon stack to be pretty similar, static spreads to be pretty similar. But the risks are pretty different. In the high coupons, there is uncertainty with regards to prepayments and so forth, but convexity is lower. And then we've seen some of the effects of that prepayment uncertainty in the second quarter where faster-than-expected prepayments and faster-than-expected prepayment expectations caused a widening in those assets. And so we think a lot of that repricing has already occurred. And to the extent that there is prepayment uncertainty, it's been our view that it's just as likely -- maybe more likely that prepayments will surprise the slow side than the build to the fast side. And the low coupons, the story is exactly opposite. That repricing has not occurred the prepayments are largely known or expected to be known, right? But what's unknown there is the convexity, right? How is the price going to react as interest rates move and as the Fed removes a combination. And so we generally -- in our positioning, we hedge our lower coupons exposures and mortgages with our servicing, and we have this high coupon exposure, which we think has already experienced the repricing as a result of the risks inherent in those assets. But on a static basis, they're all pretty similar, which gives, as I said, mid- to high single digits depending on everything.

Okay. Have you done any kind of material repositioning across coupon stacks since you last reported?

No, I wouldn't say so. It's been mostly maintenance in terms of moving things around a little bit. We continue to add servicing and low coupon mortgages on the margins.

Okay. Curious to get your kind of thoughts on just spreads here, kind of the direction, do you think they've widened out about...

I am not sure who is frozen, if it's me or Mark.

As much as they're going to? Or do you see more...

I'm sorry, the connection wasn't very good. I think you -- can you repeat the question?

Yes, yes, sure. The question was just kind of what your thoughts are on the directional of mortgage spreads here, they've been widening out a fair amount in the last couple of months. Do you think that's done, mostly priced in? Or do you see more room for volatility there?

I do see more of volatility as the Fed removes a combination. One thing that's different, I think, today than what you saw in the taper tantrum in 2013, is that most asset classes are rich, historical type levels, right? And so I do think it's unlikely that mortgages can widen in isolation of other things. There's just too much money on the sidelines waiting for investments. But I mean I think that when the Fed stops buying $40 billion a month, every month and prepayments are still zipping along, someone's going to have to buy that amount of mortgages or mortgages move on. And so I do think that will occur, as I said, compared to history, historical levels, 10 basis points tighter than where the Fed has been involved, 30 basis points tighter than where they've not been involved. I do expect that to happen, whether it happens in an orderly fashion or disorderly refashion, I guess, I don't know, but I would be surprised if it didn't.

Okay. How are you thinking about the relative attractiveness of -- versus -- of generic versus specified pools here?

Well, by generic, do you mean generic specified pools? Or you -- by generic, you mean TBAs?

Yes. Just TBAs or more generic collateral.

Yes. Well, as I already mentioned, the attractiveness of the role specialist is still there for this month, for next month, for the month after then through December, the roles are so special. As things change, that is going to come off and one will have to rely on the inherent spread of the assets in the local bonds, which as I said, are mid- to high single digits by themselves, which aren't bad, but have been wider and more attractive, as I said. Should rates rise, right, the prepayment protection embedded in some of the specified pools will become less valuable, but everyone knows that, and then we hedge to longer durations to reflect that. One thing that I think would be true for our portfolio and maybe many other portfolios out there in the world, is that -- because the roles have been so special, we have a fairly large TBA position, right? And as that changes and as -- especially if current coupon spreads widen and specified pools widen, I think, you will probably expect to see our relative positioning move away from TBAs towards specified pools as that occurs, but we'll need to see some more signs if that's happening before we do that.

Okay. How are you thinking about the appropriate amount of leverage in the current environment?

So we think about leverage maybe a little bit differently than many people. I think for -- but they way I like to say it is this, for portfolios without MSR in it, leverage and risk are really the same thing, and you can easily go back and forth between one another. Because of the properties of MSR, which hedges the current coupon mortgage exposure, of our portfolio, it's not the same thing, right? And so I always ask really, when you're asking the question, where do I think leverage can go? I think you're really asking how much money do we think we can lose, right? And when we have a whole bunch of servicing, we can withstand, and we talked about this just a moment ago, if I expect spreads on lower coupon, current coupon mortgages to widen, we are very well hedged with regards to those spreads because of the presence of MSR in our portfolio. We are exposed, as we discussed in the second quarter to spreads in the higher coupon sector, but our low coupon, current coupon mortgage exposure is very well hedged. And so our leverage is low right now. But the way I like to say it is our exposure to mortgage spreads is also very low, right? And we can -- in our earnings materials, every quarter, we have a little chart that shows what will the book value exposure be for a 25 basis point spread widening in current coupon mortgages, and that number is very low. And because of our view on Fed tapering and removal accommodation, we're keeping it very low. That said, should we buy more MSRs? Should mortgage -- should current coupon mortgages widen out as we think they might? Then you will expect -- so you should expect to see us add more mortgages, right, which will increase our leverage somewhat. So if our leverage right now is in the 6 area, which is what we disclosed last quarter, we're very comfortable seeing that number move up to 8 or 9 times in an environment where mortgages are more attractive.

Okay. Great. In the second quarter, you increased your MSR portfolio allocation from 15% to roughly 20%. What do you view as the appropriate mix of MBS and MSR when it comes to capital allocation?

Yes. I think that's a -- it's a dynamic question, and it changes with market environments and with interest rate levels and so forth among the factors that we look at are the hedge properties of the MSR or the hedge properties of the MBS relative to the MSR in different environments, and where that could move should things change. So we don't have a target in mind, but we can certainly accommodate more MSR here, and we're actively in the market trying to buy more of it. So...

Okay. And how is supply of MSR? I mean we were just speaking to one of the larger acquirers of it half an hour ago, and they commented on finally starting to see a really healthy increase in supply. A lot of originators who are not generating positive cash flow that were before now are kind of being forced to become...

I think you're frozen, again, Mark.

More active sellers. Is that kind of consistent with acquired MSR here?

Yes. I think that's largely true. I mean to put some numbers around that so, back up for a moment. So we acquire MSR really through -- and probably many people, through 2 channels. We have a flow channel. And of course, we participate in bulk auction market. The flow channel, we have a network of between 1 and 2 dozen flow sellers, some provide more loans to us during sometimes, some during other times. But generally, I would say, as a rule, the flow volumes that we experienced, largely track with prepayments. And so when prepayments were at their peak, we were seeing very heavy flow volumes. Now that they've come off a little bit, we're seeing somewhat less. The bulk market has been very active for the reasons that I think that you just mentioned that we've talked about in the past that many servicers when mults were low and rates were low and primary secondary spreads were wide, didn't have the need or the desire to sell into the market at those levels and then retain them. And I think the normalization of the primary secondary spread today as well as rates, while they were a lot higher as everyone knows. But rising rates has allowed these guys who hadn't -- had used potentially to monetize a little bit to recognize some multiple expansion and makes some extra money there. So I think we are seeing some of that come to market. And they start to say to put some numbers around it. If an average year in bulk volumes is, say, $200 billion to $250 billion of supply. We saw about $55 billion of supply in the first quarter. We saw about $100 billion -- I forget that it was $130 billion or so in the second, and we've seen $100 billion already in the third, right? So we're already north of $300 billion in bulk supply from small guys, medium-sized, big guys. And I expect that to continue through the balance of the year?

Got it. How is the demand for that supply lined up? Have you seen some attractive pricing given that strong supply?

It's competitive. There's lots of buyers in the world for bulk servicing and it's interesting, some packages that I think should have higher demand, have less; and some that I think are less, have high demand. And so it's situational. And we participate in them. We don't buy every package for sure, but we're targeting certain returns, and we're able to achieve that. And so we're pretty satisfied with our activity.

Okay. And could you talk a little bit more about the number of flow partners you have, and how tough the mix you're seeing between flow and bulk has been?

We have north of a dozen flow partners right now that are delivering to us. As I said, the flow that we have as it moves with prepayments and it largely offsets our runoff, and then we could be active in the bulk market. So we can't predict when packages are going to come to market and what they're going to look like and whether we're going to be the best in business and so forth. So that's more opportunistic in a way the flow right now is more predictable.

Okay, understood. Yes. Moving on to prepay speeds. How have those been tracking this quarter? And how does that compare to kind of like recent expectations?

So again, I think, the prepayment market is a tale of Two stories. The low coupons, I think, have shown themselves -- as I said, everyone has expected them to be very reactive to even small changes in rates. And so over the last 6 to 9 months as we've seen the rates go from the lowest to the highs back to lower rates. We've seen prepay speeds on 2s and 2.5s and the correct coupon complex in general, react quickly to those changes. Loan sizes are up in those areas. And so they're very reactive. In the high coupon sector, I think, some time ago, people had called burn out to be elusive. And certainly, prepay speeds were stubbornly fast for some period of time. But I think what happened there was that their lack of initial reactivity caused some market participants to declare burn out over absence and to project those fast feeds far into the future. And I think what we've seen over the last couple of reports, is that prepay speeds on high coupons have started to slow. They've slowed more than day counts would suggest. And so I think those have generally been a little bit slower than expected. Mary indicated that we're up a little bit this quarter so far, and really the drivers of that have been the performance of the higher coupons as a result of moderating high coupon speeds.

Okay. Great. Turning to the Fed and the imminent taper. Can you just talk about what are your expectations there for when, how long, at what pace? And how you think the market ultimately reacts to all this?

I have no special insight into that, that other people don't have. So I mean it doesn't matter what I think. It sounds like it's soon, right? Maybe we will get an announcement next week about a start date. But even if they don't start in November, they'll probably start in December. All the general consensus is that they should be done on the mortgage side by the middle of next year. I have no reason to think anything different than that. As I said, I mean, the removal of $40 billion per month of buying is significant and someone else is going to have to buy that or spreads are going to widen.

Okay. I think as you've indicated, obviously, you don't think that's fully priced in now. How does that impact how you kind of position the portfolio here?

Well, in terms of our leverage, as we said, it's lower than it could be. Should spreads widen out, I would expect that to increase. But as long as spreads are where they are, we find attractive opportunities in the paired construction of owning MSR with RMBS, where we don't care what happens to the spreads, right? I can put together a package when we buy MSR and hedge it with probably coupon mortgages in the low teens returns. And I don't care whether mortgages widen or not a little bit. That said, our leverage is low, we have significant dry powder to put to work should mortgages widen out some in order for us to take advantage of that.

Okay. Got it. Just given the opportunity you said you see here, how are you thinking about the prospects of raising more capital? Is the investment opportunity attractive enough that if you get an opportunity with the shares above book to issue more that you'd be active?

Yes. Well, as you know, we recently did raise capital, right? And the thinking there was that we wanted to be able to react quickly to changing market environment. That's still true. We always think about whether a capital raise would be good for shareholders if you have something to do with the money. And if we see that opportunity again, you might expect us to behave similarly. And we see plenty of attractive opportunities in the MSR market, in particular, but especially when paired with RBS.

Got it. And how are you thinking about the kind of the proper balance between common preferred equity in the capital structure?

Yes, that's a good question. On the one hand -- well, I mean, everyone knows preferred is really leveraged to the common shareholder. And of course, the cost of that capital acts as a hurdle rate for your marginal investments that you have to make more than that, otherwise. But if you do that, that's accretive to the common; if you don't, then it's diluted to the common. We're very happy with our relative mix at the moment. I think as you know, Mark, we call some of our preferred shares earlier in the year. That brought out our ratio of preferred stock to total stock holds equity down to around 25%, which we think is the right amount for our portfolio mix and our portfolio composition. MSR -- again, going back to the leverage question a little bit or statement, as I said, preferred can be thought of, right, as just leverage to common. And so if I can put together a package of mortgage servicing with RMBS, which has very little risk relative to mortgage spreads, that's a way -- that's a perfect use for preferred, right, as opposed to common, but we think that the mix that we have is the right amount for now.

Okay, great. Anything interesting happening around -- on the financing side, whether it be on the repo cost, ability to extract a little bit of leverage out of your servicing assets worth talking about?

Yes. Maybe I'll let Mary do that one, if that's okay?

Sure. So we have several MSR facilities in place. We are continually in conversations with our counterparties to have those reflected at market rates. We have plenty of capacity. And so that market is looking really good as is the agency repo market has been very stable.

Okay. Great. Just talk about the outlook for the dividend here, given kind of the current level of core earnings and ROEs that you expect to generate from the investment portfolio.

Sure. Our dividend philosophy really hasn't changed. We set the dividend to reflect the economic returns of the portfolio over time and with more than one quarter in mind. We feel good about our current level, estimated gross returns on new investments at MSR paired with RMBS are in the low to mid-teens, and we will now continue to evaluate the dividend in the context of those return expectations.

Got it. And anything in D.C. that you guys are following from a broader policy or mortgage and housing-related policy that you're keeping a close eye on that could potentially have an impact on the portfolio?

Well, sure. I mean there's news reports yesterday about a new FHFA director. I think I've read the same newspaper articles as everyone else. So I can't tell you more than -- about him other than what I have read that could potentially be interesting to see how that impacts GSE reform or prioritization or guarantee fees and their potential impact on prepayments and all of those things. So there's a lot there to pay attention to. I'm not sure that there's that much that we're going to actively do as a result of those things. We're still sitting here executing our strategy of buying servicing and hedging it with Agency RMBS. So I don't think there's anything there that would cause us to change that. Or another related set of things that are out there in the world are in the CFPB or other potential regulatory postures that they could take that will change or that will impact servicers. I don't particularly think, again, that will change the way that we think about our servicing portfolio or profile. I think us and many of the servicing industries have -- are already operating in a heightened compliance environment and are doing everything they can to be borrower friendly and to help ours through the stress that they have during this crisis. So even though as we switch from one administration to the next, where maybe the postures are slightly different, I don't think that will change the way that we think about our portfolio either. But obviously, we want to pay attention to all these things to see whether it does change our risk or increase our risk or provide opportunities for us to take advantage of certain things.

Okay. Great. Well, certainly been a very helpful conversation. I think we'll hand on that note. But let me thank you both for your time and insights. We really appreciate it.

Thanks very much for having us, Mark. Appreciate it.

Thank you, Mark.

Thanks.