Two Harbors Investment Corp. (TWO) Q2 2012 Earnings Report, Transcript and Summary

Good morning. My name is Mimi, and I'll be your conference facilitator. At this time, I would like to welcome everyone to Two Harbors' Second Quarter 2012 Financial Results Conference Call. [Operator Instructions] I would now like to the call over to Christine Battist, Managing Director of Two Harbors. You may begin.

Thank you, Mimi, and good morning. Welcome to Two Harbors' Second Quarter 2012 Financial Results Conference Call. With me this morning are Tom Siering, President and Chief Executive Officer; Brad Farrell, Chief Financial Officer; and Bill Roth, Co-Chief Investment Officer. After my introductory comments, Tom will provide some insights into the current macro environment and potential impact to our strategy. Then, Brad will highlight some key items from our financial results, and Bill will review our portfolio performance, provide an update on the deployment of capital from our recent stock offerings and discuss the market outlook. 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. This call is also being broadcast live over the Internet and may be accessed on our website in the Investor Relations section under the Events and Presentations link. In addition, we'd like to encourage you to reference the accompanying presentation to this call, which can also be found on our website. Before management begins its discussion of its second quarter results, we wish to remind you that remarks made by Two Harbors' 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 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. Before we discuss second quarter results, I'd like to draw your attention to our recently launched webinar series, beginning with our views on investing in subprime. The webinar can be found on our website. We intend to post additional webinars in the future to provide investors and analysts with management insights regarding the markets and our business. I will now turn the call over to Tom, who will provide some highlights as summarized on Slide 3.

Thanks, Christine. Good morning, everyone, and thank you for joining our call today. We had remarkable performance in the second quarter, delivering 6.9% in total return as we recorded comprehensive income of $0.60 per share. We are pleased to report total stockholder return of nearly 79% since our launch in October 2009. Our book value increased $0.27 to $9.94 per share during the second quarter. Finally, we generated core earnings of $0.35 per share. Brad will provide additional comments regarding our financials in a few moments. In mid-July, we completed a public offering, raising approximately $592 million in net proceeds. We plan to use these proceeds primarily to purchase Agency bonds, but we will also purchase non-Agency securities when we find them at attractive prices, as well as acquiring single-family residential properties. In addition, during the second quarter, we commenced an at-the-aftermarket offering program more commonly referred to as an ATM and sold nearly $78 million worth of additional common stock during the quarter. Although we are quite early in the quarter, non-Agency prices rallied smartly in July, causing Two Harbors book value to rise noticeably since June 30. As a result of this in a flatter yield curve, we believe that yield in net interest spreads on Agency and non-Agency RMBS that are currently available for investment are generally lower than we have historically realized in our portfolio. That said, we believe the opportunity to produce attractive returns for stockholders still exist. Please turn to Slide 4. During last quarter's call, I highlighted some insights on the key macroeconomic factors that would impact our business in the mortgage and housing sectors, including home prices, unemployment and interest rates. This quarter, I'd like to provide an update based upon the changing macroeconomic landscape in the light of recent economic status, including a weakening economic recovery, concerns around the fiscal cliff created by potential tax increases and spending cuts at the end of 2012 and continued concerns about Europe as global markets deteriorated in the second quarter. The European situation is unsettled, and it's not clear that the Congress of nations with disparate fiscal policies and a common monetary policy is viable. Dissolution of the euro or EU represents a significant risk to global markets. With the November Presidential election looming, uncertainties abound and near-term policy changes seem unlikely. Home price performance is meaningful both to our non-Agency portfolio and to our strategy of buying single-family residential properties. Despite an uncertainty surrounding the housing market, we are continuing to see signs of stabilization of home prices and even improvements in certain markets such as Phoenix. For example, national home prices have risen 4 consecutive months according to Zillow in around 0.2% year-over-year as of June 30. This is the first annual increase since 2007. Importantly, buying single-family residential property gives us realtime data on home pricing in key markets, which provides meaningful insights for managing our non-Agency portfolio. As it relates to interest rates, another key factor that we monitor, it appears that low funding costs will be available for the next few years, which is a positive for our business. At the same time, we don't think it is a good time to be taking interest rate risk, and we believe that our portfolio is structured in such a way that changes to interest rates will currently have a modest impact on our book value. During the second quarter, weak economic numbers and concerns about global growth led to lower rates, which generally stimulates refinancing activity. To put current interest rates into perspective, U.S. 10-year Treasuries fell to an all-time low in June at a 1.46% yield. As Bill will share in greater detail, we try to position our Agency portfolio with securities that have embedded prepayment protection. Current unemployment data continues to be stubbornly high and is an ongoing concern. This is a key metric for our portfolio, as next to loan to value, employment is the most powerful determinant of a homeowner's ongoing likelihood to pay their mortgage. So while there appears to be a stream of uneven economic headlines, we believe there is great opportunity ahead for our business. We believe mortgage REITs like Two Harbors can be key participants in the restructuring and rejuvenation of the housing market, as we provide a new source of capital. Please turn to Slide 5. We have continued to make progress purchasing single-family residential properties. As of June 30, we have acquired roughly $72 million in properties. Since quarter end, we have continued to purchase homes. We have added $48 million amounting to a total portfolio of approximately $120 million during the end of July. We are purchasing in cities where current yields are attractive, discounts to replacement costs exist and absorption metrics are favorable. We are currently invested in 8 different markets in Arizona, California, Florida, Nevada, and Georgia. In respect of our asset securitization program, loan production remains muted, and our exposure is reflective with that. The government is currently financing over 90% of all mortgages. We will be opportunistic, yet, patient because the economics of this initiative must make sense versus our other investment alternatives. This was a fantastic quarter for our investment team, and we are optimistic as we look ahead. Brad, I'll turn the call over to you now.

Thank you, Tom. I'd like to focus on 3 topics this morning. I'll start with an overview of our earnings and financing, comment on our investment in real estate properties and conclude on our book value. Let's move to the financial summary on Slide 6. Core earnings at $0.35 per share, represented a 14.3% annualized return on average equity. As a reminder, core earnings is largely a function of our portfolio size, our investment spread and our expense management. I would also add one more item to this mix, which is the cost of hedging our portfolio in an effort to protect book value. I will comment further on this in a moment. So first let's discuss our portfolio's size. On February 24, we completed a public offering of 34.5 million shares for net proceeds of approximately $337 million. Consistent with our expectations and historical experience, we completed the deployment of proceeds within approximately 2 months. This translates into a deployment completion time frame of late April and as a result, impacted core earnings for the second quarter. Relative to our investment spread, we saw lower projected yields on securities acquired in recent months, driving a marginally lower net interest margin for the quarter. Bill will comment further on today's investing environment. Third, our expense ratio as a percent of average equity remain consistent with the first quarter. Finally, our hedging strategy, which is designed to protect book value, also contributed to lower core earnings as a result of the cost of credit default swaps, hedging our non-Agency RMBS portfolio and amortization run-off in our interest-only securities, which hedged interest rate risk on our Agency RMBS portfolio. These hedging costs drive lower core earnings while providing protection on our book value. GAAP net income of $24 million for the quarter not only included core earnings but was also negatively impacted by other-than-temporary impairments on our non-Agency RMBS and mark-to-market losses in our hedging portfolio, namely losses on swaps, which hedged our interest rate exposure and swaptions. As we have previously discussed, because the fair value changes in these hedging instruments are recorded in earnings, while the offsetting fair value changes in our RMBS portfolio are in equity, GAAP net income will experience volatility quarter-to-quarter. This is illustrated in the Appendix on Slide 15. This quarter we had an OTTI adjustment of $4.5 million. This adjustment is in line with our portfolio of growth and continues to represent an immaterial amount relative to our overall holdings. On the financing front, I am very pleased to report that we have increased our financing counterparties as of June 30 to 23 counterparties, up from 21 at the end of the first quarter. And we have also recently renewed our facility with Wells Fargo. In addition, we took steps to extend our maturity profile, including the addition of long-term repos. Bill will comment further on this. I would like to add a few accounting-based comments associated with our investment in real estate to assist investors. For the quarter, we had a weighted average investment in rental properties of approximately $32 million as compared to our June 30 balance of $71.7 million. This weighted average measure is important on 2 fronts. First, it illustrates the immaterial nature of our holdings in the second quarter and its correlated immaterial impact on our income statement. Second, the significant portion of our investment to date has been acquired within the last 30 to 60 days. This short holding period of our investment illustrates that the cost basis on the balance sheet remains consistent with the fair value of these properties in the marketplace. As such, our stockholders' equity remains consistent with our economic book value as of June 30. Now let's move to Slide 7, which contains sequential quarterly book value roll forwards that we believe are meaningful. As Tom noted, our book value per share was $9.94 for this quarter, a $0.27 increase relative to book value per share of $9.67 at the end of the first quarter. The key drivers to this increase were the positive impact of fair value strengthening in both our Agency and non-Agency strategies net of hedges. Now I'll turn the call over to Bill.

Thanks, Brad. I will begin today with an update on our portfolio, followed by commentary on the market and our outlook. Please turn to Slide 8. Our portfolio delivered another quarter of attractive returns. We are quite pleased with our total return on book value for the quarter of 6.9%, which we believe is due to our opportunistic hybrid approach. Before I move into results for the quarter, I'd like to touch on the capital rate we completed in July. We are about 40% through deployment, focusing on the Agency securities with similar attributes that we have in our June portfolio. We believe the deployment will take approximately 1 to 2 months, which means that we expect to finish up around the end of September. Our results for the second quarter were driven by strong underlying portfolio performance as well as unrealized gains from both Agency and non-Agency's net of hedges. At Two Harbors, we are focused on delivering an attractive, risk-adjusted return as measured by comprehensive income. And as Brad noted earlier, core earnings can sometimes be impacted by the cost of hedging. On the bottom left of this slide, we have a chart depicting our portfolio metrics by strategy. Our aggregate asset yields for the quarter was 4.6%, and our aggregate net interest spread was 3.6%. Both metrics were slightly lower than last quarter but still produced an attractive investment ROE. Funding cost, including hedges, were 1%, which was in line with the prior year quarter. Our Agency cost of financing increased slightly due to modestly higher repo costs and the extension of maturities in our repo book. Our Agency net interest spread was 2.5%, down from 2.8% in the first quarter. Factors impacting this tightening included the overall yield -- lower yield environment, continued Fed purchases of securities during the quarter, the flattening yield curve and the market's outlook regarding the increasing likelihood of QE3. While this environment has been good for our book value, it is lower -- led to lower spreads on reinvestments. On the bottom right of Slide 8, we have included some benchmarking metrics. As you can see, a simple duration-hedged Agency strategy would have had a return for the quarter of 3.7%, and mortgage credit returns as measured by an ABX, index, would have been in the 3% range. A simple capital allocation strategy of 50% Agency and 50% non-Agency subprime would have generated a blended return of 3.4% for the quarter. We believe that our return of 6.9% is compelling compared to these indices and speaks to the importance of security selection and dynamic capital allocation. As regards security selection, I would like to take a minute to highlight its importance with an example from our Agency portfolio. Please turn to Slide 9. Here, we compare the performance of generic Fannie 4.5s, which we don't own, to low loan balance 4.5s, which we do. Even though the latter costs more than the generic, the payout can be well worth it. In the second quarter, generic Fannie 4.5s appreciated 1.5% and experienced a 3-month CPR of 25.6%. Low loan balance 4.5s appreciated 2.8% and realized only a 5.6% CPR. When you combine the better price performance with the difference in loss due to prepayments, we see that the low loan balance pool outperformed by 1.3%. One other point to note at the bottom of the slide is that despite being about 4 points higher in price than generic Fannie 4.5s at June 30, the low loan balance pools at the June 30 price still have a much higher expected yield and lower expected prepayment speed. We own approximately $1 billion of these bonds, which represents about 12% of our Agency portfolio. The performance of these and other prepaid protected securities was a major contributor to our strong Agency portfolio performance in the second quarter. On the non-Agency side, our strategy delivered an annualized yield of 9.6% and a net interest spread of 7.3% in the second quarter. We continue to be pleased with the performance of our portfolio as underlying fundamentals continued to improve. Delinquencies have declined relative to the last several years, and the 12-month current pay metric is increasing, which points to fewer borrowers having problems making mortgage payments and to more successful loan modifications. Also, we have seen a slight uptick in prepays, although there has been nothing meaningful yet. Furthermore, the housing market at the lower price end is stable, and in some cases, improving, which bodes well for subprime bond performance. The non-Agency market enjoyed a modest rally in the quarter with subprime bond prices up a few percent. As we show on Slide 10, our portfolio has grown to $9 billion in Agency securities including in Versailles and $2 billion in non-Agencies or about an 82-18 asset split. Our total portfolio grew by nearly $1.6 billion as a result of capital raised and appreciation of our holdings. Our asset mix is comparable to last quarter, although the allocation to Agency is a bit higher. As you can see, the portfolio composition was relatively consistent in mix and type of securities. We continue to emphasize Agency securities with prepayment protection and subprime bonds for our non-Agency portfolio. On the top right, you can see that we shifted our capital allocation slightly to Agency securities following the rally in non-Agencies. There is more detailed information on our Agency and non-Agency holdings in the appendix. Next, I would like to discuss portfolio metrics and our risk profile. Please turn to Slide 11. Despite generally higher prepayments in the overall market, our Agency CPR remains in the 5% to 6% range, increasing modestly to 5.6% from 5.2% in the first quarter. While we continue to expect prepayment increase due to both the lower rate environment and the influence of policy initiatives, we believe that our securities are unlikely to experience a significant increase in prepayments, as 97% of our Agency portfolio had some degree of prepayment protection. Our capital allocation figures are based on applying leverage in the range of 6 to 7x for Agency, and 1 to 1.5x for non-Agency, which is consistent with our approach since Two Harbors was formed. You can see at the bottom that in the second quarter, our aggregate portfolio had a debt-to-equity ratio of 4.3:1. This is higher than at March 31, which was low due -- lower due to the timing of the capital deployments of our February 2012 stock offering. We currently estimate our leverage going forward to be in the mid-4 range, up slightly from a range of 4 to 4.5 previously due to a higher allocation to the Agency strategy as we deploy capital from our July stock offering. As Tom mentioned earlier, with rates so low, we don't believe this a good time to take interest rate risk. As of quarter end, we estimate an impact of approximately 3% on our equity for an up 100 basis point move in interest rates. This modest increase versus the first quarter comes from having gained duration on our swaptions as a result of the second quarter rate rally. As a reminder, swaptions lose hedging effectiveness in a rally, so we gained duration in the portfolio. But the swaptions increase in efficacy in a selloff, which is the beauty of using them as part of a mortgage hedging strategy. Currently, the cost of long-dated protection is cheap, and we are pleased to report that we have over $3 billion in notional protection via swaptions with over 6 months to expiration. More details on our swaps and swaptions are included in the Appendix. You will note that the bulk of our swaptions average 4 years to expiration, and underlying swap tenders are almost 10 years. While rates may not go up anytime soon, we believe these long-dated swaptions will protect us if and when they do. Lastly, our interest rate swaps average pay rate as of quarter end was 0.87% with an average maturity of about 2.5 years. As Brad mentioned, we increased our counterparties during the second quarter and have recently extended for another year to Wells Fargo facility, which funds some of our non-Agency holdings. To further fortify our financing for non-Agencies, we have also entered into both 3-year and 4-year repo lines for up to $200 million in funding, of which about $50 million was being used as of quarter end. While a bit more costly than standard repo, we believe the substantially longer-term provides stability and certainty to our funding mix. As a result of this continued endeavor to lengthening [ph] funding terms, the weighted average days to maturity on our RMBS repo borrowings at June 30 increased to 86 days, up from 80 at March 31. More financing details are in the Appendix. I'd like to wrap up today with some general comments on the market as summarized on Slide 12. Rates are low and have continued their march lower since quarter end. A variety of factors have led to this as Tom mentioned. Funding rates are low and are likely to remain so for quite a while. Prepayments are higher, and that will probably persist as well, but market participants do not expect they will be as fast as they were in 2003. Finally, there is the prospect of QE3. As a result of all these factors, the curve is much flatter, and Agency RMBS had performed quite well. This has been a boom to our existing portfolio since quarter end but somewhat challenging for new investments. On the non-Agency side, we have seen a strong rally since quarter end, as the stabilizing housing market and improving fundamentals have led to widespread buying by investors looking to lock in attractive yields. Additionally, the technicals remain strong with currently few distressed sellers and a significant amount of capital that has been raised for distressed U.S. assets, which needs to get invested. The Maiden Lane 3 auction of CDOs have gone very well, with some bonds trading above the value of the underlying assets. All of this is occurring amidst an environment of ever shrinking yields in other market sectors. Thus, the yields available to the non-Agency market are markedly lower than what we saw in the second half of this year. On Slide 12, you can see the expected yields on our portfolio as of June 30 and also representative yields available in the market today. The lower yields mean higher asset prices, and as a result of this, our book value has grown meaningfully since quarter end. While we are only 1 month into the quarter and are certainly pleased with the increase in book value, the rally obviously results in generally lower expected returns on capital being deployed today, particularly within the non-Agency market. In conclusion, we are happy about our performance this quarter, our current capital allocation and the characteristics of our portfolio. We will continue to be mindful of security selection and believe that opportunities will continue to exist for attractive returns for our stockholders. I would now like to turn the call back over to Mimi for the Q&A.

[Operator Instructions] Our first call is from Bose George of KBW. Bose George - Keefe, Bruyette, & Woods, Inc., Research Division: My first question was just on the yields on the new investments. On Page 12, you showed the yields on June 30. But then for July, it's kind of broad. So just wondering if you can narrow that down a little bit in terms of the net yields on the stock [ph] you're putting on.

Thanks for joining us, Bose. The -- yes. I mean, that's a great question. Let me start with the non-Agency first. We tried to capture the range of where we're seeing yields on prime through the subprime. So the range is, obviously, pretty broad there. Some of the bonds are at the lower end. The bonds that we've typically focused on, which have been the more deeply discounted subprime, where we can make pretty Draconish assumptions are near the higher end of that range. So given that, that's a sector that we continue to like, I think you should expect to the extent that we get non-Agency, they'll be closer to the high end than the low end of that range. On the Agency side, there's a tremendous amount of variability in yields, depending on whether you're looking at hybrid ARMS or reverse mortgages or 30-year current coupons. So we didn't actually capture the entire yield range because, as you know, some of the paper is in the 1% to 2% range. But we tried to be more indicative of the security -- the sectors that we particularly like and provide a range that was more indicative of that. Bose George - Keefe, Bruyette, & Woods, Inc., Research Division: Okay, great. And then just like you're switching to the single-family business, so I was just wondering how large that could get. And also when could you potentially start putting some leverage on there?

Yes, Bose, it's Tom. The honest answer is we don't know how big it could be. It's going to be a function of opportunity and supply, so obviously, it's got to be attractive relative to our other investment alternatives. And given the size of our portfolio and the size of the single-family allocation, right now, it finances really easily within our portfolio, so we can potentially use our portfolio to effectively apply leverage to it. If it were to grow meaningfully, in context to the portfolio, at some point, maybe that becomes untenable. But today, it finances quite well within the REIT.

Our next question comes from Douglas Harter of Credit Suisse. Douglas Harter - Crédit Suisse AG, Research Division: I was just wondering if you could talk about your philosophy of whether you'll be sort of leveraging up the nice book value appreciation you've had or would you just use that portfolio appreciation to take down the leverage?

Yes. I mean, in the short run we have to be mindful that the market is -- can be volatile, and so you need to be careful about leveraging gains that are capital. But as it becomes evident that those gains are, frankly, permanent and there for a longer period of time, that is part of our capital base. So we would look at that as capital that could be leveraged and would be leveraged in the context of our overall portfolio construction.

Yes, and I would also note that we do carry a fairly sizable amount of cash on hand, and we continuously monitor excess liquidity. So we take those into account in how we look at our leverage and fairly conservative in how we approach it. Douglas Harter - Crédit Suisse AG, Research Division: Great. And then just thinking at -- with your stock above 11, if you could sort of refresh us about the outstanding warrants?

Sure, Doug. It's Tom. Thanks for joining us. With respect to the warrants, obviously, they're in the money now. And so it's possible that people may exercise those warrants. But other than that, we really don't have any information to share about them.

Want to remind him what the terms are?

Yes, they're struck at -- for everyone's [ph] benefit, they're struck at $11. They expire in a little over a year. So people may start to exercise those. And obviously to the extent that they do, that will be released in upcoming financial data in our various filings. Douglas Harter - Crédit Suisse AG, Research Division: And just for perspective, how would you -- at this price -- I mean, I guess, this -- that would be accretive to your book value to the extent that people exercise them at $11?

Well, obviously versus June 30 of book value, it would be -- as we said in July, our book value rallied smartly. So obviously, book value is always fluid. So it's possible -- it's possible it is accretive to book value. But at a certain -- it may not be, too. It's a function of what book value at any given moment.

Our next question comes from Mark DeVries of Barclays.

Just want to go back to the current July yields you're seeing. The yields you gave for Agency, is there any way you can give us some sense what that would translate into a hedge adjusted spread?

Yes, sure. Mark, thanks for joining us. Yes, when we looked at the market earlier in July, it looked -- and we said on the call when we came to market with an offering that we saw, low teens returns available. And the way you get there using -- if you took the midpoint of our 6 to 7 on Agency, call it, 6.5, the asset yields at that time were at the high end of what you see on Slide 12, and funding cost, including hedges was sort of in the 95, 100 basis points. So that would get you a NIM of, call it, roughly 175 plus or minus and an ROE in the 14 area. So what we've seen since then is we've seen the market continued to rally. So you could expect yields will be lower than the high end of the range, and yet hedging costs have also come down. So spreads are not significantly different than what we saw at that time.

Okay. Got it. And Bill, I'd be interested to get your thoughts on how you would expect non-Agencies to react if we get QE3. I mean, if I remember correctly, we've had pretty divergent reactions to QE1 and QE2 or I think risk assets really rallied in QE1 and sold off on QE2. Which direction would you expect non-Agencies to take up at the QE3?

Yes. I mean, we're -- I think QE2 sort of took hoist in a time when there was a lot of other stuff going on, which wasn't that favorable for risk assets. We're in an environment now that -- as I mentioned in my prepared remarks, the technicals are highly favoring risk assets and frankly, any assets with yield. I mean, yields right now are very tough to come by. We've seen a substantial amount of money raised to buy distressed U.S. residential and commercial assets. I think there's a belief that housing is at or near bottom or maybe off the bottom. And so people feel safe to go back in the water if you will. So if there's assets available that have any sort of reasonable yield and people feel comfortable with their assumptions, I think that, that's what we've seen so far. And I don't think that QE3 would do anything to change -- if anything, it might help people feel more comfortable in risk assets or assets with yield to the extent that QE3 diminishes yields further.

Yes, Mark. I would -- It's Tom. The thing that I would add to that, too, we think what's really sort of in the mean around non-Agencies are what's going on in the housing market. Obviously, non-Agencies are greatly impacted by probability of default and recovery. And to the extent that you have a flattening in the price in the housing market or God forbid, even a rally, that's very impactful on non-Agency price performance. And we think that is probably more of a driver in the market today than any sort of government policy. We really think that's why people are being attracted to the non-Agency market.

Okay, that's helpful. And then just last question, have -- you're not into the rally enough that you might actually consider selling some of them or do you think still think they're kind of trading below what you view is their intrinsic value?

Well, Bill and I say everything's for sale at a price except for our wives and children. But we continuously evaluate the market. I will say that, that while non-Agencies have rallied substantially, they're still arguably very attractive relative to other investment alternatives. So it's something that Bill and the team constantly evaluate. So we may sell securities at some point. We may not. It's something that we monitor on a daily basis. And one thing that I would point out and we really mean this sincerely, our initiative in the residential housing market really gives us good data, real live data about what's going on in the housing market. And at the end of the day, the performance of non-Agency securities are going to be determined by the underlying loan performance, and so that's the pulse that we can constantly monitor.

Our next question comes from Trevor Cranston of JMP Securities.

Can you guys talk a little bit more about the increased hedging cost this quarter on the default swaps and IOs? Is that more a function of larger hedged positions? Or is it more related to things like faster prepayment speeds increasing the amortization on the IOs?

Yes. So on IOs, and I'll defer to Brad on this to add some accounting related comments, but basically IOs after [ph] the amortized cost, the yield that we realize on that will be driven by how fast prepayments are. So to the extent that we've had a -- basically a continuous rally and the speed had picked up, the realized yields on those are obviously lower. So that's obviously a contributor. Now at the end of the day, the way we look at it is that's actually -- if you think about the total or economic return of the IOs, right, it's basically whatever the realized prepays are plus the change in price. So we compare that as a hedging tool to swaps. I mean, clearly, if you put swaps on and you had a big rally, you're going to have more cost from the negative carry as well as the mark to market. So it's not that different, but the IOs did pick up in speed over the last 3 to 6 months, which drove a lower yield. And on the -- and then on the credit side, as you know, we do, from time to time, take modest hedge positions that we believe will protect our non-Agencies, and you'll see in the Q, which comes out tomorrow, what those positions are. In your typical CDS, there's a running cost that you pay for protection, whether it's on high yield or investment grade or single name or ABX or anything else. And so there was a slight drag from that as well. But we believe that including that in our mix is very important to protecting book value. You may recall the third quarter last year, we had a tremendous benefit from having some protection in play when the non-Agency market sold off dramatically.

Yes. This is Brad. I'll just maybe add a couple of points more quantitative around the size. If you go to our -- Slide 16 in our deck, we have had a bit more disclosure around the numerical impact on our CDSs. If you look at Footnote 2, we do note that in Q1, we had $2.7 million of net cost on the CDS and that upticked to $3.9 million in the second quarter. As well as we did disclose the gains on our IOs, the kind of yield numerical number was 6.7 in Q1 relative to 7.6 in Q2, which actually equates to a drop in yield as Bill mentioned. So hopefully, that gives you a little more quantitative color in addition to Bill's comments.

Our next question comes from Joel Houck of Wells Fargo.

The -- I guess, to start off with on the single-family strategy, you have a comment in here that says long-duration, potential home price appreciation. That, to me, implies that you're looking at this as a longer-term asset. A couple of questions. What is a typical holding period for these properties? And second is how do you look at the expected IRR of this strategy versus buying a seasoned credit bond, because we get lots of questions from clients about the strategy and while I know it's still relatively small, there are others looking at the strategy and it just strikes me as interesting because with the carrying cost associated with the strategy, one would think that you guys see massive amounts of value here, otherwise, why bother?

Sure, Joel. It's Tom. In respect to the holding period, the answer to that is we really don't know. What's driving the strategy is that we can buy, as we said, homes at attractive current yields and discounts for replacement cost. And we wanted -- we are in markets where we think absorption metrics are -- provide a wind to our back. But we will say honestly if there's a sharp rally in the housing market tomorrow, our holding period might be quite small. There's a lot of research data out there that says that in key markets, supply-demand metrics will come back into balance, anywhere from 2 to 6 years depending upon the market that you're talking about. But it's really going to be a function of the price of that house, relative to what we think fair value is. Today, we -- this initiative is quite nascent, so we don't want to be disingenuous and spit out a lot of numbers around it. But our early experience has been quite pleasing, and we're quite excited about this program.

Yes, the only thing I would add is, obviously, within the REIT, as well, there are a number of kind of Safe Harbors to comply with holding periods. So generally speaking, we are looking at the long-term asset to fall inside those guidelines.

Yes. I mean, we don't expect the housing market to recover overnight. As we've said, we're seeing a flattening of prices in some markets. We're seeing somewhat of a rally in other markets. But we -- our business is to optimize shareholder value. So we'll continuously monitor this versus our other alternatives, and it's got to compete well. We never want to be in a business just to be in a business. It's got to be attractive for our shareholders. And so as I responded to an earlier question, how big we get in this will be just an opportunity of supply and opportunity.

Okay. That's helpful. I guess, maybe dovetailing more in the Agency side, I mean, given where spreads have gone and particularly where prices are, not just with the generics but the spec pools, obviously, you gave a nice example here of just about $1 billion trading at 111. What -- I guess I'm a little surprised that we continue to see in -- all of the REITs are continuing to raise capital and deploying the Agency space. I understand that most of the capital is deployed -- months ago, you still have some to go. What are your thoughts in terms of the allocation? Because it seems like the tone of this call is more positive toward housing, which I tend to agree with. I don't know what others think. And second, is there any way to express a short view inside a REIT with respect to the Agency sector? I mean, if you felt like the Agency space was too overvalued perhaps we hit QE3 and that pushes up the price more, what would the strategy be to protect value? Would you just simply sell and take down leverage? Or are there ways to efficiently express a short view on the Agency side within a REIT?

Yes. I mean, I guess I have 2 remarks for that. One is that despite the fact that spreads are lower -- from an historical perspective, the ability to generate double-digit investment returns from an Agency strategy is still attractive. So -- and that's consistent with my remarks to the question that Mark asked earlier. The second thing I would say -- and this is not commenting necessarily on what we may or may not do, but certainly, to the extent that you want to take a position that protects against the mortgage prices in a REIT or, frankly, any investment vehicle, you have the ability to shorten TBA -- short TBAs. We have done that occasionally over time to protect certain parts of our Agency bucket. A perfect example is when there was substantially higher prices and their -- HARP was being discussed about being expanded, high coupons were at risk. And we took a substantial position on the short side in TBAs and high coupons that we held for several months until we fully understood what the implications were. So certainly, we and others have the ability to do that. But keep in mind as a REIT, our investors are looking for exposure to mortgage market. So we would be reticent to have effectively no position in the mortgage sector for an extended period of time.

Yes, I guess, that's helpful. I guess, the performance has been outstanding. It would seem like if you were to -- and I'm interested to hear your comments, Bill. If you looked out over the next 3 to 5 years and, ignoring potential QE3, which might be coming, what's your viewpoint on the relative long-term value of non-Agency versus Agency on a 3- to 5-year basis?

Well, I mean, we have roughly half our capital allocated to each strategy. So I think that kind of tells you how we feel about it. In non-Agency, this is just a question of making sure the math works on new purchases in terms of generating attractive ROE and finding bonds that we like at the right price. And on Agency, I think, it's frankly -- the reality is it -- the evaluations on agencies are driven largely by where rates are right now. So it's not necessarily protecting on, let's say, mortgages as much as just protecting your interest rate risk. If rates go up 200 basis points, mortgage prices are going to go down. That's why we don't maintain a high exposure to interest rates. And that's why we have a substantial amount of protection, both in swaps and swaptions and you can see that in the appendix. So I think that, that's really kind of at the heart of what I think you're asking, which is that rates are low and prices are high, and the way to protect that is not to take much interest rate risk.

So you're looking at -- you may say, well, the market, if it's still revalued, we're going to protect ourselves through hedging, but we're not going to make a directional bed. If it happens, we're protected, and we're not going to sit out there and expose ourselves to a broad selloff in mortgages.

Yes, I mean, basically, yes. We're trying to generate an attractive dividend and protect our grow book value. And that's really been our endeavor since we started the company almost 3 years ago. And we continue to have that same approach. So yes, you're correct.

Okay. And lastly, I think you made a comment a while back and just seems -- to see how you still feel about -- I mean, you made a comment about it's the great time to be an Agency investor, just given the structural impediments to refinancing. That has been very--good call. Obviously, it's played out. Do you still feel the same way, structurally, that prepayments speeds are going to be low? And if we are more like Japan, that perhaps the Agency strategy has a lot more life than perhaps what some may be thinking given where rates have gone?

Yes, thanks, Joel. It's Tom. Bill and the team have done a fantastic job standing out of the crosshairs of policy initiatives. And that's why our prepayments speeds have been so low and so consistent. And I have every expectation that they'll continue to be able to do that. And so yes, we're still quite excited about the opportunity within the Agency space. They've done a very good job being thoughtful around what bonds are prone to be included in various government policies and stayed out of the way of all that. And so for those reasons, we're quite confident about the future in the Agency space.

Our next question comes from Jason Weaver of Sterne Agee. Calvin Hotrum - Sterne Agee & Leach Inc., Research Division: This is actually Calvin Hotrum standing in for Jason. Can you guys comment at all on your book value since quarter end?

Sure. It's Tom. So in respect of that, yes, the non-Agency -- if you look at ABX, it's rallied 2% to 5% since quarter end. Generally, people would tell you that cash bonds have modestly outperformed the synthetics. And additionally, we've seen a continued flattening of the yield curve within the Agency space. So it's very early in the quarter, so we don't want to address this quantitatively. But hopefully, that gives you some qualitative perspective.

Our next question comes from Gabe Poggi of FBR. Gabriel J. Poggi - FBR Capital Markets & Co., Research Division: It's Gabe. Just a quick question. You mentioned that your extending the non-Agency funding maturity. Do you guys -- can expect to kind of continue to do that over time, pay up but just to have a little more extension on your funding of non-Agencies?

We have about $2 billion market value. And as you know, our leverage target of 1 to 1.5, means -- roughly call it $1 billion. We've got the Wells Fargo facility as you know, which is a 1 year, and we can -- and we have this $200 million of 3 to 4 years. So we feel like we have pretty good percentage of potential borrowings that we have for an extended period of time. Now that being said, it's all a question of opportunities that we get presented. Certainly, if someone calls us up tomorrow and presented a fabulous deal for, whatever, 2 years, 3 years, 4 years, we would absolutely consider that. Agencies, even during the crisis, funded really very easily. Non-Agencies as you know, you couldn't fund at all. So we feel like while things are good now, we're just protecting and fortifying our finance-ability of those assets.

Gabe, one -- it's Tom. The one thing I would add particularly in the non-Agency space, funding of non-Agencies are absolutely as healthy as they have been today since the crisis has unfolded in respect of number of counterparties that are willing to fund them, terms, maturity, et cetera. So there's really been a return to a sense of normalcy in that space. And so today, it's as good as it's been since the crisis.

Our final question comes from Boris Pialloux of National Securities.

I just have like 2 quick questions. One is relating to your non-Agency funding. I mean, if the rates -- actually, the yields are coming down, would you actually consider increasing your leverage target? I know that you have 1 to 1.5, but just trying to figure out if you -- if this will be you policy for -- in this type of situation.

It's Tom. Bill and I have been around -- despite our youthful appearance, we've around a while. And how people have traditionally have gotten in trouble in the mortgage market is levering up at the wrong time. And so we feel quite comfortable with the amount lot of leverage that we're applying to the non-Agency space. And you shouldn't expect that, that's going to change meaningfully anytime soon.

And second is did you view regarding credit supports have changed because of, let's say, of eminent domain spread and the mortgage modification efforts?

I'm sorry, could you repeat your question?

Basically, in your subprime bond investment strategy, I know that you have some type of credit supports. But now you're having talked about eminent domain and you also have also mortgage modification efforts from the bank. So just trying to figure out this has changed your views regarding credit support.

Yes. No, so -- yes, there's obviously a lot going on, not just with the bank's efforts to help homeowners, but obviously, there's a number of many different policy initiatives as well as the ones we mentioned in the eminent domain. I mean, at the end of the day, one of the things that we really like about the deeply discounted subprime sector is that almost anything that you do to help a borrower is good for those bonds. And I'll give you an example. When we -- we've had in the past and if you look at the webinar, you'll see that we give an example of a non-Agency bond where we -- there's only, say, 30% delinquent, but we're assuming 70% of them default. So there's a lot of guys that are paying their mortgage that we assume default, as well as where we take a big loss. So if you think about borrowers who potentially get helped such that they don't default or even if there is a modification, which results in a minor loss, that's a tremendous benefit to that type of bond. Furthermore -- so in terms of any of the modification efforts, typically, the kind of bond that we own benefits from anything that helps the homeowner. And then regarding eminent domain, frankly, it's an extremely new proposal. I think our understanding is the municipalities are studying this and trying to determine if it's worthwhile. Clearly, to the extent that you did see something like that come about, it would result in a prepayment to the trust, so that would be a good thing for discount bonds. But I have to tell you that there is great concern, and it's been mentioned and noted that this is potentially unconstitutional. So we're not assigning any value -- any positive value to this at this point. But certainly, we're monitoring it, and we would take into consideration anything that does happen on that front.

Thank you. There are no more questions in the queue. I'll turn the call back to Mr. Siering for concluding comments.

Thanks, Mimi. Before we close today's call, we wanted to announce the timing of our upcoming Analyst and Investor Day in New York on October 10. More details regarding this event will be forthcoming in the coming weeks. We hope you can join us in person or via the webcast for this event. We think it will be a very informative day. Thank you, and have a great week.

Ladies and gentlemen, this concludes our conference for today. You may all disconnect, and thank you for participating.