Two Harbors Investment Corp. (TWO) Q4 2014 Earnings Report, Transcript and Summary

Good day, ladies and gentlemen, and welcome to the Two Harbors Investment Corp. Fourth Quarter 2014 Financial Results Conference Call. [Operator Instructions] As a reminder, today's conference is being recorded. I'd now like to turn the conference over to your host for today, Ms. July Hugen, Director of Investor Relations. Ma'am, you may begin.

Thank you, Ben, and good morning. Welcome to our Fourth Quarter 2014 Financial Results Conference Call. With me this morning are Tom Siering, President and Chief Executive Officer; Brad Farrell, Chief Financial Officer; and Bill Roth, Chief Investment Officer. After my introductory comments, Tom will provide a recap of our fourth quarter 2014 results, Brad will highlight some key items from our financials, and Bill will review our portfolio performance. The press release and financial tables associated with today's conference call were filed yesterday with the SEC. If you do not have a copy, you may find them on our website and the SEC's website. This call is being broadcast live over the Internet and may be accessed on our website in the Investor Relations section under the Events & Presentations link. We encourage you to reference the accompanying presentation to this call, which can also be found on our website. Reconciliation of non-GAAP financial measures to GAAP can also be found in the presentation. We wish to remind you that remarks made by management during this conference call and the supporting slide presentation may include forward-looking statements. Forward-looking statements reflect our views regarding future events and are typically associated with the use of words such as anticipate, target, expect, estimate, believe, assume, project and should or other similar 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 www.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, July. Good morning, everyone, and thank you for joining us today. Turning to Slide 3 in our presentation, I'd like to review our full year and fourth quarter financial results. 2014 was an excellent year for us, despite a volatile interest rate environment. We generated a return on book value of 15% for the year. This metric is a testament to Bill and the investment team's excellent portfolio management and risk mitigation strategy. This is particularly impressive given our low levels of leverage. Our book value at the December 31 was $11.10 per share, representing a quarterly return on book value of 1% when combined with the fourth quarter dividend of $0.26 per share. In 2014, we delivered comprehensive income of $578.2 million, representing a return on average equity of 14.4%. During the quarter, we had comprehensive income of $42.2 million, or $0.12 per weighted average diluted share. For the quarter ending December 31, we reported GAAP in core earnings of negative $0.10 and positive $0.23 per share, respectively. Let's flip to Slide 4. 2014 was a very successful year with regards to the development of our operational businesses. We meaningfully increased the number of sellers in our network. We completed 3 securitizations in the second half of 2014 and our pipeline volumes are strong. Our MSR business continues to expand as we aim to create long-lasting flow relationships with high-quality sellers, leveraging off of relationships from our conduit business. In the fourth quarter, we were pleased to announce our expansion into commercial real estate with the addition of a seasoned commercial real estate team. We intend to initially allocate $500 million in equity capital to this initiative, further diversifying our portfolio. The commercial real estate loan market is over $3 trillion in size and over $1.5 trillion in loans are expected to mature in the next several years. This business certainly meets the criteria we desire in respect of opportunity, scale and investment runway. We will keep you informed as this develops. Please turn to Slide 5. Let's briefly discuss the macroeconomic and political environment. As expected, the Fed made its final reduction in asset purchases in December 2014, but plans to continue to make significant reinvestments of MBS paydowns for the foreseeable future. In 2014, the 10-year interest rate fell from 3.3% to 2.2%, near the lows of 2013, and the curve continued to flatten in the fourth quarter. Employment, consumer tailwinds and HPA are important indicators of a healthy housing market. Unemployment metrics improved throughout 2014, falling from 6.6% in January to 5.6% in December. Additionally, national home prices increased 5.5% on a rolling 12-month basis as of November 30, according to CoreLogic. In fall 2014, the FHFA released a proposed rule-making change regarding membership in the Federal home loan bank system, which if adopted, could have an impact on our membership. We consider ourselves members in good standing with the Home Loan Des Moines, and we believe our interest rate -- our interest align directly with the mission of a home loan system. Additionally, our secured financings are consistent with our safety and soundness mantra. We submitted a common letter to FHFA in response to the proposed rule-making and a copy can be found under Executive Insights in the Investor Relations section of our website. We strive to be a thought leader within the industry and have increased our engagement on regulatory and policy-related topics that directly impact our business, including conforming loan limits, guarantee fees in the private label securities market. 2014 was a transformational year for Two Harbors, with respect to our operating -- operational businesses. We continue to broaden our business model to include investments in attractive sectors of the mortgage market, including our planned expansion into commercial mortgage loans. In 2014, we also celebrated our fifth year in operation and are pleased to have returned a 125% to stockholders over that time period with respect to share price and dividends, compared to the Pine River Mortgage REIT Index of 57% over the same period. With that, I will turn the call over to Brad for a discussion of our financial results.

Thank you, Tom, and good morning. Please turn to Slide 7. Book value ended the year at $11.10 per share versus $11.25 as of September 30, and $10.56 at the beginning of 2014. Book value is driven by comprehensive income of $42.2 million, or $0.12 per weighted share in the fourth quarter, and $578.2 million, or $1.58 per weighted share for the full fiscal year. Our full-year valuation gains in both our rates and credit strategy were marginally reduced due to interest rate volatility and wider credit spreads in the quarter. The company declared dividends of $0.26 and $1.04 for the fourth quarter and full year, respectively. Please turn to Slide 8 for a summary of our financial results. Core earnings of $0.23 per weighted share represented an annualized return on average equity of 8.1% and were in line with the prior quarter and our expectations. While this was a relatively consistent quarter for core earnings, I will comment on a few of its underlying drivers. First, our implied debt-to-equity ratio, including our TBA position, remain low at 3.0x, but modestly increased in the last half of the quarter from 2.7x at September 30. Swap expense was higher as a result of the hedge position we held for the majority of the fourth quarter. This should trend lower next quarter as we repositioned our overall hedges, which included a reduction in our notional swaps in the early December. Our expense ratio, which dipped 30 basis points last quarter to 1.2%, returned to the 1.5% level that we ran in the first 2 quarters of 2014. This falls within our expectations due to the variability of cost associated with supporting the operational businesses. Next, I would like to discuss our taxable income, dividend distributions and 1099 treatment of those distributions, all of which are highlighted on Slide 9. During 2014, the regenerated tax earnings of $388 million, which included taxable realized gains of approximately $73 million on the sale of certain RMBS securities. As a reminder, we continue to hold unrealized gains of nearly $900 million in the RMBS portfolio as of year end. For tax planning and optimization purposes, we were able to utilize a capital loss carried forward from 2013 of $50 million, which reduced 2014 REIT taxable income to $338 million. Since our 2014 dividend declarations of $379 million exceeded taxable income, but did not exceed 2014 tax earnings, our dividends will be treated as taxable to stockholders rather than a return of capital on the Form 1099. Please reference the dividend information on the Investor Relations section of our website for additional information regarding the distributions and their tax treatments. Please turn to Slide 10 for an overview of our financing profile. Our repo capacity continues to function in a normal manner with no meaningful shifts in financing haircuts or rates. We maintain a diversified counterparty mix, including 25 active counterparties at year end. The maturity profile of our repo contracts was an average of 64 days to maturity at December 31, down from 100 days at September 30, as a sizable portion of our maturities were previously rolled into January. As of December 31, we have secured advances with the FHLB of $2.5 billion, with a weighted average maturity of approximately 119 months. Our weighted average maturity increased from 44 months at September 30, as we added 20-year borrowings at competitive floating rates. The FHLB facility remain just one of the many tools we use to diversify and optimize our funding mix. For more information on our repurchase agreements and FHLB financing, please see the appendix on Slide 26. Now let me turn the call over to Bill.

Good morning, everyone. We are proud to have delivered stockholders a 15% total return on book value in 2014, particularly given our posture of relatively low leverage and little interest rate exposure throughout the year. Our fourth quarter return of 1% was notable in a difficult environment. Our book value performance for the quarter was impacted by several factors, including hedging losses due to heightened interest rate volatility, particularly in October, underperformance of specified tools and wider mortgage credit spread. Please turn to Slide 11 to discuss our portfolio composition. As of December 31, our portfolio was $16 billion in assets, including $11.9 billion in rates, representing 56% of capital, and $4.1 billion or 44% of capital in credit. With respect to our rate strategy, capital allocation to Agency RMBS and MSR remain flat quarter-over-quarter. However, we opportunistically added approximately $1 billion of lower loan balance pools, increasing our leverage slightly. In general, we continue to keep our basis risk exposure and leverage relatively low. Our credit portfolio remains predominantly weighted towards legacy non-Agency RMBS, with a focus on lower-priced sub-prime bonds with upside to better performance and prepays. That being said, the capital allocation to these assets declined from 38% a year ago to 31% at year-end, as opportunities to add in the legacy space were limited throughout the year and we continued to sell certain bonds that we believe were fully value. While we decreased our capital allocation to legacy non-Agencies, we increased our allocation to new issue mortgage credit through our conduit activities as well as by purchases in the market. Our capital allocation to conduit activity increased from virtually nil at December 31, 2013 to over 9% at year-end. Also during the quarter, we added some GSE credit risk-sharing bonds, as spreads on these bonds widened dramatically and we view the risk return profile is favorable at these wider spreads. Please turn to Slide 12, as we review our portfolio performance and yield. Our rate strategy yield increased 10 basis points quarter-over-quarter to 3.7%. Within this strategy, Agency yields increased 10 basis points to 3.4%, and MSR yield increased 20 basis points to 9.1%. Prepays on Agency decreased modestly in the quarter. On the credit side, yields on legacy non-Agencies decreased slightly, while yields on new issue non-Agencies increased by 30 basis points. Our net economic interest in securitization yield also increased 30 basis points as subs and IOs grew as a percentage of our retained interest upon the completion of Agate Bay 2014-3. Non-Agency prepayments were stable quarter-over-quarter. As you can see on the bottom right of this slide, our annualized net interest rate spread was 2.9%, down modestly from the third quarter. For more on our rates and credit strategies, please refer to the appendix, Slides 21 through 25. Moving to Slide 13, there are a few items I would like to highlight with respect to hedging. During the quarter, we adjusted our hedges, reducing the net notional of our swaps by $6.9 billion and adding optional protection via swaptions totaling $4.8 billion. We believe this change positions us better for the Fed moving interest rates higher on a more measured basis, but also covers us for a dramatic selloff in rates, such as we saw in 2013. Further, the optional nature of many of our hedges helps protect book value in the case of a continued drop in rates. Consistent with our typical posture and as you can see in the table on this slide, we still maintain a fair -- fairly low overall exposure to rising rates. See Slides 27 and 28 in the appendix for more information on our hedges. Please turn to Slide 14, where we will spend some time talking about the conduit. 2014 was a banner year for our mortgage loan conduit, as we made great strides adding seller partners, which helped drive loan volumes. We completed 3 securitizations throughout the year, representing approximately $1 billion of mortgage loans. Our total net economic interest in securitization trusts by market value increased by approximately $45 million quarter-over-quarter to $535 million at December 31. Our prime jumbo pipeline, which includes loans and interest rate loss commitments was approximately $1 billion at year-end. Our recent origination run rate has been roughly $300 million per month, putting us on track to have substantially more volume and complete more securitizations this year than in 2014, obviously, subject to market condition. We ended the year with 33 approved sellers, and we anticipate adding an additional 15 to 20 this year. We believe this high-quality network of sellers will provide a rich source of opportunity across a variety of products over time. From a financing perspective, the FHLB facility has been beneficial to our conduit program, as we are able to offer attractive pricing on a consistent basis to our seller partners. This is obviously helpful for homeowners who fall outside government programs, as attractive rates allow them to afford a home. Please turn to Slide 15. As we noted on our third quarter earnings call, we launched high LTV and non-Prime programs during the quarter. Both of these programs are in their nascency, although we have begun to lock loans with LTVs between 80% to 90%. As a reminder, this program is focused on high credit quality borrowers who prefer or require a lower down payment. The non-Prime program is geared towards borrowers of average credit quality who have the financial wherewithal to buy a home, but haven't been able to get a mortgage due to exceptionally tight credit standards. As we noted on our last call, we expect it will take some time to drive volumes in these products, and we look forward to growing these programs in 2015 and beyond. Let's now discuss mortgage servicing. The regulatory landscape from mortgage servicing and servicers remains dynamic, and we expect that this could result in large volume transfers in the next few years. We believe we are well-positioned to add MSR over time in both bulk and flow transactions and aim to add more flow sellers throughout 2015. We remain committed to growing our MSR program over time as it provides multifaceted benefits across our portfolio. We are also excited about the opportunity to further diversify our portfolio into commercial real estate debt. Given the large market size and attractive returns, we feel this opportunity will help drive stockholder value. We will have more to say about this as we go through the year. Turning to recent market events. In January, the 10-year treasury rallied about 50 basis points and mortgages widened on fears of faster prepayments due to lower rates. Given our overall hedge profile and substantial holdings of prepayment-protected pools, we are pleased that our book value is up slightly through January. As always, with these types of remarks, I would caution that we are only 1 month into the quarter. In closing, 2014 was an excellent year for Two Harbors. We delivered a strong return for our stockholders, while at the same time, carrying a conservative risk profile with respect to leverage and rate exposure. We solidify the operational aspect of our business through the growth of our conduit and MSR initiatives, and also laid the foundation to invest in commercial mortgage loans. I will now turn the call back to Ben to take questions.

[Operator Instructions] And our first question comes from the line of Douglas Harter of Crédit Suisse.

You had mentioned the continued attractiveness of the MSRs. Have you guys been able to add any additional flow deals over and above the PHH deal that you have?

Thanks for joining us. Basically, we haven't made any announcements in any additional flow sellers at this point. We are focused primarily on increasing our sellers on the prime jumbo side. But I will tell you that one of our goals for this year is to add flow sellers on the MSR side, many of which that could come from our current sellers that we have on the prime jumbo side, which, as I mentioned, are 33 at year-end. So that's something that we think will be beneficial over time for the MSR initiative.

And then as you're looking at kind of the flow versus bulk opportunity, I guess, how do you kind of weigh the opportunities there and the attractiveness of those opportunities?

Well, I mean bulk is something that kind of shows up when somebody decides to sell something. And so that's not unlike in the securities market, if there's securities out for bid. So we're always examining bulk offers or opportunities that are presented to us. The flow situation's a little bit different, because as you know, there are many originators that sell flow as part of their cash flow management, so it's a little bit different dynamic. And being positioned on the flow side, we think provides a more stable flow of MSR to us over time. So basically, we're happy and willing to look at both and they're just a little bit different from a dynamic standpoint.

Got it. And then shifting to the sort of the more of the MBS side, you guys have been running with lower leverage for a while. We've sort of entered a more volatile time. Is this -- have the opportunities improved enough with this volatility to sort of make you want to increase leverage more meaningfully? Or are we not quite there yet?

Yes. That's a great question. Yes, I mean, we saw mortgages -- Agency mortgages anyway and we also widened in January quite a bit, and we've seen a little bit wider mortgage credit spreads. Bottom line is, the Agency ROEs, while they've improved slightly, they're still in the single digits. We did see, as I mentioned, some opportunity on the credit risk sharing with the wider spreads, but there is not that -- this -- basically, the best dollar we can spend today, frankly, is -- comes out of the conduit. We're triple laser trading in the market today, that's roughly 10% ROE hedged, and subs and IO are better than that. So that's obviously something we're very focused on. But we wouldn't intend to take our leverage up on the Agency side unless we saw a dramatically higher expected ROE.

Our next question comes from the line of Trevor Cranston of JMP Securities.

I guess one more question on the MSR portfolio. Presumably, as rates have dropped here in the fourth quarter and again into the first quarter, the negative duration of that portfolio has increased. And I guess it seems like it might make sense as that's happening to change around the swap books some, which obviously happened in the fourth quarter. Can you talk about any changes you might have made to the hedge book to date in the first quarter?

Thanks for the kind remarks. I think that -- well, first let me just make a couple of comments on MSR, right? If you look at the WACC on the -- our portfolio, it's in the high 3s. So clearly, it hasn't been in the money most of last year and it's close to at the money currently. So the duration of that obviously changes as rates change. We manage our hedge position for the whole -- our entire book, and as you know, we manage it pretty actively. So if you look -- I think it's instructive, if you look at Slide 27 and 28, and you'll see most interestingly, I think, to note on Slide 28 that our swaptions, our payers are now at roughly 8 billion and our -- we actually have about 5 billion of receivers. So what we're doing is, we're -- and we took our swaps down, right? So we increased our swaptions on the both payer and receiver side, and we took our pay fixed swaps down. So without getting too math-y on this call, I think the point is, is that what we're trying to do is create a profile where we're protected from various movements in the curve and rates. The perfect example, as I mentioned on the call, our book value is up slightly in January, even though we saw a big rally in the 10-year, flatter curve and at the same time, wider mortgage spread. So that's -- I think the important point is just to know that we're very active in moving our hedges around as the market changes based on what our exposures are.

Trevor, it's Tom. I just -- I think the investment team did a marvelous job in respect of their timing, switching to more optional protection, because obviously, we've had a heck of a rate move in relative terms and interest rates. Our timing, I think, was very good. And I think we're very well-positioned for shifts, up or down, in interest rates right now.

Got it. That makes sense. And then on the commercial real estate side, can you maybe give a little bit color kind of on what you're thinking in terms of the initial focus in terms of what kind of property types you might want to focus on or avoid, and kind of what sort of geographies you might be looking to exploit or stay away from?

Yes, sure. Let me just take a minute and give you sort of a very high level overview. First, as of the end of the year, we hadn't invest -- made any investments yet. I think the deployment will probably largely be back-ended in 2015. As was mentioned on the call, it's obviously a very large market with a lot of volumes expected over the next several years. So we're going to be focused on loans in the U.S. and property types that we consider could be office, multi-family, retail, industrial, hospitality, or self-storage, we said that we're going to initially allocate $500 million in equity capital. And our target assets on the loan side could be senior loans, mezzanine loans with a typical size. Our B notes and preferred equity, pretty much any part of the capital stack, specifically debt-oriented. You can probably expect typical loan size to range in the $10 million to $100 million size. Most loan terms are sort of in the 3- to 10-year range and typically float off a LIBOR.

And just like our other credit allocations, we're very MSA-focused, too, Trevor. Obviously, market conditions are quite good different depending upon what city or state you're talking about.

Yes. And I guess the last comment I made is we see expected ROEs in the low double-digits, which certainly, we think is attractive, especially given that most of the assets are floating and would be levered to higher rates. So we're really excited about it. We don't really have much more to say than that at this point, because it's really early. But the next several quarters, we'll obviously hope to have a lot more to talk about.

Yes. Fair enough. That's helpful. And is this -- would you anticipate using FHLB financing for these loans?

Yes, great question. We're working with them quite a bit, looking at their collateral requirements. Some of this -- some of these senior products might be fit really nicely in their "guidelines. " So yes, we're definitely looking at them. I think it might be a nice fit for a portion of the production.

Our next question comes from the line of Rick Shane of JPMorgan.

I will say I'm not big on compliments on earnings calls, but I really do appreciate the disclosure. I think you guys do a very good job helping investors understand what's going on here. I want to explore Slide 13 just a little bit more. I know you just went through this a little bit with Trevor, but there's an interesting sort of element of, heads, I win; tails, you lose, to the swap strategy right now. And I think what's happening here is that you guys are trying to, in a longer, lower rate environment, maximize spread, which seems like a good strategy. But I would be curious if you extended this book value exposure to change in rates table out to plus 200, plus 300, would we see, because of the swaptions, that it starts to flatten out, that you basically have a lot of sort of tail risk insurance in place and that you sort of lose that normal, as rates continue to rise, further deterioration of book value?

Yes. Rick, that's a good observation. The interesting thing about this table on Slide 13, it's a standard disclosure of an immediate parallel shift right, which I don't think any of us on this call have ever seen an immediate parallel shift of exactly 50 or 100 basis points. And it's obviously driven out of the model. But the bigger point of the table, first of all, is just to give a rough idea of we take in a lot of interest rate exposure or not so much. And if you look at the table on the left, you can see that, generally, we don't take very much and our results over time have obviously shown that in different rate environments, we've done pretty well. I think in terms of your up 200, up 300, obviously, it's a little challenging to see that happening today, but markets always find a way to surprise us. If you look at our assets, they're really very stable. So the prepay -- obviously, the HCEMs don't really move much, they're fairly short. We generally have higher coupon prepaid protected pools on the Agency side. In those, the cash flow variability is much less, because there's less volatility in prepayments. So one of the things we're thinking about is those cash flows are generally fairly stable. You don't get a massive amount of extension. And then if you look at our payer swaptions, we have about $8 billion of payer swaptions, which is, like I said, I'd refer you back to Slide 28, typically, those are you can see, they're 56 months to expiration, so basically, 5 years of protection into, on average, about a 7.5-year swap. So the whole point there is, if you look at what's going on in the world today, right, central banks all around the world are easing, European rates are extremely low, and it's hard for us to see long rates going massively higher. But the point is, what we want is we want tail protection against something unforeseen that would drive rates massively higher, and that's the whole point of those swaptions. So while we don't -- we haven't put out any numbers for up 200 or 300, we're pretty comfortable and immediate. We're pretty comfortable that those payer swaptions are both long-dated and plentiful enough to protect us. I hope that gets at what you were -- your point.

Our next question comes from the line of Dan Altscher of FBR.

I just wanted to follow up maybe in a little bit more detail from Trevor's question related to what you're planning on the commercial loan side. Appreciate the color there. But I guess, are you looking to maybe buy loan packages as opposed to actually building out an origination team and doing it directly? Is that kind of the idea?

Yes. Thanks a lot. Yes, I mean, as you know, we announced that -- I mean, Two Harbors' obviously, externally managed by Pine River. And on the Pine River side, we brought in this very senior and well-accomplished seasoned team to basically build out a commercial real estate debt platform. And this is really focused more on loans and origination as opposed to buying CUSIPs in the market. So I think you should expect to see that our holdings will pretty much substantially be all loan-oriented. I'm not ruling CUSIPs out at all, but that's the focus.

I guess my question was more so in terms of we're going to -- are we, as Pine River or Two Harbors going to have a team that's actually going to be actively going out trying to source product, as opposed to there might be a large insurance company for instance that's writing a senior and we want to partner with them to take [indiscernible] or take a B-note. What's the game plan related to that, I guess specifically?

Yes. I mean, look, the team that's going to -- the folks that are building this team out, clearly, are going to be looking to originate loans, okay. And that doesn't mean that they're not going to buy parts of loans that someone else originated or team up with anybody. But they're going to be originators and underwriters who are working directly with either loan brokers or borrowers.

Okay. I think I got it. I think that's helpful. And just following up maybe on Rick's question, at least related to the parallel shifts in the curve, what we've seen from a couple of other folks is some sensitivity analysis around spread widening or basis. Do you have any kind of sensitivity around a 10 -- incremental 10 basis points widening it to book value or change to book value, 25 basis points or something similar like that?

Yes, I think...

[indiscernible] that might be a little bit more applicable in kind of where we are at this point?

Sure. Yes. We -- that's not something we typically put out. I -- what we've generally done is referred people to the Appendix, where you can see what our assets are. We've got close to a couple billion HECMs, which are very short. Our exposure in 30 years is typically in the higher coupons and so most of the stuff we have is fairly short duration. Second of all, obviously, we've got a substantial credit book, most all of which is a petty significant discount. And so if you look at what happened in January, mortgage spreads widened out quite a bit and yet we ended up making money anyway. And I think that talks to the fact that we are typically keeping leverage low and basis exposure low, because if you look at historically where spreads are, they're not particularly exciting in terms of jacking up leverage and taking more risk. The other thing is that our MSR book, obviously is a great basis risk hedge. And so, while we don't really calculate and put that out there, I think it's -- you could probably spend some time looking at what we have and determine what you think our exposure is.

Okay. And then just kind of a geeky question maybe, probably for Brad. Just related to the dividend and taxable income for end of 2014. Do you kind of have rough guesstimate as to what the actual kind of cash taxable income was versus maybe the non-cash taxable income that's generated in the year?

We haven't put that out, and we typically don't. But if you really think about the drivers of our $388 million that we disclosed, obviously, the realized gains are cash. And if you think about the majority of our yield, our run rate or carry whatever you want to refer to it, majority of that is cash flowing. We always do have a bit of what's referenced as kind of phantom income on non-Agency securities. But if you look at the scheme of the number relative -- in our legacy portfolio, the size of it, it's a pretty small amount. So I guess the higher-level answer is, the majority of that is cash flowing return, and that's kind of how we think about our dividends. So if you -- it wasn't -- it's not a surprise, as you would expect to us, that the $379 million comes in nicely with the $388 million, and that's kind of how we think about it.

Okay. So I guess, if I look back to the Qs or the Ks and just look at the, I guess, discount accretion on the non-Agency, that's probably a decent enough ballpark to kind of get me there?

Yes. Dan, it's Tom. I guess this is what they call in radio a teaser. So it's our expectation on our Analyst Day to have more disclosure around our mortgage spread exposure. So hopefully, you can join us then.

Our next question comes from Chris Gamaitoni of Autonomous Research.

Can you give us an updated outlook on the mix -- kind of the goal mix for the conduit business, how much AAAs versus how much subs, IOs?

This is Bill. Yes, so on the conduit side, as I mentioned, our run rate's been going about $300 million a month roughly, for the last several months, so for you math experts out there, I think you can figure out what an annual rate would be. Clearly, that would be a big pickup versus last year and the year before. In terms of subs and IOs versus AAAs, basically on the AAAs, the way we think about that is, is that our general intention is to do a securitization, sell off the AAA, potentially other pieces that we -- like AAs or As, and keep the credit risk, because we're doing all the underwriting and we understand the credit there. Now that being said, one of the things that we have said before is that, it's important for us to be consistent providers of attractive rates as we build our program, and the FHLB allows us to do that. Because to the extent the market gets disruptive and AAAs widen out and are worth retaining, we're happy to do that and fund them at the FHLB until the market settles down and we can either sell them out at tighter spreads or continue to do securitizations in a natural manner. So Chris, the bottom line is, I can't predict what the natural -- what the mix would be, but our general intent is to retain subs and IOs. But from time-to-time, we're happy to retain AAAs if it's warranted by where they're pricing.

Yes. Chris, it's Tom. It's difficult for us to answer that, because we love all our children equally, right? And so it'll just depend upon what the return profile is of this or any other thing going forward.

That makes sense. That intent is really what my question. Obviously, it's market dependent. And the other part of that is, who -- in your conduit business, most of the servicing appears to be done by Cenlar, who's actually -- who's typically a sub-servicer. Who's holding the MSR in that transaction? I'm just trying to get a sense of kind of the outlook of being able to buy additional flow from these counterparties, understanding who is earning the asset versus doing the servicing generally in these transactions?

Yes. On the -- Chris, it's Bill. Yes. So on any of the loans that we aggregate and then securitize, we're holding the servicing. And then, we contract out -- we contract that out to a sub-servicer.

Okay. Perfect. So if for the high-LTV subprime, the same process would occur I'm assuming?

That's -- yes, that's correct.

Our next question comes from the line of Joel Houck of Wells Fargo Securities.

Question has to do with bulk MSR purchases. Obviously, there's a distressed company out there, who just disclosed this morning that they're going to sell between $5 billion and $20 billion of Agency MSRs on a monthly basis. Without commenting specifically on price, there is some, I guess, confusion around whether or not a regulatory monitor will come with any asset sale. Can you guys comment or shed some light on what your understanding is with respect to asset sales -- just in general, not necessarily that particular instance. In general and potential -- if a regulatory monitor would come with it, was that something you would shy away from?

Yes. Thanks, Joel. It's Tom. This is obviously kind of an evolving situation and we really can't comment on things that we may or may not do in the market going forward. Obviously, the MSR market's something that we monitor very closely, and so -- but what we may or may not do in the future, I really couldn't comment on.

I mean, is the regulatory monitor a deal breaker? Or you don't even want to comment on that?

Yes. As I said, it's a developing situation, so I think it would probably not be appropriate for me to make a strong comment one way or the other.

Okay. That's fair enough. Let me ask a different question then. So obviously, the legacy non-Agency exposure came down last year, and you guys mentioned on your prepared remarks. How much of that is a function of -- as you kind of alluded to that you didn't necessarily like the prices or the HPA kind of outlook going forward. Is it a combination of both, or just simply hey, HPA is still okay, but we need to have better pricing in order to take on -- make new purchases?

Yes. Joel, it's Bill. Thanks for joining us today. Yes. I mean, the HPA situation, obviously, we saw a nice run-up and then generally we've seen that taper down to a smaller growth in prices, not only this past year, but going forward. Frankly, the legacy portfolio, it's just really more a question of price. We have bonds that we bought at very attractive prices, and I think we talked about one -- I'm trying to remember when it was, that we bought $60 plus or minus, that at one point, was trading almost at par. And bonds get to levels where there's -- just their yields aren't attractive and there's no more upside. So the decline in our holdings is a combination of pay downs as well as sales. But we certainly wouldn't like anything more to be able to buy some more attractive yields with upside, it's just -- it has nothing to do with HPA, really. It has to do with price.

Okay. That's helpful. And then, I think the last thing I have is, can you maybe comment on the incremental return on CRT bonds versus the legacy non-Agency portfolio? I mean maybe a little compare and contrast?

Well, in terms of -- I think I mentioned a little bit on -- in an earlier one, legacy non-Agencies today, ROEs are in the mid- to high-single digits. In our case, you might find something that stands out from that which has got some good upside and et cetera, and we have been able to do that from time-to-time, but it's not a systemic occurrence. The GSE credit risk funds are really interesting. They look -- they were really interesting when they first came out, and then they tightened dramatically and the risk reward there didn't look that interesting. But recently, in the fourth quarter particularly, it widened out dramatically, and we saw returns there depending on the tranche, but without going into lots of detail, they were double-digit expected returns. So that's why we went and bought some more recently. So I would say, if you want to compare and contrast, once again, it depends on price. But more recently, the credit risk sharing bonds have been more attractive and offered better returns than where the legacy have been trading, but that's not true all the time.

Thank you. And that does conclude our Q&A session. I'd like to turn the conference back over to Mr. Tom Siering for any closing remarks.

Thank you, Ben. I'd like to thank everyone for joining our fourth quarter conference call today. 2014 was a fantastic year for Two Harbors and we are quite proud of the growth of our operational businesses. We are excited about opportunities in 2015, and look forward to keeping you updated on our new initiatives. We will be hosting our Annual Analyst and Investor Day on March 19 at the New York Stock Exchange with Billy Beane of the Oakland A's as our keynote speaker. Additionally, we will be attending the Crédit Suisse Financial Services Forum on February 11. We would welcome the opportunity to speak with you at these events. Have a wonderful day.

Ladies and gentlemen, thank you for your participation in today's conference. This does conclude the program, and you may all disconnect. Have a great rest of your day.