MFA Financial Inc (MFA) 2013 Q4 法說會逐字稿

Ladies and gentlemen, thank you for standing by, and welcome to the MFA Financial, Inc. fourth quarter 2013 earnings call. (Operator Instructions) And, as a reminder, this conference is being recorded.

I'll now turn the conference over to 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 Management's beliefs, expectations and assumptions as to MFA's future performance and operation. When used, statements that are not historical in nature, including those containing words, such as will, believes, expects, anticipates, estimates, 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 those described in MFA's Annual Report on Form 10-K for the year ended December 31st, 2012, and quarterly reports on Form 10-Q for the quarters ended March 31st, June 30th, and September 30th, 2013, and other reports that it may file from time to time with the Securities and Exchange Commission. These risks, uncertainties and other factors could cause MSA'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 fourth quarter 2013 financial results. Thank you for your time.

I would now like to turn this call over to Bill Gorin, MFA's Chief Executive Officer.

Thank you, Danielle, and I want to welcome everyone on this snowy day to MFA's fourth quarter 2013 financial results webcast.

With me today are Craig Knutson, MFA's President, Gudmundur Kristjansson, Senior Vice President, Steve Yarad, CFO, and a number of other executives.

Now, despite variability to monetary policies, interest rates and prepayments, MFA generated consistently good results throughout the year and in the fourth quarter. This is due in large part to our focus on strong fundamental analysis when investing in credit sensitive residential mortgage assets.

To summarize the quarter, in the quarter we generated net income of $74.8 million or $0.20 per share. Book value per common share increased approximately 3% to $8.06 as of December 31, from $7.85 at the end of third quarter. This growth was due primarily to Non-Agency MBS price appreciation.

In the fourth quarter we again transferred a sizable amount, in this case $47 million, from credit reserve to accretable discounts, bringing the total transferred in the year to approximately $208 million. Due to the increases in accretable discount and to changes in the forward curve the loss adjusted yield in our Non-Agency MBS portfolio increased to 7.77% in the fourth quarter.

Now, hopefully, you all have access to the webcast, turning to slide 3, you will see, as I mentioned, net income per common share was $0.20, the dividend was $0.20, and the estimated taxable income per common share was also $0.20. With that dividend we end the year approximately $0.17 per share undistributed in terms of taxable income distributions.

Turning to page 4, which I think is a pretty unique metric for our Company relative to all financial companies throughout the year. Despite change in interest rates and prepayment speeds our key metrics remained generally consistent throughout 2013. Spreads started the year at 2.32%, ended the year at 2.34% with not much variation in between. Debt to equity ratio started the year at 3.1 times, ended the year at 3.0 times, again, without much variation during the year. In general, throughout the year, through all four quarters over 55% of MFA's assets remained in Agency hopeful assets, with the remainder in Non-Agency MBS.

Turning to page 5, you see, as I mentioned, book value started the quarter at $7.85, grew to $8.06, primarily due to changes in the value of Non-Agency MBS.

Page 6, and hopefully this will be the last time we need to have a separate slide on taxable income in our earnings call. We've now distributed to stockholders an amount equal to all taxable income for years prior to 2013. We currently estimate that for 2013 re-taxable income was approximately $374 million, while distributions not attributed to prior years was approximately $313 million. Now we have until the filing of our 2013 tax return, which should be around September of 2014, to declare the distribution of any re-taxable income not previously distributed.

Again, talking about consistency, we ended the quarter with our interest rate sensitivity as measured by net duration below one. Gudmundur Kristjansson will give more detail about the interest sensitivity of MFA's assets and our distribution.

Thank you, Bill. On slide 7 we show the duration of our assets and hedging instruments. In the top part of the table we display the duration of our Agency and Non-Agency assets broken down into buckets by coupon resets. We estimate that the total duration of our assets is 2.1.

And in the bottom half of the table we summarize the duration of our hedging instruments, which currently are [integrated] swaps with maturities of up to 10 years. The duration of our hedging instrument is negative 3.7. When we combine durations of assets and hedging instruments we estimate that our net duration is [0.09].

Although net duration is a good indicator of interest rate risk, it is also important to understand that extension risk in our portfolio is limited. This is because 70% of our Non-Agency securities are hybrid arms and 60% of our Agency securities are hybrid arms, while the rest of our Agency MBS securities have 15 years or less to maturity.

These structural features limit the extension rates of our mortgage backed securities. Because of the limited extension risk in our portfolio we did not need to add interest rate hedges in the fourth quarter and are comfortable with our interest rate risk as it is.

And, with that, I'll turn the slides over to Bill, again.

Great, thanks Gudmundur. And now Craig Knutson will provide some detail as to the improved housing market metrics and, more importantly, the impact on MFA's portfolio.

So, thanks, Bill. Turning to slide 8, we continue to see improvement in the credit fundamentals of the underlying mortgages in the Non-Agency portfolio, and the key metric that we focus on here is the loan-to-value ratio or the LTV. Now, as I'm sure you know, the loan to value ratio is a fraction, where the numerator is the loan amount and the denominator is the property value, and so we have two moving parts here. The first is many of the loans, in fact, more than half of the loans in the portfolio are now amortizing, so as loans amortize, as homeowners make monthly payments of principal and interest they're lowering their loan amount, that amortization lowers the loan amount.

So what does that do? Well, that lowers the numerator of that fraction, thereby lowering the fraction, lowering the LTV. But in addition we also have home price depreciation, so the property values are increasing in value, so that increases at the same time the denominator of that fraction, which again further lowers that loan to value ratio.

So why do we care about loan to value ratio? Well, the lower the loan to value ratio, the fewer defaults that we would expect to occur in the future, and in addition we would expect lower loss severities on loans that do default.

So, as a result, again primarily of lower LTVs we have lowered our estimate of future losses in the Non-Agency portfolio. As Bill said previously, in the fourth quarter we transferred $47 million from our credit reserve to accretable discount, which is for the year over $200 million that we moved from the credit reserve to accretable discount.

We move to slide 9, so credit reserve reductions, what happens when we move money from our credit reserve to accretable discount? Well, what I'll point out is the accounting here is different than it would be for instance loan loss reserves for a bank. When a bank changes their loan loss reserves that change flows directly through their income statement and, therefore, through retained earnings.

For us, the treatment is different. When we reduce our credit reserve and increase our accretable discount we increase the yield on those assets, on those Non-Agency mortgages, we increase those yields prospectively. So it increases our future earnings as it increases the yield on those securities. So we moved $207 million during the year 2013, that $207 million will be recognized as additional income over the remaining life of those assets.

Now just to give you some perspective, the total credit reserve transfer over 2012 and 2013, again, if we freeze all the other elements, and there are other moving parts, prepayment speeds, forward curve and so forth, but if we just isolate the affect of the credit reserve reductions that is probably increasing the Non-Agency security yields by approximately 70 basis points, but again this is prospective, it's future earnings as we recognize that over time over the remaining life of those assets.

Finally, I'll note that the credit reserve reductions do not impact our book value. Again, this is different than it would be for a bank. Our book value is determined solely by the fair market value of those securities. So, again, these changes do not directly impact book value.

If we turn to slide 10, so slide 10, two graphs here. The one on the left is the average Non-Agency LTVs. So I'd call your attention to the gray line, you can see in the gray line that the average LTV of delinquent loans in the portfolio has declined from the beginning of 2012 to date from about 120% to less than 90%. So, again, this is average LTV, but that's very significant because those lower LTVs should lead to fewer defaults and lower loss severities on loans that do default.

Now while this average LTV and the lower average LTV is certainly a good thing, it's only one element. And the graph on the right illustrates another very important point. So if we look at average LTV, again, lower average LTV is good, but a 65% LTV loan when averaged with a 125% LTV loan may show an average of 95%, but the fact is that 125% LTV loan is still very much at risk of default.

So the graph on the right shows the percentage of loans with LTVs that are over 100%, they're still over 100%, and you can see in the beginning of 2012 probably two-thirds of the delinquent loans in the portfolio had LTVs over 100, and now that number is less than a third, it's down to almost 25%.

And then, finally, I'd call your attention to the orange line on the right graph, so these are the loans that are current in the portfolio, these loans are not delinquent. And, again, some of those loans we expect will default in the future, but as those LTVs and as the percentage of those with LTVs greater than 100% decline, that's gone from 50% down to about 15%. Those loans are certainly at less risk of default and, again, if they do default we would expect lower loss severities as they have lower LTVs.

Moving to slide 11, slide 11 and 12 show the two largest geographic concentrations. So slide 11 is California, which comprises approximately 46% of the portfolio, and you can see the key counties where we have concentrations of loans. And, in general, those home price appreciation numbers for the year are up 20% or more, which again speaks to the LTV.

Slide 12, which is Florida, again not up as much as California, but that's about 8% of the portfolio, and there are double-digit increases, more than 10% in all those key counties where we have concentrations.

So good news for the two largest states, but I would also note in the middle of a blizzard in New York both those states are much warmer than we are here today.

Slide 13, so this is another key metric, so we look at the transition rate. So these are loans that are current today, at what rate are loans that are current today transitioning to delinquent? And you can see that after peaking in early 2009 that the transition rate has declined, we're back to almost early 2008 levels of loans transitioning from current to delinquent. So, again, another very good sign for the credit fundamentals.

And, finally, slide 14, so this just shows how we account for these securities. So if you look on the left-hand side you can see the blue part, the purchase price is 73%, so think of that as a price of 73%, so average purchase price is 73% of par. Now, therefore, there was a discount of 27%, in this case if you take out OTI, 26% was our net purchase discount.

So what do we do with that purchase discount? Well, if you look over on the right, you can see that the majority, the vast majority of that purchase discount is still held in a credit reserve. It's that purple portion at the bottom, the accretable discount, that's the portion that we accrete into income along with coupon income. The credit reserve, which is still a little bit over a billion dollars, is not accreted into income.

So the other way to look at this is our purchase price, over on the left, is 73% of par, and because our credit reserve represents 19% of the face amount of these securities then, in essence, we're assuming in those yields that we get back $0.81 on the dollar, so we paid $0.73 on the dollar for these securities and we're expecting to get back $0.81.

And, with that, I will turn the call back over to Bill.

Thanks. Operator, we're available to take any questions.

Thank you. (Operator Instructions)

[Dan Altar] with FDR.

Hey, thanks, good morning, everyone. Greetings from a snowy Washington, D.C. winter wonderland.

Thanks. Glad to have you on the call.

Yes, we made it. I was wondering if you could talk a little about the Agency portfolio? Clearly, the prepayment speeds were down quite a bit, which probably helps the overall asset yields this quarter. Where are we looking around for first quarter so far? My sense is that may have been maybe a better than normalized or a better than expected number, per se?

Hi, Dan this is Gudmundur. You're right, prepayments dropped a lot in the fourth quarter. They came down by about 30%. And in the first quarter speeds went down a little bit, and I expect that the first quarter to come in a little bit lower than the fourth quarter. Looking forward on that, I would caution, saying that right now we're at the seasonal low in terms of turnover and CPRs, so for the foreseeable future, I mean even next quarter it'll probably be a little bit lower, but then they should pick up as we move into the year.

Okay, yes, that makes sense. And switching to a little bit of a different topic, you did increase the buyback authorization just recently, but it didn't look like you really bought back any shares in this quarter or at least in the fourth quarter. How do you think about that moving forward, was the authorization just to increase the flexibility or is there actually like an intent to buy back the stuff?

Good questions, Dan. Let me give you the longer answer. So, as you know, in 2013 we distributed special dividends of about $300 million to investors, so we were very shareholder friendly and, basically, I mean that was capital and you couldn't use for a share buyback.

Strategically, based on the consistency of our earnings, I would say our strategy is working. In no way are we competing with a nonprofit maximizing government entity that can print money, so we're very happy with our strategy. I almost call our strategy ground and pound versus other people that might be somewhat defensive of this time in their strategy.

That being said, in the fourth quarter some days the discounts were very intriguing, and we actually did purchase 2.1 million shares at an average price of $7.20 in the fourth quarter. So it wasn't just to give us the flexibility, we implemented that.

Okay, that's really interesting, that's good. And then just one quick one on the hedge portfolio, it looks like you got rid of the TBA shorts that were in last quarter, was that just maybe like an experiment or just trying to see how those would work or is there anything fundamentally wrong with them that weren't working or basically why did we get rid of the TBA shorts?

Yes, the TBA shorts was not experiment, it was could there be unknown uncertainty as the market accepted the concept of tapering? And we think we're sort of through that, so it was just extra protection for what seemed like an extra volatile, less predictable period of time.

Okay, got you. Thanks so much, and enjoy the snow.

Thanks, you, too.

Douglas Harter with Credit Suisse.

Thanks. You gave -- Craig, you gave some good color around the impact of the switching of, the releasing of the reserves from -- into basically earnings over time. Is -- can you talk about how much of the impact you also saw from yield curve shifts in the Non-Agency yield this quarter, and sort of what we've seen in the first quarter, if we should expect some reversal of that?

Sure. To be honest, I think the affect of the forward curve, I don't have the exact number, Doug, but the affect of the forward curve was not that significant in the fourth quarter yields, that was primarily due to credit reserve adjustment. So, yes, there is some sensitivity to forward curve, but like I say primarily it's been the credit reserve that's driven that.

And did we see more of it in the -- more of a pick up in yield in the fourth quarter just from the cumulative impact of everything that you had talked about and the fact that you'd been doing it sort of over the course of the year and it just kind of builds to the point where it gets to be more meaningful?

Exactly, we tried to isolate that and that 70 basis point yield number is an attempt to strip out all the other affects, and that's the cumulative affect of 2012 and 2013 of those credit reserve adjustments.

Got it. Okay, so there's nothing, there's no reason to think just because what's happened in the yield curve kind of quarter to date that that yield shouldn't be sustainable, all things being equal?

Again, tomorrow is a new day, but I would agree.

Okay, all right, I appreciate that, Craig. Thank you.

Steve Delaney with JMP Securities.

Good morning, everybody, and congratulations on yet another very solid quarter. I wanted to -- thanks for the clarity on -- when you had some unique things last year with taxable EPS with that change and accounting that for tax, so it's helpful to have this disclosure, the $61 million and the $0.17.

Bill, I was wondering last year you actually paid two specials. I get the impression from the way you presented this that since you have until September to deal with this in terms of distribution that you'll just use this amount to sort of maintain and contribute to the 2014 dividends and possibly have a carryover at the end of next year into 2015, am I reading you right on that?

Well, one, the Board sets the dividend policy. We do not.

Yes.

But I think what you're saying mathematically makes sense, that just keeping the regular dividend clearly will be Board distributed for 2013 by September, and we're not -- it doesn't appear as if that shortfall would be growing.

Got it, got it. And I guess shifting to the book value, nice increase there. Seeing some of the earlier reports there seemed to be comments and some of the book value moves that actually exceeded expectations, it seemed to be focused on the seasoned hybrids, and we kind of can see what the 15-years are doing, sort of down a point-and-a-half, maybe in the fourth quarter up a point-and-a-half now, sort of a round trip. But it seems like the seasoned hybrid arms have materially outperformed, say, new hybrids or 15-years, and I was wondering, Gudmundur, if you could give us any color from what you're seeing kind of within your portfolio how the hybrids, your more seasoned hybrids have performed in 4Q and 1Q?

Yes, I mean in general short cash flows, whether that's seasoned hybrids or 10-year or seasoned 15-years or high coupon 15-years for that matter, did pretty well in the fourth quarter and continue to do well in the first quarter. So I would definitely agree with that.

Okay, so you're seeing continued -- whatever happened in the fourth quarter you're seeing some pull through and continued strength there, kind of with the broader market?

Absolutely, continue to perform well, not necessarily that they're increasing in price, but whatever they gained, the momentum that they've gained has kind of stayed in the first quarter.

Steve, what you're saying, we agree with. There seems to be great demand in a short asset.

Yes.

Yes, at this point in the right cycle I think people (inaudible) the banks just want to be at the short end. And, Bill, just one last thing and I'll hop off, this is big picture -- just looking at 2014 it seems like there are a lot of emerging kind of opportunities for a broad residential mortgage REIT as opposed to sort of a more narrow Agency REIT. It gives you guys a lot of flexibility. Just curious if you have any thoughts about what might be some of the best opportunities for you in the broader residential credit area as you look out 2014? Thanks.

Right, no, thanks. And, look, we appreciate the opportunity to share our strategy in this forum, so thank you for that. So approximately five, six years ago we adjusted the strategies from Agency assets, which basically are dependent upon interest rate sensitivity and prepay sensitivity, to Non-Agency assets, especially when they cheapened. We believe that the assets don't move, the movements are not highly correlated, that that is shown in the consistency of our returns where I would say over the year maybe the yield on the Agency portfolio was trended down, while the yield on the Non-Agency portfolio has trended up.

I would say our competitive advantage is credit analysis, so but we're completely supportive of people widening beyond pure Agencies, because again you're dealing with a noneconomic competitor in the government. We've looked at mortgage servicing rates, for example, but there we think that asset, while it's great it's not correlated with Agencies, it is a negative correlation, is dependent upon prepays, which are therefore dependent upon interest rates. And we're not saying that we're the best predictor of interest rates or prepays or the weather, but I think we've done very well by focusing on the credit performance of mortgage assets of a certain vintage and we continue to find a lot of opportunities there, Steve. So I think you're going to see more of the same in 2014.

All right, appreciate that color, and good job, guys. Thank you.

Thanks.

[Joel Hough] with Wells Fargo.

Yes, thanks, and good morning. And, again, thanks for the color on the strategy. It seems like strategies amongst your peers are evolving, and then you guys have kind of stuck to your knitting.

Just back on the yield on interest earning assets, it was up 21 basis points in the fourth quarter. I know you disclosed the 70 basis point cumulative benefit, I guess, from the credit reserve releases, but in the fourth quarter with that jump, I mean how much of that was attributed to the credit reserve release specifically and how much was just due to lower prepayments fees on Agency MBS?

All right, so we could tell you the change in yield on the Agencies, and by the way we'll be filing the 10-K soon today. But I believe the Agency yield went from 213 in the third quarter to 237 in the fourth quarter, and I think that's where you're seeing the impact of the slowdown in prepayment speeds. And the Non-Agency yield, as we've already said, went up from 733 to 777, the majority of that is due to improved credit and some portion of it is due to change in the forward curves.

By the way, I want to point out the change in the forward curve is really only significant for the adjustable rate Non-Agencies, the fixed rate Non-Agencies are not impacted, which is why we're comfortable saying most of the change is due to change in credit assumptions.

Okay, good, thanks for that. And then just the last question, I guess, is more on a strategy question around the Agency, obviously, the seasoned arm strategy has been a good one. If we assume the Fed continues to taper and eventually gets out of the market, some believe that could cause OAS to widen out significantly for fixed rate, 15 to 30-year product. If we were to see that, I mean hypothetically what would MFA's approach be if they were good risk adjusted returns kind of after the market had moved? Or would just say the strategy is working for us, we're not really interested in moving out on the curve with respect to the Agency opportunity?

So your question is would we ever consider going beyond the 15-year Agency?

Yes, basically.

So the answer would be we'd be reluctant to, but with the appropriate hedging if the spreads were appropriate we would consider it definitely.

Okay, thank you, guys, very much.

Thank you.

Daniel Furtado with Jefferies.

Good morning, everybody, thank you for the opportunity.

Sure.

The first question, I guess you know that Non-Agency continues to perform very well. Congratulations there, everyone, but -- and here's the but -- at the same time it continues to run off pretty steadily, we all know that the universe of Non-Agencies shrink, and I think you talked about this a little bit with Delaney's question. But, more specifically to this strategy, how active are you on the new purchase front today? And how do you see that market going forward?

Let me start with my usual speech, which is on every one of these calls the questions either are aren't we too early or aren't we too late? You might have heard this before, but that being said we're very active in the markets.

Yes, and, Dan, the market, while it is shrinking, I think it's still probably $700 billion or so and there have probably been in new supply or bonds being traded in the last month it's several billion dollars. And we've probably put, so far this year we've probably put $200 million to work in Non-Agencies. So it's not hard to reinvest runoff, at all. A runoff on Non-Agencies typically runs about $60 million, very rough numbers, $60 million a month, and about half of that pays down resecuritizations. So it leaves us about $30 million to reinvest, which again is not very difficult. So, yes, at some point, at some year there will certainly be less supply and less opportunity, but at least for now we don't feel like we're constrained, at all.

Got you. And so I guess not looking for guidance, but despite the contraction on say the principal on this portfolio, it wouldn't be out of the question if we were to see the principal balance on this piece grow over the course of time even from here?

Well, as Bill said, about 55% of our assets are hopeful Agencies, and as you know that's a limiting factor. So I think that's been fairly consistent for the last year or so.

Understood there. And then speaking of the re-remic bonds, what are your thoughts about potentially leveraging the next step down as the AAAs burn off and the next sequential bond then becomes a defacto AAA, is that something you would look to do? And, if so, about how far out do you think we are before that becomes a viable option for you guys?

Well, it's a good question, Dan. We've actually already done that. On the first securitization that we did, the DMSI deal that we did back in 2010 that A1 paid off, and we've sold the A2 and actually the A3 on that, as well. So that was a remic transaction, which makes it easier to sell those securities. Several of the other deals were debt for tax structures and not to get too complicated here, but those we'd have to wait until the senior bond pays off to then sell a second bond because we can only have one sold at a time.

But another option on those Non-Agencies would be if the senior bonds pay off and we are now the holder of all remaining pieces, we could also potentially collapse those deals and get the underlying bonds back. So we certainly have a lot of flexibility.

Like I say, it's a good question and something that we talk about a lot. Our guess is that the second two deals that we did are probably going to pay off those senior bonds, will probably pay off in the middle and then the latter part of this year. So, again, we have a lot of flexibility and a lot of different options open to us there.

Great, thanks for the color there, Craig. Have a great day.

Thanks, Dan, you, too.

Rich Shane with JPMorgan.

Hey, guys, thanks for taking my questions this morning. A couple things. One, I guess I'd like to take a slightly different take on the question that Steve Delaney had asked related to the rollover dividend. Going back to some comments that Bill had made last year, which played out pretty consistently with his expectations, we've seen the convergence between GAAP and tax. Tax actually fell $0.03 this quarter and is now right in line with the dividend. I'm curious if you think that that trend is going to continue? And then, obviously, we see that $0.17 rollover as basically cushion to continue to make re-taxable distributions.

Good question. So, one, in order to forecast taxable income we'd have to forecast a number of things, which we're not prepared to do, but the trends you point out is correct. And the way we look at it taxable income sets your minimum distribution, in no way does it set your maximum distribution, and you had some strange tax results because we purchased senior most pieces at very large discounts, which required us to make these large distributions over the last year. So we don't see taxable income as a ceiling, and I think going forward we certainly would focus on taxable income and economic income.

Got it. Okay, that's helpful. And then the second question is this, and I'm surprised this didn't come up, this question, and maybe we're missing something, but historically one of the metrics that you guys have focused on is core income and there's always been a reconciliation associated with GAAP to core. That didn't appear this quarter, I'm curious what the rationale to that was and actually there's a nuance that we always pulled a weighted average share count from the core reconciliation, so we don't have a weighted, at least we've been unable to find the correct weighted average share count for the quarter?

Rich, it's Steve Yarad. Thanks for your question on that. The reason that we started the practice of disclosing core income a number of years ago, it had a lot to do with the fact that the accounting we had to do for linked transactions. And you might recall if you go back a year or two that the impact of linked transactions on our financial statements was fairly significant. Over time those linked transactions have runoff and we have much less of that in our portfolio now. And so the impact on our income statement of the mark-to-market on the linked transactions is now virtually diminimous and has really no impact on using that on Agency portfolio or our disclosed leverage numbers.

So we've taken the decision in this quarter to take out the GAAP to core reconciliation, which in the past was the past heavily by linked transactions. You'll notice in the press release we still listed out the gains on sales of the securities so you can pretty easily back into what the core number would have been for the quarter.

In terms of your question on the weighted average shares, when the K gets filed shortly after this call, a little later today, you'll see in the EPS footnote the weighted average shares and that'll help you do the calculations you need on that front.

Sure. I'm assuming that just not including it the way you did this quarter was sort of a disconnect between removing that slide. I would ask if you could going forward just to throw it in the press release so that we can dial in our numbers right away?

Sure.

Awesome. Thank you, guys, very much.

(Operator Instructions)

Chris Donat with Sandler O'Neill.

Hi, good morning, it's Chris Donat. Thanks for taking my question. I had one question I wanted to clarify on the timing of how you address the credit reserve and accretable discount, am I correct that you look at a third of the portfolio every quarter, is that how you do it? And then for your LTV calculations, when you're looking at the home values is that something you're looking at real-time or is it on a similar sort of cadence that one-third every quarter for the home values?

So, Chris, we do look at the credit reserve every quarter. It's not exactly a third of the securities every single quarter, but there's no event where we don't look at a security at least every nine months. So we might, our credit reserve worth might encompass 45% or even 50% of the portfolio in a particular quarter, but again there are no bonds that go unlooked at for more than three quarters. I'm sorry, the second part of your question?

Just on the LTV calculation, the value part of that, you're looking at that basically real-time or monthly, right, you're not assessing home values on some sort of lag basis are you or what has that been?

We look at that monthly, but again we don't have BPOs on these properties so we're doing zip code level work.

Right.

But zip code level work is typically median home price in that zip code, so to the extent that the specific mortgage is a lot less than the median home price or a lot greater then obviously we're making some assumption that their prices move in line with the median prices, but those are pretty much real-time and we get those monthly.

Okay, and then just asking on the share repurchases, looking at the average price for the quarter it looked like a 10% discount to book is where you pulled the trigger, is that a fair way to think about where you might be active in the future or does it depend on a bunch of other factors?

Yes, it depends on a number of factors.

Okay, just thought I'd try to quantify it, but thanks, appreciate it.

(Operator Instructions)

And, Mr. Gorin, we have no one else in queue.

Thank you very much, Operator, and thanks, everyone, for participating. We hope everyone gets home safely today. Thanks a lot.

Thank you. And, ladies and gentlemen, that does conclude our conference for today. Thank you for your participation and for using AT&T Executive Teleconference. You may now disconnect.