Two Harbors Investment Corp (TWO) 2017 Q2 法說會逐字稿

Good morning, my name is Bridget and I will be your conference facilitator. At this time, I would like to welcome everyone to Two Harbors' Second Quarter 2017 Financial Results Conference Call. (Operator Instructions) I would now like to turn the call over to Maggie Field with Investor Relations for Two Harbors.

Thank you, Bridget and good morning, everyone. Thank you for joining our call to discuss Two Harbors' Second Quarter 2017 Financial Results. With me on the call this morning are Tom Siering, our President and CEO; Brad Farrell, our CFO; and Bill Roth, our CIO. After my introductory comments, Tom will provide a brief recap of our quarterly results; Brad will highlight key items from our financials and Bill will review our portfolio performance. The press release and 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 at sec.gov. In our earnings release and slides, which are now posted in the Investor Relations section of our website, 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 on our website in the same location.

Before I turn the call over to Tom, 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 reflect our views regarding future events and are typically associated with the words such as anticipate, expect, estimate and believe or other such words. We caution investors not to rely unduly on forward-looking statements. They imply risks and uncertainties, and actual results may differ materially from expectations. We urge you to carefully consider the risks described in our filings with the SEC, which may be obtained on the SEC's website at sec.gov. We do not undertake any obligation to update or correct any forward-looking statements if later events prove them to be inaccurate.

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 that we issued last night. As reflected on Slide 3, during the quarter, we generated comprehensive income of $85.9 million or $0.25 per weighted average share. Core earnings were $0.28 per weighted average share. Additionally, we delivered a total return on book value of 2.2%. Our results were primarily driven by solid performance across our strategies as our team took advantage of attractive market opportunities. In our rates strategy, we remain committed to building our MSR position to pair with Agency RMBS, as we believe that this allows us to deliver better returns with less risk. Our credit strategy continues to be further enhanced by a strengthening economy and the strong fundamentals in the housing sector. We believe that there is significant opportunity to continue to capitalize on our valuable non-Agency RMBS position.

Our performance this quarter, again, demonstrates the benefit of our hybrid model and our ability to deliver strong returns for stockholders. The most notable highlight of the quarter was the formation of Granite Point Mortgage Trust as a standalone entity for our commercial real estate platform. Granite Point began trading on June 23rd on the New York Stock Exchange, under the ticker symbol GPMT. We believe that the creation of Granite Point will better reflect the embedded value in the commercial business that we built over the past 2.5 years.

As demonstrated through the formation of Granite Point, we have a strong commitment to maximizing shareholder value over the long term. Since our inception in October 2009, we have delivered a total stockholder return of 192%, which compares favorably to the Bloomberg REIT mortgage index return of a 124%. Our alpha generation over the past 8 years has been driven by our focus on dynamic capital allocation, superior asset selection and a sophisticated approach to risk management.

Additionally, we continue to use our balance sheet to benefit stockholders. Specifically, post quarter-end, we were able to successfully access the preferred equity markets. The underwriters exercised their full overallotment option, resulting in total offering proceeds net of offering costs of approximately $278 million. We intend to use the proceeds to invest in Agency RMBS, MSR and credit assets. Clearly in the current paradigm, the preferred market is a cheaper form of equity capital than what is available in the common stock market. Further, we believe this capital raise will be accretive to our earnings.

Turning to Slide 4, we believe that we are in an excellent position for the future. Our team continues to execute on our previously articulated plan for 2017, which is focused on running a more efficient business model while maximizing returns for our stockholders. Through close attention to asset selection, we opportunistically allocate capital to the most attractive investment opportunities across our Rates and Credit strategies. We continue to find ways to diversify our financing profile and manage our balance sheet and capital structure to optimize earnings and shareholder returns. Most importantly, we believe that our sophisticated and disciplined approach to risk management, of which MSR is a key component and a distinguishing factor, allows us to deliver strong results and book value stability through a variety of rate environments.

I will now turn it over to Brad, for a review of our financial results.

Thank you, Tom.

Please turn to Slide 5. Our book value at June 30, was $9.87 per share compared to $9.91 at March 31. During the quarter, we delivered comprehensive income of $85.9 million or $0.25 per share, which largely offset our dividend per common share of $0.26 and our series A preferred stock dividend.

As Tom noted, Two Harbors acquired 76.5% of the outstanding common stock of Granite Point in a formation transaction that closed concurrently with Granite Point's IPO. Due to our controlling ownership interest in Granite Point, we will consolidate Granite Point on our GAAP financial statements until such time as the shares received are distributed to our stockholders.

During the consolidation period, we will reflect a non-controlling interest on our financial statements for the portion of equity and net income not attributable to Two Harbors. In a moment, I will discuss additional GAAP and tax accounting details concerning Granite Point.

Before doing so, let's turn to Slide 6 to review our core earnings results for the quarter. Strong core earnings exceeded our run rate expectations and resulted in weighted average core EPS of $0.28.

Second quarter core earnings benefited from $1.6 million of higher net interest income due to higher average balances in our RMBS portfolio, partially offset by an increase in average borrowing rates and balances.

During the quarter, we held higher average debt-to-equity, which increased to 4.7x from 4.4x in the first quarter. We also realized a favorable difference of $5.3 million in swap expenses quarter-over-quarter due to an increase in short-term LIBOR and decreased notional swap amounts, as we preferred optional protection in the form of swaptions. Our other operating expense ratio increased slightly to 1.9% from 1.8%. The main driver of this increase was the recognition of $2.5 million in de-boarding and transfer fees, as we repositioned a sizable amount of our MSR portfolio across our sub-servicer network. We believe these transfers were important to our effectiveness in our servicing oversight role and will also result in lower costs of sub-servicing in future periods.

Turning back to the Granite Point transaction, let's turn to slide 7 to look at a pro forma balance sheet of Two Harbors at June 30. The intent of this pro forma is to illustrate Two Harbors' balance sheet as if the Granite Point entity was not consolidated and we had completed the distribution of the Granite Point shares to our shareholders by June 30. As you can see on the bottom right, our pro forma GAAP book value would have been $8.05. This lower book value per share will result in several important considerations to understand as a stockholder.

First, as we removed the lower-levered CRE portfolio of Granite Point from our balance sheet, our overall debt-to-equity will nominally increase as illustrated here by the 5.5x. Aside from that, we don't expect any material changes to our leverage metrics. Second, while we will have numerically lower core earnings, due to the lower capital base, we believe that our core earnings on a relative basis will be flat to accretive compared to the past few quarters. We are optimistic and believe that Two Harbors will benefit from a simpler business model across our Rates and Credit strategies. And going forward, we believe this will also allow us to maintain similar or modestly better dividend yields on book value.

Turning to Slide 8, let's briefly discuss the special dividend of Granite Point common stock. While these shares are currently subject to a 120-day lockup period, we have stated that it is our intention to distribute these shares in the form of a special dividend to our stockholders, separate from our cash distributions. For tax purposes, the fair market value of the special dividend will be the closing price of the Granite Point common stock on the day of distribution. For example, if the distribution had been made on June 30, the fair market value would have been $18.92 per share, multiplied by approximately 33.1 million shares or $625.7 million. Importantly, it is our intention to distribute enough cash dividends this year to effectuate a return of capital that will be approximately equal to the fair market value of the Granite Point stock distribution. This may result in a special cash dividend later in the year, although, we expect the amount to be relatively insignificant. With respect to the distribution ratio, if we use our June 30 shares outstanding of approximately $349 million, the distribution ratio would've been roughly 1 share of Granite Point for every 10 common shares of Two Harbors. Please note that this dividend remains subject to the approval of our Board of Directors and is subject to change.

Please turn to Slide 9. Our diversified and dynamic financing profile includes traditional repurchase agreements, FHLB advances, revolving credit facilities and convertible senior notes. We had $13.3 billion of outstanding repurchase agreements at June 30 with 25 counterparties. The outstanding repo balance includes the consolidation of the Granite Point financing facilities. We have not observed any disruptions as the repo markets continue to function efficiently for us. We continue to thoughtfully manage our FHLB capacity. Our FHLB advances totaled $3.2 billion at June 30 with a weighted average borrowing rate of 1.52%. Going forward, we anticipate that our balances may decline as we transition some Agency RMBS to repo financing as today's repo rates and terms are more favorable for financing these assets. As always, we are continuously monitoring and evaluating the best sources of financing; not only to be effective in our diversification and risk management, but to also achieve lower costs. We are financing MSR through both revolving credit facilities and the convertible debt issuance. At quarter end, we had $40 million of short-term borrowings under revolving credit facilities, with additional available capacity of $50 million. We have provided some additional details on our borrowings on appendix Slide 25.

I will now turn the call over to Bill for a portfolio update.

Thank you, Brad, and good morning, everyone. We delivered another quarter of strong performance across our Rates and Credit strategies. Our Credit strategy performed particularly well as we continued to capitalize on the opportunities available in legacy non-Agency RMBS. As we have noted before, the benefit of our hybrid model is the ability to allocate capital to the most attractive investment opportunities across our Rates and Credit strategies at any given time.

Looking at Slide 10, you will see both our portfolio composition at June 30 and our capital allocation quarter-over-quarter and year-over-year. As highlighted on the right-hand side of the slide, if the Granite Point shares had been distributed as of June 30, our capital allocation to Rates and Credit would have been approximately 65% and 35%, respectively.

As Tom mentioned earlier on the call, post quarter-end, we completed a preferred stock offering. We believe investment opportunities in our Rates and Credit strategies currently offer low to mid-double digit ROEs and, as of the end of July, we had deployed the majority of the proceeds into Agency RMBS, MSR and credit assets.

Moving to Slide 11, I'd like to go over a few of the drivers of our portfolio performance in the second quarter. Interest rates did not move substantially in the quarter, although the curve flattened and higher coupon agencies underperformed modestly. Non-agency spreads tightened positively benefiting our credit strategy and overall performance. Both our annualized portfolio yield and Rates yield were relatively unchanged from the first quarter. The addition of legacy non-Agencies at lower yields than our existing holdings was the main driver of the yield decrease from 9.2% to 8.5%, which also impacted our overall NIM slightly. In the first half of this year, we purchased about $400 million of legacy non-Agencies, bringing our total credit holdings to $2.3 billion. While these securities were purchased with conservative assumptions at baseline yields in the 5s, we believe there to be substantial additional upside. To highlight this, the bonds that we bought this year have already appreciated in value, bringing our realized return on those assets into the mid-teens on an annualized basis. We believe that these legacy non-Agencies will continue to have upside potential.

Please turn to Slide 12, as we discuss our rate strategy. Our objective is to provide strong returns through a variety of rate environments. A core component of this strategy is our investment in MSR. We believe that the combination of agency pools and MSR drives a higher return with less basis risk than agencies hedged only with swaps. The reason this is important is because it is widely anticipated that the Fed will reduce their reinvestments in agency RMBS, which will likely cause mortgage spreads to widen. MSR benefits from wider mortgage spreads thus mitigating the potential impact to our book value. MSR is a significant benefit to our portfolio in this scenario as compared to an agency-only strategy.

In terms of portfolio activity in the rate strategy, we focused on increasing our prepay protected pool holdings from 72% up to 83%. On the MSR front, we added $5.8 billion in UPB of MSR from flow-sale arrangements during the quarter. We also closed on to 2 bulk purchases totaling $13.1 billion UPB. It is our expectation that near-term flow-sale MSR volume will run about $2 billion per month and we intend to continue to grow MSR as a component of our Rates strategy going forward.

Let's move to slide 13 to discuss our Credit strategy. Residential credit performed very well in the second quarter and our strategy continues to benefit from fundamental improvement in the housing market and positive market technicals. Housing prices are up 6.7% on a rolling 12-month basis, which is supported by the tailwinds of affordability, low housing supply and strong demand. These tailwinds are quite meaningful and we believe future residential credit performance will continue to be strong, driven by increasing prepayments, lower delinquencies, defaults and severities. To benefit from this, we have positioned our portfolio primarily into deeply discounted legacy assets due to the yield pickup these can realize from improving performance. To emphasize this point, we continue to realize both better credit performance and faster prepayments than our initial expectations. And as a result of this, in the quarter we released another $13.9 million in credit reserves against this portfolio, bringing the total amount of credit reserves released over the past 12 months to $38.2 million. We still see opportunity to realize strong returns in this sector and have used some of the proceeds from the preferred stock offering in the third quarter to purchase similar bonds with upside performance potential.

In closing, our investment strategy going forward remains rooted in our core competencies of understanding and managing interest rate, mortgage prepayment and credit risks. With that in mind, we will continue to opportunistically allocate capital to the most attractive investment opportunities across our Rates and Credit strategies, while maintaining our sophisticated and disciplined approach to risk management.

With that, I will turn the call back over to Bridget for Q&A.

(Operator Instructions) Our first question comes line of Doug Harter with Credit Suisse.

Could you talk about the outlook for credit? I know you said you continue to see opportunistic but, I guess, with the absence -- or with commercial being split off, do you think, you'll be able to sort of consistently allocate capital there and sort of maintain the allocation? And are there any other kind of newer credit investments that you're looking at?

Yes, that's a great question. I think there's a couple of things I would point out. First of all, as you might recall from past discussions around our non-Agency position, the average life of our holdings somewhere in that 7 to 8 years. So we expect the portfolio to have quite a long average life, which obviously we're excited about because we think we can benefit from the tailwinds that we discussed on the call for this portfolio. In addition to that, there are -- there's numerous other sectors that we are looking at currently and also that we believe, over time, will have value to us. I think CRT is going to continue to grow. We have not been excited about those spreads lately. And so, we're not really focused on that, but as the non-QM and NPL securitization and other securitization of private loans continues, we're pretty confident that we're going to be able to take advantage of that over time.

Yes. Doug, It's Tom. And if I could just add to that, as Bill mentioned, we have a quite an extended life of our non-Agency book. And if that book were to shorten up it would, in my estimation, only be because prepayment speeds continue to accelerate, which, in some respect, is a problem, but a very high-class problem as it would result in higher book value and obviously more money to spend for our shareholders. So if I have to have problems, I'd rather have the high-class type but, as Bill mentioned, we think our credit book will be in place for quite a while going forward.

And our next question comes from Bose George with KBW.

In terms of the incremental return opportunities, can you just talk about how rate compares to the returns you can get now in credit?

It's Bill, Bose. Yes, so they're quite different profiles but we're really actually pretty excited about both of them. On the rate side, Agency is paired with MSR and we're -- I think we mentioned about $2 billion a month of -- on a flow basis in terms of what we're getting, that's in the low to mid-teens -- low to mid-double digits, excuse me. And so that obviously is also attractive because with the Fed poised to reduce their balance sheet, to the extent that agency spreads widen, MSR is going to provide a very nice buffer against widening agency spreads, at which point NIMs will be more attractive but, from a book value standpoint, we'll be much better protected than, say, an agency-only REIT. And then on the credit side, it's a very different profile because, while you're going in, yield on these legacy assets is not as high as it was, say, 5, 6 years ago, there's still a tremendous amount of upside. I talked about on the call that we bought bonds with yields in the 5s earlier this year, but it's not even been 6 months and our realized return has already been in the mid-teens on an annualized basis. So what you're looking for there is sort of a high single-digit baseline with no upside kind of assumption. But where your realized return, which obviously would be good for book value, can be much higher and, in fact, that's been our experience is that we have realized substantial upside from these. Hopefully that answers your question.

Yes. That does, thanks. And then actually just switching over to servicing, I think, last quarter there was discussion about the funding side of MSR's improving. Just wanted to get an update there. Are you're getting closer to seeing asset level funding for the MSRs?

Thank you for that follow-up question. I don't know -- we don't have tons to update. We are, as we mentioned previously, speaking with several counterparties. We do think the market is creating structures that are going to provide the liquidity and value protections that we are looking for in financing that asset class. I'd say, the Ginnie Mae market is probably further along than the conventional. But we're still very optimistic that we'll have something in the works and, optimistically, done by the end of the year to establish more leverage in that asset class.

That was Brad, this is Tom, Bose. We continue to think this is a significant area of opportunity for the company.

Okay great. Because let me just ask one more, just kind of a broader, not a market question. I just wanted to get your thoughts on the incident in June where Wells Fargo -- the trustee withheld proceeds for some of those called bonds, just wanted to see, if you thought there was any read-through that we should think about?

Well, yes, that's a challenging one because it certainly was unprecedented and, as we know, there's already litigation that's been put in place. So, I don't know that we have a real strong opinion on how that's going to play out. With regard to our situation or our portfolio, this isn't really something that we're that -- we're not concerned about it at all, actually, because the bonds that they're focused on are in deals that are callable and there might be advantage to the owner of the call rights to exercise that call. So those are typically much cleaner, prime or potentially high quality Alt-A deals. As you know, almost all of our portfolio is in deeply discounted subprime bonds. The average delinquency in the deals that we have is in the low '20s, so given that the bonds are at deep discounts and the delinquencies, this isn't really something we would anticipate impacting us at all, but it is certainly going to be interesting to watch to see how it plays out.

And our next question comes from the line of Steve Delaney with JMP Securities.

Looking at the pro forma leverage and equity allocation that you provided, 5.5x, 6 -- debt-to-equity and 65% allocation to Rates, can you comment on how you see those within sort of the spectrum of where leverage or the equity allocation might go? It would certainly look, just on the surface, that those percentages are higher, both in terms of leverage and allocation, to the rate strategy than we've seen over the last several years.

Yes. Keep in mind that the -- we did use some leverage from the convertible bond to benefit the MSR and as you know there is certainly the possibility for more financing available for MSR. In terms of the allocation, the leverage that is on our balance sheet as of June 30 obviously reflects sort of an unusual period of time with the Granite Point and the IPO. But on a going forward basis, I think the rate strategy -- if you look back in time, a rate strategy leverage or agency strategy is typically somewhere in the mid- to high single digits and credit strategy 1.5x to 2.5x. So on a blended basis, we look reasonably comfortable with that and we would expect to be somewhere in that range as we move forward.

That's helpful.

Steve. I think, if you did a decomposition of existing Two Harbors and pro forma of Two Harbors, the leverage within the pro forma book really hasn't changed. What it's really largely due to, the nominal higher leverage, is the off-boarding into Granite Point of the commercial real estate assets, which of course carry with them lower -- [less] than agency. So if you look at our overall leverage on the pro forma book, it really hasn't changed about meaningfully. There's bit of noise in there but, from a thematic standpoint, it really hasn't changed.

Yes. Steve. I want to add one more thing, which I actually think is really something worth pointing out and that is if you have an agency -- let's just talk about the Rates for a minute, if you have agency-only strategy, in other words, just agencies and swaps or Treasuries, whatever your preferred hedge mechanism is, based on whatever leverage you run, that actually generates certain amount of exposure to wider spreads. When you couple agencies with MSR, and I think we talked about this before, we talked about this on our webinar, you can actually have, using the same leverage on agency pools combined with MSR, you actually can have a higher return with less spread exposure, which actually implies that, if you use a little bit higher leverage on agency, you can still have less risk from spread exposure. So I think that's important to also add, the benefit that the MSR adds to the Rates strategy is the ability to be comfortable with comparable leverage or even higher leverage and still have less risk.

And Bill, that's not always, I think, intuitive in terms of the MSRs and the spread issue. And I assume, the way -- where we get there in terms of benefit is that obviously that spread widening functions into the effective cost of borrower refinancing, and so you did get a -- wider spreads or higher rates are both going to, I guess, make the MSR have a longer, slower CPR.

Right, well. I mean, just to give you an idea, so our MSR at June 30 roughly hedges about $4 billion of pools. So If you think about our total pool position against Rates alone, it's obviously much, much lower. It's lower by about $4 billion. So the pools in MSR are relatively stable profile as relates to spread and rate.

Yes. one more quick -- thank for that. One more quick comment, just in terms of your structure of your capitalization of the company. You had previously done a convert, I guess, in the first quarter, and now this larger preferred issue along with the original preferred, it looks like maybe $700 million of long-term and -- or investable capital in addition to your common equity base. Looking at that kind of pro forma, it looks like that $700 million would represent 24% or so of common equity after the spin out. Is there -- when you look at your balance sheet and your sources of capital, does that 24% give you some incremental space for additional notes or converts or preferred from sort of where you will be standing in November? And that's my last question.

Well, Steve, you always ask great questions, so ask as many as you'd like. You really should think about the convertible bond offering and the preferred differently, we certainly do. So with respect to the convert, we really think of that as financing for MSR. And whether or not we would revisit those waters will probably we largely due to the outcome of our goal of achieving discrete MSR financing through 1 or 2 different methods. We're fairly far down the path in that goal, but obviously there's more to be reported in that regard. As far as the preferred stock goes, you should -- the way we think of that is simply a cheaper form of equity capital than what's available in the common stock market. And so one of our primary goals of the company is to use our balance sheet and capital structure in a smart way to optimize shareholder returns and so whether we tinker with convert or preferred offerings going forward, it's just going to depend upon a lot of things including market conditions obviously being the primary one.

And our next question comes the line of Rick Shane with JPMorgan.

A lot of this has been touched on, but I'd love to sort of think about this both in terms of the short and long-term, in terms of capital allocation. You show basically 2/3 to Rates, 1/3 to Credit. I am curious, with the new capital that came in from the preferred, were you roughly in line with that mix? And then longer-term, what are sort of the brackets around that so we can think about how you will approach the market opportunistically over the long-term?

So as I mentioned on the call, we had done a pretty good job putting the capital to work from the most recent preferred offerings as of the end of July. And I think as you should expect, probably that Rates probably got a little more than it's -- the share that we showed on Slide 10. But keeping in mind that we're trying to be much more opportunistic on the credit side. Obviously, pools you can buy with a phone call and MSR we're getting every month. But we do anticipate that credit will get a fair share of that. Longer-term, and as you know, you been following us for -- over the years, is we really try to be opportunistic with where the market presents attractive investment opportunities for us. So I think it's highly unlikely that we'll be 100% one way or the other and we think it's -- having a credit exposure when credit is attractive makes complete sense to us and it also helps diversify and stabilize the overall portfolio returns. But to try and put numbers on it when we're not sure what the market is going to give us is fairly challenging. I think in the short run, in other words, the next, whatever, 12 to 18 months, it's probably going to look more like what it does today than anything significantly different.

Yes. Rick, I mean, just to add to that. I mean, the way we think about it is this is how we see the world today. But we don't want to be committed to doing this thing or that things, we just want to be committed to doing the best things for our shareholders. So this is how we see the world today but, as Bill says, the future is not promised to any of us. We just want to do smart things with good risk return profiles.

Fair enough. And again, I mean, I think that's what we're trying to figure out which is that, with spreads wide on agency and pretty tight on Credit, is this sort of 2/3, 1/3 towards the boundary of where you want to be or is the boundary 75%, 25% in the current environment and gravitating back to 60%, 40% over time.

We just -- we don't know is the honest answer. We think that a runway for MSR is quite encouraging and we think that's the significant opportunity for us going forward. We think that's a distinguishing factor, but the markets aren't promised to any of us and so what that mix is in 2018 or '19, it's very difficult for us to say and we don't want to be disingenuous. We just want to be committed to doing the smartest things as we can.

No. Look guys, I really -- I appreciate the honesty of that answer and sometimes the best answer is I don't know, so I appreciate that. One question, I just wanted to make sure I understood this. You talked on the credit securities that you purchased earlier this year about realized gains. Did you actually enter and exit those trades? Or I just wanted to make sure I understood that precisely?

What Bill said -- again, this is Tom. What Bill said was that non-Agency that we had bought earlier in the year had appreciated in value and we didn't make any comment on realizing gains on those. It's unlike us to whip things around or trade things around, unless we think they've really achieved fair value or exceeded fair value. We still think, the book is cheap...

Your comment had surprised me because it -- I thought I'd heard realized in it, it didn't. It seemed little incongruous with the strategy all along?

Yes. Just to be clear, the -- just -- I think you're interpreting realized to mean that we sold them all and that was not the intent of that remark. In other words, we bought them and they have been marked up higher and, as a result of that, the annualized return was in the mid-teens. But that -- you should not interpret that to mean that we sold them all and took those gains from that process.

Yes. I think, perhaps, you're reading the word realized too literally in that context.

And our next question is from Jessica Levi-Ribner with FBR.

Most of have been asked and answered, but can you talk a little bit about the new sub-servicing arrangements? You mentioned that it should result in a lower cost of sub-servicing. How can we think about that in our models?

Yes. Thank you for that question. We have some -- we have a sizable sub-servicing network, which allows us a variety of benefits. It allows us to establish meaningful relationships with the broad seller network. It also allows us to reposition the portfolio where we think either we're going to feel like we get the best service as an overseer of MSR and also pursue opportunities where we think we can save money over the long term. In this case, we felt we actually could achieve all of those things by shifting a sizable amount of our portfolio across our subservicer network. As we noted it, we recognized about $2.5 million of the boarding fees or transactional fees with that. And we generally estimate that, that recovery will be between 12 to 15 months and will obviously generate additional earnings beyond that. So I think we will see that from time to time. We don't see it, the size of this, happening very often but, again as I said, we want to make sure our seller network is broad. We want to make sure that we're getting the best servicing performance to protect our MSR and then where we see opportunities, we're willing to take that de-boarding cost and remove servicing to benefit our future cost as well as make sure we get the best performance. That's kind of generally how we think about it and that's kind of a summary of that 12- to 15-month payback period for that cost.

And just what's the -- what's your desired allocation to MSRs? I know you made the comments that there's going to be roughly $2 billion per month on a flow basis, but is there a cap for you guys with MSRs?

Well, I kind of go back to what Tom said earlier, which is we're trying to put our money into what we think is the most attractive holding that will drive our returns for our stockholders at any given point in time. So in -- historically, we've talked about 20% to 30% of capital, which, given that we have now got some leverage available, albeit it's not a substantial amount of leverage, which we -- you wouldn't expect, necessarily, on that asset class. That only allows us certainly to continue to grow our MSR portfolio quite nicely for the foreseeable future. Now the other thing is, if you think about the returns I talked about, sort of lower to mid-double digit, to us that's as attractive as anything that we're seeing. So it's certainly the best dollar that we can put to use today. So I would say for the foreseeable future you can expect us to continue the path that we're on. I hope that kind of tells you way we're thinking about, I don't think we think about it in terms of a specific dollar amount or equity capital amount.

Yes, I mean, the near-term a feeling isn't proximate. So we anticipate continuing to grow the book going forward and how big it gets is going to be a function of market conditions, again, and how they pair with agencies, et cetera. We think the run rate for MSR is great and we -- you should reasonably expect that book to continue to grow, certainly through flow agreements, potentially through future bulk deals. But that is the path that we're on and we see no reason to avert from that path.

And our next question is from the line of Fred Small with Compass Point.

Sorry, my question got answered, thanks.

I'm not showing any further questions, so I'll now turn the call back over to Mr. Siering for closing remarks.

Thank you, Bridget, Thank you, everyone, for joining our second quarter conference call today. We are pleased with the progress we have made on our plan for 2017 and look forward to speaking with you again soon. Have a wonderful day.

Ladies and gentlemen, this does conclude the program. You may now disconnect.