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

Great. Good afternoon, everyone. Thank you for joining us. Happy to have Two Harbors back with us here at the Crédit Suisse Financial Services Forum. Joining us from the company today is CEO, Bill Greenberg; and CIO, Matt Koeppen.

With that, why don't we get started? I would say 2020 was definitely kind of a year of transition for Two Harbors and both of you guys individually. Can you just talk a little bit about kind of how stepping into the new roles you did last year? And what was obviously a challenging year and kind of how that unfolded?

Thanks very much, Doug. And then I just wanted to just first say, thank you very much for having us here and for hosting this conference. We always like this conference, and you guys always do a great job. So thank you to you, Doug, first point, as well as a Crédit Suisse. We're happy to be here. Yes, there's no doubt about it. 2020 was a very challenging year. And it's especially challenging to step into my new role in the midst of this market volatility and the pandemic issues, which required everyone to transition to working from home right away as well as the portfolio repositioning that we embarked upon during that time period as well. So it was really a lot going on for us during that time. Nonetheless, I think when looking back at that time period, I'm pretty happy with the way things ended up. I think the team really performed very well during all this. We were able to execute an awful a bunch of stuff that we worked on during the year. We completed the internalization of the company. In August, we rightsized our portfolio, and we repositioned the portfolios in the agency plus MSR-only focused REITS. And most recently, as I think you've seen, we were able to take steps to rightsize our capital structure through issues of -- from a refinancing of the convertible note and call in some preferred stock. In terms of me, personally, one of the things that's been most interesting to me, maybe not surprisingly, you might say, is as I've gotten to work more closely with people in some of the other departments that I was less exposed to before, people in the finance department or the legal department, the IT department, the HR departments, I have, of course, been very, very impressed with just how smart and dedicated and engaged everyone is. And I'm obviously very appreciative of them, very proud of the whole team and everything they've accomplished this year, and I'm super optimistic for the future. So stepping into the new role has really given me a new perspective, maybe sort of obviously on the company and the future.

And I would add, and also thanks for having us today, Doug. I agree with those comments. And from my perspective, I think it hasn't been as dramatic of a change as what Bill has been going through and experiencing. Having been co-CIO for a year or so really put me solidly into that role. So it's mostly a continuation of those duties from my perspective. One thing I would add is that one thing that's been a constant, Bill and I have been working here together since 2012, and we've developed a great partnership. So I think we're both looking forward to continuing on that path, and I'm also optimistic for the future of the company.

Great. And Bill, as you mentioned, you guys completed your internalization in August of last year. Can you just talk kind of how you view -- were there any kind of changes to the day-to-day running of the business? And then also kind of where you are in terms of kind of cost saves from that internalization?

Yes. Sure. As a direct result of the internalization, there's very little to be -- there's very little changes in business strategy that will occur. As Matt said, we've been running the portfolio together for the -- we've been co-CIOs for the last year. We were important members of the development of the strategy in the years before that. So the business strategy is one that we've helped to create, that we believe in fully and the internalization doesn't really change any of that. As far as the cost savings or in terms of what was projected versus how things are unfolding, I would say that, in general, we're running better than what we had initially anticipated. I don't have the exact numbers in front of us. Mary has those in detail. But I'd say, in general, we're running better and more favorable in terms of cost savings that we projected.

Great. Now why don't we pivot to kind of just the broader market? Obviously, kind of been -- rates kind of front and center with kind of a significant move. If we could just start there, kind of how you're viewing this move in rates and kind of how you think that might continue to play out over kind of the -- both the short term and kind of the longer term?

Sure. I can start with that one. I guess -- so we've seen a couple in the last couple of months, I think we've seen a couple of bouts of curve steepening, and both of those really were combined with some current coupon spread widening. I guess, I would say -- there's a few things you could say about that. I guess I would expect that if the curve continued to sort of bear steepening like that, I think you'd see a continuation move in mortgages as well. I think the things that have been kind of been driving that really spurred in the last couple of months from sort of positive developments on the vaccine front and positive developments on the fiscal stimulus front, all of that leading to expectations of future economic growth, growth in GDP, unemployment falling, maybe actual realized inflation and higher inflation expectations, which then would lead people to probably conclude that maybe the Fed could slow down or taper its buying activities to some extent. So that's what we've seen. The steeper curve is nice for mortgage investors, of course, right? So that's going to have the effect of probably causing a reduced refi response, so we could fall into a little bit slower speed environment, which I think would be a relief for us and others. One interesting thing that we've seen also in this little move recently is that the primary-secondary spread, which I think we all know has been holding at steadily wider levels actually decreased pretty significantly. I mean, I think depending on exactly how you measure it, it's probably within 10 or 20 basis points of long-term averages now. So that's getting much more normalized. Things are moving quickly, but it's certainly been more normalized. And so what you can see in further interest rate increases, it might actually flow right through into the primary market rate, which again, would dampen fast prepayment seeds.

Great. I guess you mentioned Fed expectations around taper. Can you just talk about kind of how you're viewing -- what your expectations are for taper? And kind of how you think that the market is positioned for that inevitable taper today versus 2013?

Sure. I think there's been a general consensus out there that the Fed probably wouldn't really begin any meaningful tapering until we got into 2022. Given the improvements today and the prospects for today's economy, I actually think that there's a possibility they could start to taper or at least indicate that they're going to taper maybe even in this year, that might be sooner than broader markets expectations, but I certainly think that is possible. In terms of positioning, one interesting thing about today is that a very, very big component of the mortgage market is owned by the Fed and banks. So the convexity is selling that you might typically see in an interest rate selloff in other environments, when the mortgage market is owned by active hedgers or in the past it's been owned by GSEs and people stay on top of that. In the past, I think that effect could be less. In addition to that, I think that the Fed is -- has a memory of 2013 and when they sort of surprised the market back then with an announcement about tapering. And so I'm sure they have that in their minds. And the combined effect of sort of the Fed being more gentle about it and the lesser impact of convexity selling probably makes it a less violent episode, I would say. I mean it will still move the market certainly, but I don't think it will be as violent as we saw in 2013.

Great. With your earnings, you had an interesting slide talking about kind of spread levels today and what that could mean kind of 1 year forward in terms of spreads. I guess just for anyone that kind of didn't see that, can you just kind of talk through kind of your outlook for spreads? And I guess in that, just how spreads have been performing recently kind of with this move in rates?

Sure. I can do that one. So we do -- we did show a fourth quarter investor deck out there today. So if people have that in front of us, they can potentially look at it. I think the slide is on Page 14 of that deck. And what it shows -- there's 2 graphs there that are displayed. And the main results of the reference, like the answer first, is that a portfolio that includes MSR in its composition is expected to have more stable returns under various scenarios of mortgage spreads than a portfolio without MSR, right? And so what the graph shows is -- there's 2 graphs. Top graph shows 10 years of daily OAS data from JPMorgan DataQuery on an MBS index. Now on the x-axis is a starting spread -- a starting OAS. On the y-axis is the change in the spread 1 year hence, right? So all the other dots above the x-axis indicates spreads were wider and all the dots below the x-axis indicates that spreads were tighter 1 year forward. And what it shows is that when we made this chart on January 4, I think it was, we were at like 16 or 17 OAS, which is indicated by the dashed vertical blue line, right? Now what the graph shows is that in 100% of the cases over the last 10 years of daily data, which do include other historical periods where the Fed was involved in purchasing and other elements of quantitative easing, that in exactly 0 of those cases where the spread is the same or tighter 1 year forward, right? Now I'm not saying they're going to tighten them out. There's certainly lots of tailwinds here. Funding rates are very low. The Fed is continuing to buy. They say they're going to be here for a long time. The rolls are still somewhat special, but as Matt said, they've come off a lot. But if you have to say about the risk, the risk is to wider spreads and not tighter spreads in our opinion. Okay. Thank you, Doug. So on average, given -- during the course of the quarter, so far, we saw the current coupon tightened a lot so that the start of OAS went from 16 or 17 basis points, the vertical blue line, to probably 10 or lower. So all the way off the left-hand chart of that graph, right? Since then, they've widened out a little bit, and we're probably between 10 and 17 right now. But still, what it shows is that the last 10 years of data, on average, one would expect, if you believe in history, that OASs can widen 25 basis points or so on average. What the bottom chart shows is, if you assume for any given starting spread level that mortgage spreads instantaneously adjust to that average widening or tightening indicated in the graph above, and you create a portfolio with -- it includes assumptions about leverage and duration hedging and all of those things, and you created 2 portfolios, 1 that is the agency-only, agency swaps, which is indicated by the solid blue line; and one with our rough amount of MSR in the portfolio, as indicated by the black line. And if you start -- if you look at the point where we are at, the dashed blue line, the vertical line, at 16 or 17 OAS, what you find is that if spreads were to widen 25 basis points, which is the average amount shown in the top graph, we would expect an agency-only portfolio to earn 0% or minus 1%, as shown where the blue -- solid blue line intersects, the dashed blue line. In contrast, a portfolio like ours with, again, roughly our hedge ratios of MSR to RMBS portfolio is still expected to return in the high single digits, right? And in fact, this graph shows what the result would be for any starting spread. And you see that for large areas of starting spreads, a portfolio with MSR outperforms, except in areas where the spreads are -- the starting spreads are on the wider end of the range. And so what I'd like to say is, today, I think it's easy. Spreads are tight. Everyone believes spreads are rich. I don't know when they're going to tighten, but I believe they will one day. But even if you're uncertain about the direction of spreads, as I often am, I think that our strategy, where we keep mortgage spreads low, keep that exposure low, is preferable to one where the spreads are higher.

So clearly, it looks like the flatter black line, which is higher in a lot of scenarios and definitely flatter, would be preferable. I guess how do you think -- what are the trade-offs or the risks that you'd have to make in order to kind of achieve what looks like kind of a higher and more stable return profile?

There's certainly some liquidity issues. The MSR asset itself is not trading on a T+2 basis. You can't, in one phone call, move the risk around that easily, trades take longer [indiscernible]. On the other hand, the funding facilities are multiyear in nature, right, and not exposed to short-term repo dynamics either. It takes a lot of effort to create an MSR platform. I think many people know we have about 110 people who work at Two Harbors. We have a dedicated department of about 35 people that -- whose function is to oversee the job of our subservicers, to make sure that the payments are going where they're supposed to be going between the borrowers and the GSEs, to make sure that the tax escrow payments are being handled correctly. We have some people in our legal department, our mortgage compliance department that review every single borrower complaint that is made against each of our subservicers. So we have a flow program where we buy flow on a daily basis, managing that process, and we have 20 sellers in this program. Manning the data flow in and out requires some use of IT and so forth. So there's a lot that goes into it, and it requires a commitment to do that. We think it's worth it because it gives you the black line rather than the blue line. So we like that. But it's an effort to do it, and I appreciate that that's not for everyone.

Great. I guess just putting -- I guess, maybe just putting a little context around kind of the environment today. Is there any update you can give us as far as kind of how the portfolio has performed sort of quarter-to-date?

Yes. Maybe, Matt, you can do that one.

Sure. Yes, well, the quarter is far from over, obviously, and nothing is -- no results from -- through February have been finalized through our normal processes. But we're estimating that through last Friday, we're seeing total return in the 3% to 3.5% range.

Great. And I guess just if you could put a little more context around that. I mean it seems like January was a very strong month from kind of a spread tightening standpoint. The last week or 2 in February have seen spreads widen. Just to kind of put that black line kind of in the context, can you just -- any color you can give us around kind of how you've performed in each of those kind of 2 parts of the quarter to-date?

Well, I think you'll find that a good bit of the performance is coming from higher coupon performance and specified pool performance, especially in the January time period. I think servicing is performing in line with expectations, but that's going to -- that will have been one of the drivers so far.

I might add, Doug, maybe this is what you're getting at is that, in January, we had said in our earnings call that our performance at that time was, I think, up around 1%. So I think what we said at that time, if you compare it with other people who reported, so many agency-only REITs that reported during that period, they were up more than us. And so I think that is very consistent with what that graph shows, which is that we would underperform when spreads tighten, right? But that when spreads widen back out, I mean I don't know what other people are doing, but I would think that would be consistent with what we've seen.

Great. Then I guess, if you could then kind of given where spreads are, if they are on the tighter side today, can you just talk about where you see returns on kind of an incremental basis, whether that be agents -- kind of just the agency pools? Or where are you seeing kind of the paired strategy and the relative returns you're seeing today?

Sure. I'll take that one. In -- on the agency P&I side of things, even with the recent [indiscernible] that we've seen in the current coupon complex, I think spreads remain at -- really at multiyear tights. In February, I think it's tightened in and then sort of widened out to net to flattish, but that's a pretty tight level. So I think in the current coupon complex, I think we would cuff sort of gross static returns as being in the mid- to high single digits for both TBAs and pools. TBAs had an interesting funding advantage in Q3 and Q4, which was a great edge, and that was caused by the dynamics of the Fed buying in the front months and originators selling in the back months. At one point, that was probably 100 basis points through repo funding, which is obviously a huge advantage. The Fed in December started buying the sort of rolling and buying mortgages in the back months to match where sales were coming through, and that really dramatically reduced the funding advantage. I think it went down to something like 20 or 25 basis points through -- instead of 100. So it's far less interesting at a tight spread without funding specialness attached to it. I think we think that similarly, mostly because they're struck off of sort of a tighter spread, TBAs new specified pools in the current coupon complex are also going to be at the lower end of the mid- to high single-digit range. So we don't find that particularly interesting. I do think that seasoned higher coupons, we have a specified pool book that's higher coupons with some seasoning and loan balance in geography stores, I think those spreads are still attractive. Of course, that's difficult to add to because they're not making any more of those. But really, I think our main focus and what we find most interesting is adding new servicing through our flow arrangements. We're finding that to be the most interesting thing available. We've participated in a few bulk transactions, but those were a little opportunistic. And most of the time, I would say bulk is probably trading, I don't know, 200 or 300 basis points tighter than in the flow market. So we see those gross static returns is getting you into the low double digits when you pair new flow with RMBS. And like Bill was talking about a little bit earlier, our platform is quite a competitive advantage to us, right? Not everybody can do this. Like Bill said, there's lots of significant hurdles to entry into the business and the licensing and approvals and the operational aspect of it in infrastructure, and we're in a good position to take advantage of that today.

Great. I mean, I guess, you're mentioning kind of the TBA specialness being less attractive today. Do you see that kind of changing the mix of your holdings of agencies towards more pools? Or kind of how do you see that progressing over the next several quarters?

Yes. That's a good question. I mean we've reduced both recently actually. Our -- which has taken our mortgage spread risk exposure down to pretty low levels. I mean we're definitely cautious about it. The whole page that Bill described gave you a good idea about probably what's in our minds about the direction of spreads at tight levels. So I certainly -- as long as spreads are at tight levels like this, I wouldn't expect us to significantly increase our exposure. It probably takes wider spreads to get us more excited about adding more RMBS.

Let's -- just could you quantify? You mentioned kind of relatively low spread exposure. Can you -- any numbers you can put around that to -- kind of what's your sensitivity to kind of a spread widening and how that would compare to kind of a straight agency-only portfolio?