MFA Financial Inc (MFA) 2013 Q3 法說會逐字稿

Ladies and gentlemen, thank you for standing by. Welcome to the MFA Financial Incorporated third-quarter 2013 earnings call.

(Operator Instructions)

As a reminder, this conference is being recorded.

I would now like to turn the conference over to your host, Danielle Rosatelli. Please go ahead.

Good morning.

The information discussed on this conference call today may contain or refer to forward-looking statements regarding MFA Financial Inc which reflect managements' beliefs, expectations and assumptions as to MFA's future performance and operations.

When used, statements that are not historical in nature, including those containing words such as well, believe, expect, anticipate, estimate, plan, continue, intend, should, could, would, may, or similar expressions are intended to identify forward-looking statements. All forward-looking statements speak only as of the date on which they are made.

These types of statements are subject to various known and unknown risks, uncertainties, assumptions and other factors. Including but not limited to those relating to changes in interest rates and the market value of MFA's investment securities, changes in the prepayment rates on the mortgage loans securing MFA's investment securities. Changes in the default rates and managements' assumptions regarding default rates on the mortgage loans securing MFA's MBS. MFA's estimates regarding taxable income and the timing and amount of distributions to stockholders. MFA's ability to maintain its qualification as a real estate investment trust for federal income tax purposes, and MFA's ability to maintain its exemption from registration under the Investment Company Act of 1940.

These and other risks uncertainties and factors, including those described in MFA's annual report on Form 10-K for the year ended December 31, 2012, its quarterly reports on Form 10-Q for the quarter ended March 31 and June 30, 2013. And other reports that it may file from time to time with the Securities and Exchange Commission could cause MFA's actual results to differ materially from those projected, expressed or implied in any forward-looking statements it makes.

For additional information regarding MFA's use of forward-looking statements, please see the relevant disclosure in the press release announcing MFA's third-quarter 2013 financial results.

Thank you for your time. I would now like to turn this call over to Stewart Zimmerman, MFA's Chief Executive Officer.

Good morning, and welcome to MFA's third-quarter 2013 earnings call.

Joining me this morning on the call are Bill Gorin, President; Stephen Yarad, Chief Financial Officer; Craig Knutson, Executive Vice President; Ron Freydberg, Executive Vice President; Terry Meyers, Senior Vice President; Harold Schwartz, Senior Vice President and General Counsel; Kathleen Hanrahan, Senior Vice President and Chief Accounting Officer; Goodmunder Christiansen, First Vice President.

Today we announced financial results for the third quarter ended September 30, 2013. I will briefly discuss recent financial results and other significant highlights for MFA. I will then ask Bill Gorin and Craig Knutson to present certain of this information in greater detail during their webcast slide presentation following my remarks.

Third-quarter 2013 and other highlights include the following. Third-quarter net income per common share of $0.19 and core earnings per common share of $0.18.

In the third quarter we declared two dividends. We declared a special cash dividend of $0.28 per share of common stock on August 1, 2013. We paid the stockholders on August (technical difficulty) 2013. In addition, on September 26, 2013, we declared a regular quarterly dividend of $0.22 per share of common stock, which was paid to stockholders on October 31, 2013.

On September 11, 2013, we timely filed our 2012 tax return. An amount equal to all taxable income for years prior to 2013 has been distributed to stockholders. We currently estimate for the first nine months of 2013, our taxable income was approximately $297 million.

Including the dividend paid on October 31, 2013, we have distributed approximately $236 million in dividends in 2013, not allocated to prior years. Book value per common share was $7.85 as of September 30, 2013. Due primarily to the special dividend of $0.28, book value declined from $8.19 as of June 30, 2013.

In addition, the third-quarter dividend per share exceeded net income per share by $0.03 so as to more closely track REIT taxable income. Due primarily to increases in accretable discount and to changes in the forward curve, the loss-adjusted yield on our Non-Agency portfolio increased from 7.15% in the second quarter to 7.33% in the third quarter.

For the third quarter ended September 30, 2013, we generated net income allocable to common stockholders of $67.3 million, or $0.19 per share of common stock. Core earnings for the third quarter were $64.7 million, or $0.18 per share of common stock. Our quarter-end, our debt-to-equity ratio was 3 to 1.

Our Agency portfolio had an average amortized cost basis of 103.5% of par as of September 30, 2013, and generated a 2.13% yield in the third quarter. Our Non-Agency portfolio had an average amortized cost of 73.4% of par as of September 30, 2013, and generated a loss-adjusted yield of 7.33% in the third quarter.

I thank you for your continued interest in MFA Financial. And at this time I would like to turn the call over to Bill and Craig. After their presentation, we will open the call for questions.

Thank you, Stewart.

Turning to slide 3, you will see there that, as Stewart mentioned, net income per common share was $0.19. Core earnings was $0.18. And the estimated taxable income per common share was $0.23. So taxable income again exceeded net income.

In the third quarter, we declared two dividends. There was a third-quarter dividend of $0.22. In addition, there was a special cash dividend of $0.28. So the total of the two dividends was $0.50. Even after these dividends, our estimated undistributed taxable income per common share as of the end of the quarter was $0.17 per share.

Now, despite change in interest rates and prepayment speeds, our key metrics have remained consistent over the last year. If you look at the yield on interest-earning assets over the last four quarters, it's been a little bit over 4%. The net interest spread has stayed approximately 2.25%. And the debt equity ratio has been very consistent, about 3.1 for each of the last four quarters.

Now, in a period of very low interest rates and great volatility, why are we seeing this result, in terms of interest-earning assets spread and leverage? The answer is, when we diversify into owning both a credit and interest-rate portfolio, we find the two assets are performing differently, and in opposite directions. While the yield on Agency assets has trended down over the last four quarters, similar to other interest-rate sensitive assets, the yield in our Non-Agency assets have gone up.

And this is because of the improving credit performance, which is increasing the yield we are realizing on these assets. That is why the yield has remained consistent over the last four quarters. But basically, that is the key to our strategy.

Moving to slide 5. We see that core EPS has declined over this period of time, despite the fact that the spreads have remained consistent, and leverage has remained consistent. And why is that? It's because we needed to distribute the taxable income which caused us to pay special dividends of $0.78 per share. Therefore the equity base -- and to some extent, the asset base -- have declined over this period of time.

Had we retained this $0.78 per share and continued to earn the same ROE that we are earning, the core EPS would have remained at $0.20. So again, very consistent performance. The equity base did have to go down due to the fact we had to distribute money to shareholders. Now, had these distributions been reinvested by shareholders, we would be in the exact same spot. They would have the same amount of earnings, same amount of dividends had they reinvested the dividends. So there was no detriment to shareholders to receiving this dividend.

Turning to slide 6. The book value did go down in the quarter. That is primarily explained by the fact that we had to distribute $0.28 per share. In addition, because taxable income -- and therefore the dividend -- did exceed earnings, there was another $0.02 to $0.03 that needed to come out, which you see impacted book value per share. So book value, as of the end of the quarter, was $7.85.

Turning to slide 7. Over the course of the year, we have communicated to you about our taxable income. We have now distributed to stockholders an amount equal to all taxable income for years prior to 2013. For the year 2013, we are slightly undistributed. There is undistributed taxable income, as I mentioned. The taxable income was $297 million, while the distributions attributable to this year was approximately $236 million. We actually have until September of next year to distribute all the taxable income for 2013. So we are now in very good shape in terms of our distributions of taxable income.

Turning to slide 8. There had been some changes to the strategy over the year. Now, this was not a rebalancing. We actually very much like the balance of the assets. We didn't have to sell, nor did we sell any agency assets. But what did happen was, we determined our Non-Agency assets were becoming more and more quarterly to interest rates during the course of the second quarter and into the third quarter. As a result, we added longer-term swaps against the duration that we were now measuring on the assets. We added $1.750 billion of five 10-year swaps at a cost of 2.13%. In our mind, locking in 5-year to 10-year funding at a cost near 2% is a very positive thing to do.

Because we are a more mature mortgage rate and we have an older portfolio, we had some longer-term swaps running off that were put on in a very different interest rate environment. So while we did add some swaps at a cost of 2.13%, the good news is, during the quarter, we had $350 million of notional amount of swaps expire that had a weighted average cost of 4.16%. Very expensive. We have a benefit of having these expire, while there is a cost of adding $1.750 billion at 2.13%.

Turning to slide 9. You see the duration for our assets and for our hedges. The duration for our assets is 2.1. But after you net out the impact of the swaps, the net duration is 0.65. Now, I would like to spend -- Craig Knutson will now present the next couple of slides and talk about the credit performance in the underlying portfolio.

Thank you, Bill.

So page 10 we show a graph of the Non-Agency portfolio LTVs. So the average portfolio LTV -- the average LTV on the loans underlying this portfolio, they have declined from about 105% in January of 2012 to less than 85% in September of 2013. Because these LTVs are lower, lower LTVs means that we expect that future defaults will be less than what we have seen in the past. There is less likelihood that loans will default in the future, the lower the LTV today.

In addition, the loans that do default, we would expect to see lower loss severities. And finally, lower LTVs also should translate to higher voluntary prepayments, which, since we purchased these securities at a discount, is a good thing. You will also note in the press release we showed that we moved $71 million in the third quarter from our credit reserve to accretable discount. And a total of $312 million from our credit reserve to accretable discount since January of 2012. Again, this is largely due to this improving LTV.

So why are the LTVs lower? Well, as you know, loan-to-value ratio is the loan amount divided by the property value. And so, simply shown here on this slide the property values underlying these loans have increased. Nearly half of our portfolio -- 45.4% of the underlying loans -- are located in California. And you can see, we have broken out by county here what we have seen happen to home prices in these counties in the last year. So, again, property value goes up, the denominator of the LTV ratio goes up, which drives the LTV ratio down. You can see that in many of these communities, we have seen better than 20% appreciation in home prices in the last 12 months.

So slide 12. The second largest geographic concentration in our portfolio is Florida, with approximately 8% of the portfolio. Similar story here. The home price appreciation numbers are not quite as high as California, but still solidly double-digit home price appreciation. Which, again, drives those LTVs down.

And then finally, what we show here is the transition rates. So this is the Non-Agency portfolio loans. The rate at which we see these loans that are current today go delinquent. So this is a more important metric than, for instance the number of loans that are delinquent, because the number of loans delinquent at a particular point in time is really a backward-looking statistic. Some of those loans might have gone delinquent a year ago, two years ago, and have been in the pipeline. So we like the transition rate as a much better indicator of a forward-looking statistic. This is the rate at which borrowers who our current go 60 days delinquent.

If you look back to the beginning of 2009, that's really where we saw the peak of transition rates. So that's when the most people were going delinquent, from current to delinquent. And you can see that has continued to decline since then. And we are now back to basically the same levels we saw in late 2007 or early 2008. So the bottom line, I think, to take away from this slide is that the worst of the housing crisis is certainly behind us.

Then, finally, speaking to the credit reserve. Again, we moved $71 million in the third quarter from credit reserve to accretable discount. However, we still have a very substantial credit reserve, $1.1 billion, which you see over on the right of the slide, that credit reserve of $1.1 billion. Now, if you look over on the left, you will see the purchase price. So this is average purchase price of Non-Agency securities, is approximately 73%. And if you look to the right, you see a credit reserve of $1.1 billion, represents about 19% of the face amount.

So the way to think about this is, on average, we have paid $0.73 on the dollar for these securities. And with this 19-point credit reserve, we are accreting those to $0.81. So essentially we are saying that we expect to get back $0.81 on the dollar on these securities that we paid $0.73 for. And with that, I will turn the call back over to Stewart.

Thank you, Bill and Craig. And I think it was a terrific slide presentation. Very well done. Having said that, what I would like to do is open the call for questions.

Thank you.

(Operator Instructions)

Daniel Altscher with FBR Capital Markets.

Question for you on the Non-Agency book. It looked like the cost of funds picked up a bit quarter over quarter. Is this related to the new swaps that were put on? Are we seeing it there or was there something else going on there?

Well, you hit the nail on the head. That's exactly what it was. I think the overall funding cost was very similar to what it had been in prior quarters. But for the first time we have allocated swap expense to the Non-Agency cost.

So is it actually just an accounting allocation issue? Or is it actually, those are the actual swaps that are being allocated there?

Well, it is allocated over the entire portfolio. So we look at all the contributors to duration. And we attribute swap costs accordingly.

Okay, I think that-- got it. Okay. A question on the transition rate swaps, which is fantastic and really speaks volumes. Is that -- I see the source of its core logic. But is that -- are those the actual transition rates for your specific portfolio? Or those more generic transition rates for a subset of Non-Agency?

Well, this is our specific portfolio.

Okay, cool. And just one quick question also on the undistributable taxable income. Is it right to be thinking about that as a potential bucket of special dividends on a go-forward basis, if all else is equal?

It is not -- if you consider it a bucket of special dividends, it is cash we have yet to distribute on taxable income that has been earned. But whether it becomes part of the regular dividend or the special dividend, you really can't say.

Okay. Thanks so much.

Sure

Steve DeLaney with JMP.

Stewart, I think we should start, at least me personally, by congratulating you for your outstanding 15-year career at MFA, leading the Company in building this high-quality portfolio we see the results of this quarter. So thank you for that.

Thank you Steve. I appreciate your thoughts.

Yes. So guys, turning to the portfolio and book value. We certainly understand the $0.28 special dividend. We were looking for something even adjusting for the $0.28, maybe closer to the $8 range. And it's not that much lower.

But just curious -- I am sure it was just a matter of ARIDs assumptions on Non-Agency prices and hybrid ARMS, both of which can be very subjective. Wondering if you could give us the way you see the market, just general market color, on both Non-Agency prices and Agency hybrid ARM prices since September 30. Thanks.

So Steve, as far as Non-Agencies, Non-Agencies are certainly up in price since the end of September. Again, it depends on the particular bond and type of security. But I would say in general, Non-Agency prices are up a point, probably a little bit more than that.

Hi, this is Goodmunder. On the Agency side, since quarter-end, prices in our universe are up by about a quarter of a point.

Goodmunder, are you saying, like your seasoned hybrids, you would be seeing a quarter of a point on those bonds and --

Well, when I say a quarter of a point, I am just talking generically about both the hybrids and the 15 years. But if you want to talk specifically about the hybrids, we had some spread-widening going through in the second quarter, and in the third quarter, as well.

And I would say, as of today, they have probably retraced about two-thirds of the spread-widening. It is not back to the absolute lows we saw early in the year, but a majority of the spread-widening has definitely been recurrent.

Okay. Because we have seen some price indications from some of the dealers showing gains as much as a 0.5 a point or 0.75 of a point. Would those be -- given that spread tightening, would those be out of line?

No, it is going to depend on the type of product. So the lower-coupon, longer-duration hybrids, you can talk -- say, lower coupon sale ones and tennant ones -- they did widen more. So they have recovered more of the widening, in terms of spread and price terms. So that would not be inconsistent, no.

Okay, thank you. And one final one. Bill, you and I have talked in the past about RMBS 2.0. And you guys are watching that. But one of the concerns you had expressed, especially with 30-year fixed-rate collateral and rates still near historic lows, you had expressed a concern about duration risk related to RMBS 2.0.

I was just curious. Not that you have to -- your portfolio is performing, you don't have to do anything anytime soon, or anything radical. But if -- we are hearing that more 5.1s are being originated in the jumbo space. And I was curious if you saw the potential for transactions that involve 5.1 collateral ARMs versus 30-year fixed. If your view towards that type of transaction would maybe change.

So Steve, if I remember, the very beginnings of RMBS 2.0, I believe those first securitizations were actually from hybrids.

They were. The early Redwood deals, yes.

Right. But I think what the market discovered was, the bank originated them, the bank wanted to keep them. And that is why you saw RMBS 2.0 evolve towards a fixed rate model. I still think that banks' portfolios are very competitive versus the execution on securitization of hybrids, Steve. Does that answer your question?

Yes. There's no question that is the situation we have today. And whether you're talking about 5.1 or 30-year, I was just curious if paper did become available, whether it was through private mortgage banks or community banks, if the ARM product generally would be more attractive to you than the 30-year product.

Yes. But also as you know, the execution on these securitizations has gotten somewhat more costly.

Yes.

So if it was hard for securitization to compete with a bank before, it's probably somewhat harder now.

Okay. Well, guys, thanks for the time and the color.

Thank you, Steve.

Douglas Harter with Credit Suisse.

You guys showed on that one slide where taxable earnings was $0.04 to $0.05 above your earnings. Is that a trend that is likely to stay in that range? Or is there something that will change the taxable earnings gap?

As we have talked about before, we expect the numbers to get closer over time. So it would not -- it is hard to forecast. It's hard to forecast earnings and taxable earnings both. But I would say that taxable earnings will probably trend down over the next couple of quarters.

Great. And then you guys also talked about the increased correlation on interest rates with your Non-Agencies. Has that held as rates have fallen from the beginning of September?

Yes, I think generally it has. I think much of the reason that those securities began to exhibit more duration is that they are trading at higher dollar prices. Right? So that the higher dollar prices mean that there is less upside in the future, if you will, of credit improvement.

So I don't think it is completely unexpected. But I would say, in general, we think they have continued to trade with the same duration we observed in the beginning of the third quarter.

Doug, we can't think of any asset that isn't highly dependent on what the Fed says at their meetings. So therefore, Non-Agencies fall into that bucket.

All right, that makes sense. Thank you, guys.

Thanks.

Rick Shane with JPMorgan.

Given rate volatility last quarter, how you handled both the asset side and the swap structure, the timing is probably pretty relevant. It strikes me, in terms of the assets, that you stayed pat. But would love to understand the timing of what you did with the swaps. Was it steady throughout the quarter, or did you take advantage of the decline in rates later in the quarter? Just a little bit of color there.

Well, you are right, we did change the asset base over the time. In terms of the exact dates and times when we added the swaps, well, you know the average cost is 2.13% for 5 to 10 years. I don't know if we have timed it right to the exact day, but we did take advantage of swings in interest rates to add the swaps, as you would imagine.

Okay, great. That's what we are trying to figure out. It was one of these things we were trying to figure, if in August, as rates were starting to move up really quickly, you were aggressive, or you took advantage of what happened in September.

Yes, look, what we are trying to communicate is, we did not add the swaps because we didn't know what the Fed was going to do day to day. We added the swaps -- and we started adding at the end of the second quarter, due to the fact that we saw increased correlation between the Non-Agency assets and interest rates.

So it was added systemically as we saw this interest-rate sensitivity go up on the Non-Agency portfolio. But it was not a forced response to the interest-rate sensitivity of our Agency book and uncertainty about interest rates.

Got it. Okay, that is very helpful. And do think you will continue to do that at this point? Can you give us a little bit of sense of what is maturing within the legacy swap portfolio as well?

Well, one, we are happy where we are, in terms of the assets and the hedges that we have. I think the swap rundown over the next -- it's really over the next quarter. I think we only have about $40 million or so of swaps running off. Forcing the good ones to run off, it costs about 4%. And then after that, the next two quarters are low runoffs. Then it really picks up in the second half of next year.

Great. Bill, thank you very much. I just want to say congratulations, Stewart. It has been a pleasure following you over the years. We wish you all the very best. And to everybody who was promoted in light of all this, congratulations to you guys as well.

Thank you very much for your thoughts. I appreciate it.

Thanks, Rick.

Vic Agarwal with Wells Fargo.

After the double-digit increase in HPA, how are you guys looking at the HPA environment over the next 12 months? And how much of that do you think is already discounted in Non-Agency assets?

Well, I guess the first thing I would say is, our future projections for default are not really predicated on an assumption that home prices will continue to go up by 10% or 20%. So we look at HPA in the rear view mirror, rather than estimate what it does in the future.

So I think our credit reserve changes are really based on observable facts within our portfolio. It is things like the transition rate, it's things like the LTV. And again, those lower LTVs, not only do those lower our expected future defaults and lost severities, but we are also seeing higher voluntary prepayment rates.

Sure. So on that point, with higher voluntary prepayment rates, obviously that has been increasing from roughly 14% in the beginning of 2012 to 18% now. Can you give us a sense of -- say if you had a 1% drop in the current LTV, how much higher voluntary prepayments may come out of the portfolio?

It is pretty hard to draw a straight line between those. The other thing that I would caution is, this CPR number that you see, so that 18%, that includes the effect of defaults. Or actually, the net recoveries on defaults. So the voluntary prepay rate for the three-month period, which you will see in the FICA table in the 10-Q, is 12.4%.

But it's very hard to draw a straight line between the two of those. It is a lot of different facts. And again, we don't do this across the whole portfolio. We do it on a bond-by-bond basis.

All right. Thanks guys.

Some of the critical numbers are -- when the LTV was greater than 100%, it was not irrational to default. Because once you improved your credit, you can go buy the house next door at a lower price. Now that the LTVs have improved and they are in the 80%s, whether it is the mid-80%s or low-80%s, you are not going to know as a homeowner. You don't mark-to-market your home every night when you go home and decide to make your mortgage payment. So perhaps the next critical number might be when it gets below 80%, and then it really becomes easier to refi.

That is a fair point. I appreciate the answers. Thanks.

Henry Coffey with Sterne, Agee.

If you just keep on keeping on, we only stock with stable book value, Non-Agency bonds go higher, I think we have all noticed that hybrid values are going higher. You continue to pay dividends, and we make 11% or 12% for a long time without a lot of distress. And that's a pretty good thing. Conversely, if you look out on the horizon, are there asset classes and areas of investment where you can deploy capital and grow earnings? Or conversely, would you ever think about buying back stock?

So the answer to the first part of the question is, yes, we continue to find good investment opportunities that generate low double-digit ROEs. Which is all we have ever really tried to achieve. Because if you go from 15% to 20%s, you are taking on more risk. And we are very happy in the 10% to 12% ROE return for the risk ratio.

In terms of share buybacks, what is important to point out is, we have actually returned a lot of capital to shareholders this year, in terms of the specials. And that is what the share buyback is, it's returning capital because you cannot make good investments. And we can make those good investments.

There actually was one day this quarter -- despite the fact we were paying the special, the stock got so low, we actually did buy that day. I think the stock was close to $7.00 on that day. So we are aware of those things. But we continue to believe that what we do is very high value-added, that people can't do this on their own, and that the double-digit ROE investment opportunities are still there.

Thank you.

Sure.

Jordan Hymowitz with Philadelphia Financial.

If you look at the charts on page 10, which is the non-agency LTV chart, and then chart 3 later, 4 later of the non accretable discount. I don't understand how -- now that we are at 80%, unless there is a big [tower] -- as to how that non accretable discount, that's the amount started coming down appreciably. And you basically have 55% coverage down to it, at this point, with 26%. Or, I'm sorry, with 20% discount. So it would be 60% LTV. Do you know what I'm saying?

I hear what you're saying. The one thing that you need to keep in mind is, the average portfolio LTV is exactly that. It is an average portfolio LTV. So the loans that have 60% LTVs, unfortunately they don't help the loans that have 120% LTVs. So --

Let me ask the question. What percent of loans still have over 100% LTV?

I don't have that here. It's not insignificant.

Would it be more than 25%?

I don't know, Jordan.

Okay. And my --

We will try to provide that for the next quarter.

That would be great. And also, now that Stewart is leaving, have you thought about changing the colors at all? Especially for those dark green, it's difficult to read. You have blue and maize.

We will try it. By the way, congratulations on Saturday. You did very well, Jordan.

We would be glad to discuss this offline. (laughter)

Dan Furtado with Jefferies.

I think Henry and potentially Steve asked this earlier. But just wondering if, when you are looking out, as we see more opportunities, would be the word to use, either between the securitization market, the risk-sharing market, nonperforming loans, MSRs.

We have seen a lot of different things crop up in the REIT, especially on the hybrid REIT side, or the diversified REIT side. And I know that you are targeting the low double-digit ROEs and you're finding the opportunities there. But is there anything that you find incrementally interesting, if you think about the next prong in the strategy, if there is one? Would you go down one of these paths? Or for the foreseeable future, is it just to keep the course you have been on?

Well, we have seen different investments opportunities. But it would be premature to talk about it publicly.

The only thing I would like to add to that, over the last 12, 15 years, we have looked at I don't know how many different opportunities, most of which we have turned down because we have enjoyed the strategy that has been so successful. How Bill and Craig will look at those going forward -- I think there will be a continuity.

They will always look at opportunities. If those opportunities come to fruition, that will be fine. But in terms of what the basis strategy of the Company has been, I think you'll find it to be on a continuous basis.

Understood. Thank you for that. And then the second question is touching a little bit more on this credit reserve.

Just so that I understand, the current yield, 7.33%, is based on about projected defaults of 32% of the underlying loans. You're saying that the 60-day-plus bucket, or the 60-day bucket only represents 17%. So basically, about another doubling of losses. Assuming all those 60s go belly up, then you are assuming another, not quite doubling, but almost doubling, 15% versus 17%.

I am just wondering. Do your forward projections -- I know, Craig, you look at this on a bond-by-bond basis. But generally speaking, do your forward projections assume incrementally worsening credit performance? Because it seems that just from a percentage basis, that to get another 15% of potential losses, you would necessarily have to assume roll rates start to tick back higher.

Not necessarily, Dan. I think -- you're right, we look at this on a bond-by-bond basis. And basically, what we do is, we bucket the loans in each bond by LTV. So the column that we are really referring to here would be primarily the loans that we still believe have LTVs over 100%. Because those are the ones that we feel are at risk in the future.

And I will tell you that we are defaulting a lot less than 100% of those loans that we think have LTVs over 100%. But again, it is a quarter-by-quarter, bond-by-bond thing. Also keep in mind that in many cases, we are reviewing bonds every nine months or so. When we sit down and review a bond, we may find that we have nine months of home price appreciation, which gives us some reason to change that credit reserve. It is not necessarily a month-to-month thing. There is a little bit more of a time period between the two.

Understood. Thank you for that clarity. I appreciate it.

Sure.

Chris Donat with Sandler O'Neill.

Another question about the slide 10 with the LTVs, and just in the move from 105 points to 85 points. Can you give us a rough sense of how much of that is home price appreciation and how much is amortization? I would guess it's 10% to 20% of the move is amortization. But just wonder if you could provide a sense around there.

Yes. The majority of it is home price appreciation. But again, more than half of the portfolio is also amortizing. And to the extent that those amortizing loans are post-reset hybrids, the rates are pretty low, 3% or so. So they probably amortize to the tune of about 3% a year. So your numbers are probably about right. If I had to guess, I would say maybe about 20% is due to amortization.

Okay. And then looking ahead, it should be steady-state there? Just looking at your non-agencies got $1.5 billion of 30-year. But then obviously, the biggest part is the 3.1% of the ARMs with less than two years. But --

So going forward, certainly the amortization component, I think you can expect it will be steady. In fact, the amortization component may increase somewhat as loans begin amortizing that have been in IO periods. As far as the more significant contributor to LTV declines, which is home price depreciation, as I said before, we don't inject projections of future home price increases into our credit reserve analysis.

Right. I appreciate that you don't. But I think the market at times is always tempted to do things like that. So, anyway, just trying to figure out what is sort of bankable on the amortization side, versus what can be applied from an analyst or investor perspective.

And then just one question on net interest income. It also seems like a steady-state function. You got a little bit of a headwind from higher hedging, a bit of a tailwind from accretable discount. Anyway, going forward, net interest income seems roughly like it should be stable here. Any thoughts there?

We are not going to make a forecast on that.

Okay. Thought I would try at least. But okay, appreciate that.

Mike Widner with KBW.

So let me just ask about on the non-agency side just to start there. Obviously you guys bought a lot of those securities at much cheaper prices. So prices are a lot higher now. Therefore, the yields available for incremental capital are lower now. Plus on top of that, you are now allocating swaps against those, so the cost of funding is higher. Where do you see the incremental net spread, if you were to put an extra $1 of capital to work today on the non-agency side?

So Mike, I would say, again it depends on the bond obviously. But in general, I would say the yields available on these securities today are 5%, give or take, loss adjusted yields.

Yes. That is not quite 7.33%, unfortunately.

No, it's not.

So, again the question -- we have not seen the breakdown of how you are allocating swaps against them. But you had been running a 2.5%ish cost of funds that now you're showing more like 2.9%. So is it fair to say that 2.9% is still representative of funds cost for buying new assets today?

I would say, as you know, we have a significant amount of longer-term financing in place for Non-Agencies. And that is more expensive than funding is, incrementally. So I would say incremental funding, repo for Non-Agencies, again, it really depends on the bond. But it could be as low as, say, 1.5%.

And that is a repo cost as opposed to a fully loaded with swaps and everything cost?

That's correct.

I would like to point out when Craig mentioned that the yield would be 5% or maybe slightly less, that is for a lower duration non-agency asset where you don't really have to add incremental hedge. So it really is your yield less your cost of funds.

Okay. So in any case then, we are still looking at -- if I use that number -- 5% yield, and 1.5%ish if you're doing repo financing and not putting swaps. When you are looking at 3.5% net spread, which is still pretty good. But how do you feel like that compares to the Agency side, with durations pretty much already having extended, and spreads still decent over there?

Just using your numbers and using the average leverage you see, I think you see the Non-Agency yield is between 10% to 12%.

Yes.

And the Agency numbers would not be very different.

So that -- which really brings us back to the question of capital allocation. And you are roughly 70%/30% or so, off the top of my head. Where we stand right now, do you have any preference on either side? There was a long time when you weren't buying any agencies at all, and then you went to redeploying capital within their respective segments, roughly speaking. And so how do you see that right now?

I don't know if there was ever a redeployment. We were at a ratio we were happy with. And what happened was, the non-agencies gained so much value that we had to add agencies to keep the relationship the same. So I think we are comfortable with the ratio we have now. We don't really see that changing. And really, it did not change. When we were buying agencies, it was because the non-agencies were appreciating so rapidly.

Okay. And so going forward from here, as payment -- as principal gets repaid and as bonds mature and all that. How do you see the -- where would you put your incremental dollars of capital today?

Right now we don't see the ratio changing.

Okay. Fair enough. I think you covered pretty much everything else. Appreciate it. And congratulations on good portfolio management and keeping the book value stable once everything is backed out. Thanks, Stewart. Congratulations on all that stuff, as well. Everyone else already said it.

Thank you, Mike, I appreciate it.

Howard Hennick with Scurlydog Capital.

You ducked this question before. But I will ask it slightly differently. You added a bunch of duration, which I think makes total sense in this environment, given the appreciation price of the non-agencies. But I don't understand how you can add that much duration on the swap side and not affect spread. And so, I will ask the question, how does that not affect spread going forward, and also the dividend?

All right, well, what I -- do you mean negative duration?

Yes, you're rating duration on the swap side. And I'm -- I traded for years. So you're rating duration on the swap side, so your duration of equity went down, correct?

Yes. So there is a higher cost of funds. But fortunately, we have had increasing yields in non-agencies.

Right. So you think they -- you are saying, basically, that you think your spread going forward, et cetera, is (inaudible) at this point is no different than your spread retroactively?

We are not telling you what our spread will be after this period. We are not forecasting that. We are telling you there are moving parts, and we are pointing out what those moving parts are. So we haven't really ducked anything.

Right, okay. So you are not -- okay. I still don't understand that answer.

And the other question, I think you basically talked to Mike about. You think right now, the incremental capital basically between non-agency and agency is roughly flat, both yielding on a fully levered basis, with the leverage on the agencies being lower of 10% to 12%. Correct? So you think there is no real difference, at this point, in terms of choosing which one to put in, which one is cheaper?

For a long time, the question was always answered -- oh, the non-agencies are definitely cheaper. Now you're saying they are equally interesting?

No. The ROEs were approximately the same. We did not say the interest rate risks were the same.

No, that's what I meant -- the ROEs. The fully hedged ROEs, or the -- as much hedge as you going to do it.

The answer to your question, the best you can answer it is that, yes, they are somewhat similar, they are somewhat pari passu. But then the question is, in terms of deployment of capital, where we think it's most advantageous.

And again, as Bill said, we are not going to look forward and go through that discussion today. But again, there are times when we feel very comfortable with the agencies. And there are times we feel a lot more comfortable with the non-agencies. And that will be dependent on the marketplace.

All right. And my last question is obviously the non-agency leverages, a lot less. What is it roughly? 2 to 1, versus roughly 6 or 7 to 1? Is that correct, ballpark?

Yes.

And would you consider increasing the leverage on the Non-Agencies to increase returns?

We have never gone in that direction. We have never changed -- we have never incurred leverage as we were trying to hit a bogey. That is not the way the Company has been run for the last 15 years. I think that Bill and Craig will follow a similar philosophy in terms of running this Company properly and not chasing the bogey.

That is not what I was saying. But okay, thank you.

Arren Cyganovich with Evercore.

I was just curious of what you see in terms of new non-agency paper, new legacy books. Have they been opening up at all, given the increases in price?

I know that DeMarco has talked about looking to sell some of the more -- getting the GSEs to sell some of the less liquid pieces of their retrained portfolios. Just your thoughts on any kind of new supply coming to market.

So this is legacy non-agency new supply?

Yes.

So yes, the GSEs have sold off and on through the course of this year. Typically what they have sold have been much older vintage, much higher dollar-price-type paper. So while they have sold some of that paper, it hasn't necessarily fit us.

We do see paper that is still available. It is not as plentiful as it was a year ago, certainly. But we are still able to add in small pieces.

Okay. And then lastly, if the Administration gets its way and can put a new director into the FHFA, do you see any kind of impact to your business from a new director?

It's kind of tough. It depends on who the director is, and it depends on what decisions they make. But the answer is, I am very comfortable about going forward.

What I have found over the last -- well, it's more than 15 years -- but in terms of what my career has been. Whenever there is confusion, there was always opportunity. And I think this Company will always look for those types of opportunities. They will continue to exist.

Great, thank you.

Jason Arnold with RBC Capital Markets.

First, I want to say, Stewart, huge congratulations on your retirement. I will look for to hearing about your plans, enjoying your spare time. In terms of questions, just obviously, we have seen a lot of volatility in interest rates here in the past quarter. I was just wondering if you could update us on your macro views on the rate end of the equation.

Well, one, we try to run the Company so that we are not dependent on having the best call in the Fed. Because it's not clear that we would. It seems to us that the Fed truly doesn't know what they are going to do, or when they are going to do it exactly. Therefore, I don't know how we would have that view.

Has fiscal policy been irresponsive to the economy, I think the Fed sees this as a world of under consumption, where you have a risk of debt-driven deflation, they probably would have tapered. But they really -- I think the Fed really doesn't have the flexibility to tinker with their very accommodative monetary policy until there is a rational fiscal policy, which unfortunately is not happening.

So we don't have a strong call. We do think this trend of under consumption or oversupply is a longer-term trend. And while the Fed may tinker on the margins on QE, it is going to remain -- monetary policy is to remain very calm, as far out as we can see.

Okay, great. Thanks for the color.

And at this time I'm showing no questions in queue.

Well, thank you. Thank you very much for the personal remarks you folks have made relative to me. Again, I have so enjoyed working with all of you. And maybe our paths will cross again. Having said that, I know the Company is looking forward to the next conference call, which will be relative to our fourth-quarter 2013 earnings call. And again, thank you very much.

Ladies and gentlemen, this conference will be available for replay starting today at noon, going through February 4, 2014, at midnight. You may access the AT&T teleconference replay system at any time by dialing 1-800-475-6701, and entering the access code 307244. For international participants, the number is 1-320-365-3844.

Again, those numbers are 1-800-475-6701 and 1-320-365-3844, with an access code of 307244. That does conclude your conference call for today. Thank you for using AT&T executive teleconference service. You may now disconnect.