MGIC Investment Corp (MTG) 2011 Q2 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the MGIC Investment Corporation second-quarter earnings call. (Operator Instructions). As a reminder, this conference is being recorded. I would now like to turn the call over to your host, Mike Zimmerman. Sir, you may begin.

Thanks, Stephanie. Good morning and thank you for joining us this morning and for your interest in MGIC Investment Corporation.

Joining me on the call today to discuss the results for the second quarter of 2011 are Chairman and CEO Curt Culver, Executive Vice President and CFO Mike Lauer, and Executive Vice President of Risk Management Larry Pierzchalski.

I want to remind all participants that our earnings releases this morning, which may be accessed on MGIC's website, which is located at MTG.MGIC.com under investor information, includes additional information about the Company's quarterly results that we will refer to during the call, and include certain non-GAAP financial measures. As we have indicated in this morning's press release, we have posted on our website the supplemental information containing characteristics of our primary risk in force and flow new insurance written, as well as other information we think you will find valuable.

During the course of this call, we may make comments about our expectations of the future. Actual results could differ materially from those contained in these forward-looking statements. Additional information about those factors that could cause actual results to differ materially from those discussed in the call are contained in the quarterly earnings release.

If the Company makes any forward-looking statements, we are not undertaking an obligation to update those statements in the future in light of subsequent development. Further, no interested party should rely on the fact that such guidance or forward-looking statements are current at any time other than the time of this call or the issuance of the press release.

With that, let me turn the call over to Curt.

Thanks, Mike. Good morning.

In the second quarter, we reported a net loss of $151.7 million. The loss reflects the continued impact of the weak housing market and high unemployment, as well as the typical seasonal patterns regarding new notices and cures. The net loss for the first six months of 2011 was $185 million, compared with a loss of $126 million last year.

New insurance written in the second quarter was $3.1 billion, compared to $2.7 billion in the second quarter of 2010. An additional $583 million of HARP refinance transactions was also completed during the quarter.

We estimate the private mortgage insurance industry's market share to be approximately 5.5%, with our own market share within our industry estimated at 24%.

We expect new insurance written in 2011 to be in the $12 billion to $14 billion range.

Persistency in the quarter was 83.3%, which was flat to last quarter, and total primary insurance in force declined to 184.4 -- $182.4 billion, down from $187 billion last quarter and $202 billion one year ago.

Losses incurred in the second quarter were $459.6 million, which was up from $320 million last year. During the quarter, new delinquent notices exceeded [peers] before rescissions and denials by 4,100 loans, compared to a decrease of 2,400 loans in the first quarter. This net increase in delinquencies was the primary driver that caused the losses incurred to increase during the quarter.

Primary loss reserves now total $4.5 billion, or an average of $24,416 per delinquent primary loan. The average reserve for delinquent loan reflects the continued mix shift towards the older delinquency categories.

During the quarter, 6,850 loans, or 19% of the primary cures, were reported as modifications, versus 700 -- 7,554 modifications in the first quarter, or 16.5% of the cures. This total was comprised of 4,741 HAMP modifications and 2,109 of non-HAMP modifications. As of June 30, 14,300 of the primary delinquent inventory were reported to us as being active in the HAMP trial process.

The redefault rate of recent modifications continues to perform better than our historical experience, which we believe is primarily due to the fact that borrowers' monthly payments are lowered by 30% to 40%.

Paid claims in the quarter were $818 million versus $687 million last quarter. The increase in paids was primarily due to an increase of $98 million in the level of pool claim payments. During the quarter, we increased the resources allocated to pay pool claims in order to minimize the amount of interest we are obligated to pay. We expect the accelerated pace of pool paids could remain elevated in the third quarter before returning to lower levels.

Within the primary delinquent inventory, at the end of the quarter there were approximately 14,500 claims that have been received but not yet paid, which is down from 19,724 loans a year ago. Likewise, within the pool of delinquent inventory, there were approximately 1,836 claims that have been received but not yet paid, which is down from 2,023 claims a year ago. Assuming normal foreclosure patterns, we believe that the second quarter marks the peak quarter of claim payments and that the overall level of total claim payments will now begin to decline.

As expected, the total estimated primary and pool loss mitigation savings continued to decline, and totaled $424 million with $170 million of that in rescissions and denials, compared to last quarter's savings of $513 million where we had $237 million in rescissions and denials. While we have seen a marginal pick-up in the level of claim denials over the last year, rescissions continue to be the substantial majority of our total.

The average premium yield was 61.6 basis points, up marginally from 61 basis points last quarter. Underwriting and other expenses were $54 million, which is flat to last year. Cash on investments totaled $7.8 billion as of June 30, reflecting the higher level of claim payments, the declining premium from the smaller in-force book, and the repurchase of 55 million of power of the 2015 senior note. MGIC's risk to capital increased to 20.4 to 1 at the end of June, due to the higher level of losses incurred -- we recorded in the quarter.

Regarding Washington activity, the deadline for QRM comments has been extended to August 1. This delay reflects the fact that many legislators, in fact in excess of 300, as well as community groups and housing industry participants, have voiced tremendous concerns that the rule, especially regarding the level of down payment, could have a very detrimental impact on the housing market, could jeopardize an economic recovery, and would set back the government's stated role of reducing its footprint in housing.

Both MGIC and MICo will be providing a response to the NPR. Both responses will highlight the significant issues contained in the current proposal, and will provide an alternative that we believe promotes prudent lending and enables prospective borrowers to access affordable housing options while reducing the government's role in housing.

The GSE reform debate also continues, and we expect that it will be quite some time, in fact after the 2012 election, before there is any meaningful clarity provided on the topic.

The QRM definition and GSE reform are linked together and should be addressed in a coordinated manner along with further FHA changes, if the administration and Congress goals are to be realized. So while there is much to play out in Washington, we continue to be encouraged by the central theme of shifting lending activity back to the private sector.

With that, operator, let's take questions.

(Operator Instructions). Steve Stelmach.

Good morning. Just one quick point of clarification, and then maybe my standard questions. Curt, did you say 2Q you think you witnessed the peak in claims for the year? Or did I misunderstand that?

No, that was correct. Assuming normal foreclosure patterns continue, we do expect it to be the peak.

Because I typically assume the first half is better in terms of delinquency experience, but you're saying that foreclosure activity tends to be more front-end loaded?

This is claims paid.

Right, no. Exactly. So claims paid typically slow down towards the end of the year. That's something I did not (multiple speakers)

No, no, not at all.

It's a function of the delinquency inventory and where they are in the timing.

What we've talked about are two different things. Noticed inventory traditionally goes down in the first half and then builds in the second half. What we're talking about now is the resolution of those notices, and that we believe we've reached the peak paid claim levels in this quarter. And as Curt mentioned, the pool claims, we think, will be about the same level, possibly lower, in the third quarter but then trend down. And overall, we believe paid levels will reduce from this point. All subject to --

(Multiple speakers) housing market and everything else. In terms of -- so you're not just talking about seasonality, you're saying this is sort of the high watermark in paid claims.

Correct.

That's very interesting. And then, just, going on what you guys mentioned in terms of the alternative QRM definition, any way you guys can do within the construct of the GSE charter that could get that penetration up further or that percentage higher? In terms of (multiple speakers)

No. I mean, the charter is the charter right now.

Right, but there's nothing in terms of, like, pricing or some other mechanism you guys could use that could get that penetration higher?

I think what would get the penetration higher is if the GSEs would drop their loan-level fees and their market adjustment fees on both purchased and especially refinanced loans.

But, Steve, the alternative QRM is a 90% -- the definition is a 90% LTV. So while the penetration of those 90s may be increasing, the 95 LTV, that's really what we're talking about with that number.

That's right.

But the definition would have to change in order to get a higher concentration of loans post the QRM definition.

Right. Okay. And then, lastly, in new insurance written, other than a better penetration rate, there's nothing there that could get that going higher, even absent sort of QRM? Are there any plans to reprice to change any credit underwriting criteria?

We continue to underwrite in a prudent manner. Again, my comment earlier would have been the one that's more appropriate here, is it really is reflective more of the GSE's pricing policies of adding fees onto purchase and refinances that if they'd drop, which I would submit they should, given the limited impact it has on the benefit to them and could easily help the housing recovery if they would drop that, it would make the conventional and privately-insured loan much more aggressive relative to FHA.

Got it. Okay, guys, I'll hop back into queue. Thanks.

Douglas Harter.

Thanks. I was wondering if you could update us on the amount of parent company cash, and then also talk about your decision to buy back some of your debt during the quarter.

At the end of the quarter, we had $823 million at the holding company. During the quarter, we were -- I guess, had an opportunity to buy a pretty good position, sizable position. Someone came to us that wanted to buy it at a discount, and we took advantage of that opportunity. So we bought about $52 million, cost of $52 million, about $55 million of par debt back, so we saved somewhere around $16 million in discount and the future interest payments.

Could you talk about your desire sort of to do that in the future, if (multiple speakers)

It would all be subject to a number of things, if we had opportunities, what the prices would be, and it's something that we would look at. But obviously, this was a one-off at that time. We don't have anything else at this time to consider.

And when thinking about that $823 million, is there any plans to put any of that down into the mortgage subsidiary or would the plans be to keep it up at the HoldCo?

It will all be subject to a need -- if we had a need at the writing company, and if so, how much.

So it's available to drop down in the event we would need some, and we also have -- if you recall, we did drop some money down into a new subsidiary that would also be available to write business in the event that we needed additional, if you will, volume with respect to risk of capital.

So we have, I guess, a two-pronged attach to that. We have capital available for the writing company; we also have the opportunity, given approvals from state insurance commissioners and Freddie and Fannie, to utilize a MIC subsidiary. So --

Thank you. And (multiple speakers) just want to say thanks for providing the disclosure of the breakdown of the cures. That's very helpful.

Sam Martini.

Good morning. Just on that last point that Doug made on the cures, can you just talk broadly about -- it looks like, round numbers, 15% to 20% of your late-stage bucket cures. It's harder to calc the 4 to 11 on a seasonal -- because you've got stock and flow numbers. I'm just curious your thoughts on how to use this data best, especially for the 4 to 11 bucket, what you think reasonable historical rates are, and if you could, in answering the question, go back beyond just Q1 2010. I don't know if you have that data handy, but I think this is helpful. I'm just trying to figure out exactly how to look at it all, given we don't have -- it's hard other than just to guesstimate the [roll] rates from bucket to bucket.

So if you could just talk about what you think reasonable historical -- what this data shows you for reasonable historical cures, especially for the 4 to 11. It looks like 15% to 20% for the 12 and longer, and if you could incorporate prior to March 2010, that would be helpful. Thanks.

You're right. On the historical data, if we go back several years, the majority of cure -- you would not see very many incremental cures after 12 months on a historical basis.

You still see some, but the level has grown because of the modifications that have been driven, as well as delays in reinstatements and people coming cleaner because of the curing on those.

So on the middle-stage bucket, I guess point being your cures happen earlier, historically. We're seeing those cures been delayed over the last few years because of all the foreclosure moratoriums, servicers catching -- working with borrowers, et cetera.

We don't have historical information past the first quarter on this call, of 2010. So to me, the best way to use this information is working backwards from the late-stage bucket, again I was assuming the cure out. You can calculate the number of implied -- that roll from each bucket to the next bucket. So on average, you'd have anywhere from 20% to -- say 20% to 40%, depending on the quarter and the level of modifications that would roll out of that 4 to 11 into the late-stage bucket. And somewhere in the 25% to 30% that would roll from that first-stage bucket into the middle bucket.

If I did all that math, Mike, what does it tell me that, maybe the last six quarters if we don't want to look prior to that, what does it say sort of a reasonable cure rate for the last six quarters for the 4 to 11 is? If you back out the roll into the late stage, what have we been running at roughly?

So if you look at it from on a beginning-inventory basis of that bucket, you're seeing somewhere around, on a quarterly basis, about 10% of those are curing, and again, I'd go back to the 25% of [tars] rolling because these are dynamic pools. We -- versus looking at each month's new notices that came in, and then tracking those for a separate 24-month period of time, if you will, a static pool type of approach.

So with this data, I mean, the only other way to provide that information would give you thousands of lines of delinquency data to do the exact roll rates. I would suggest you just use that implied quarterly roll rate and take the average of the last few quarters and use those trends going forward.

We do give the monthly data now that -- it doesn't break it out by aging category, but you can start to see the ins and the outs, at least at the aggregate level. But I would suggest for right now, you have to use the average of the last three to four quarters to reflect the more recent activity relative -- in particular, relative to modifications.

So if the economy picks up and employment starts to pick up, then that's going to affect the roll rate of those earlier-stage buckets. But other than that, it's really modifications and just the -- the data that's available as we presented it there.

If we have 39,000 loans in the 4 to 11 bucket that were shown to have cured in the last 12 months, that, you would say, works out to -- adjusting for the new coming in and the aged going into the late-stage bucket, that gets to, like, $0.10 a quarter, 12 cents -- I'm sorry, 12% a quarter, something like that?

If you use the average of the beginning and ending for the quarters and the average cures that came out of those in the quarter, yes. That seems to be an appropriate way to do it.

And you don't have this data handy for, say, 2009, 2008, 2007, something like that? This is (multiple speakers) that bucket also not have cured at this level? Or is this bucket more normal?

Sam, I don't have it available on the call. I don't know the answer to that question. Maybe that's something we can look (multiple speakers)

(Multiple speakers) look into, Sam.

And then disclose it either in our monthly -- with an update on our monthly information or on the next call, but I don't have that information available today.

Okay. Thank you, guys.

Mike Grondahl.

Yes. Three questions, guys. One, could you talk a little bit about the increase in severity? I think the average claim paid went from, like, 48,000 to 50,000.

And secondly, when you commented that you think claims paid peaked in the June quarter at $818 million, what type of decline do you expect?

And then, maybe lastly, just generally, cures kind of seemed to fall off or decline more abruptly this quarter. What do you think caused that, and what is kind of working in cures and what isn't? Thank you.

On an average claim paid, I guess I wouldn't read too much into it. Quarter to quarter, there will be some movement, but as we look at it within the flow channel, within the bulk channel, and through the geographies, you do get a quarter-to-quarter blip up or down. But it just seems to be normal movement quarter to quarter. So I -- we don't expect any change in the average severity at this point. So, again --

The only thing I'd say is there's a little bit higher California, Arizona, Nevada mix this quarter than last quarter. That's the only thing driving it.

Within those geographies, the average claim paid is relatively stable.

What was the other question? (Multiple speakers) the pace. I think, as Curt said, it will peak, but it will decline very gradually in part because the foreclosure process is taking much longer than in the past, so we can see that if things continue at this slow pace, it will be a slow decline over the next couple of years.

Secondly, in the quarter, Curt mentioned that 100 million of the increase was due to pool paids. And you'll note that we had increased reserves throughout the last several quarters in anticipation of that.

We had the notice inventory, we resolved the claims and paid them this quarter and mitigated some penalty interest. And then, the reserve is down accordingly. So those paids effectively came out of reserves, accordingly. And then, Curt mentioned that we anticipate the pool paids in the third quarter to be lower, and then trending lower again in the second -- I mean, the final quarter.

And then, Mike here, your last question was relative to the cures, I think the main -- the seasonality is a large driver when you look at historical information. Obviously you have ours for now a little over a year, but you can look back at MICo data where you've seen the net inflows or that cure ratio that MICo puts out as being under 100% fairly consistently throughout the second half, really beginning in April, May, and continuing for the second half of the year.

But also, year over year, you have a dramatic decline in the number of modifications that took place. Second quarter of last year, we had almost 16,000 modifications on the primary side versus the 7,000 that occurred this year. So a big driver on a year-over-year basis is the decline in the number of modifications.

Got you. Okay. Thank you.

Chris Gamaitoni.

Thanks for taking my call. Along with the delay in servicer issues getting to claims paid, do you have a weighted average timeline you think it will take to pay all the claims that are inherent in your book today?

We've got a 10-year forecast. The in-force will continue to draw new delinquencies over time, probably more so being driven by employment at this point going forward than home prices. But the in-force will play out over a long period of time here.

From the timeline perspective, I think -- there's other sources of information. For example, Lender Processing Services puts out monthly data, and the time from -- to complete a foreclosure has been extended several months as a result of all this.

When that engine restarts, I'd say your call is probably as good as ours. But that's going to -- traditionally, it was 12 to 15 months for a delinquency to mature to a claim payment, and it's clearly much longer than that now. When it starts to come back to more historical averages, we don't know.

Some of the delinquencies take over four years to cut -- get to a claim receive status, given what's happening on the foreclosure side.

What kind of factor in your reserving methodology would you apply to those --

(Multiple speakers). The longer that it's in the higher reserve factor, obviously.

So the longer it's in, the higher the propensity it's going to go to claim. Some of the new cure information that Mike Zimmerman spoke about a few minutes ago, one of the things, the point we wanted to make, was even some of those notices that are 12 or more months behind on their payments, a good portion of them still cure.

And that is due to modification, correct?

In part. Some come current. Some are -- they sell the [part], they get charged off, so some people do come current, so they may have been playing the game and push comes to shove, they come current.

And I guess, finally, have you had any conversations with the GSEs or regulators about additional modification options? President Obama has spoken about wanting to go back to the drawing board in housing, and it seems that given HAMP wasn't successful as originally designed, there may be something -- some other type of program that could be initiated beneficial to you.

There are discussions that are ongoing on that subject matter relative to refiing loans above 125% with the GSEs, and also looking at principal reduction type situations relative to delinquent loans. So, those discussions continue. So, what gets done from that, I have no idea.

Thank you for your time.

Matthew Howlett.

Thanks for taking my question. I'll wait for the Q on this, but if you have the incurred loss split breakdown early between the reserves incurred for the current year versus for the prior years. Do you have that handy?

No, we don't. That will be in the Q.

Got you. And then, just a question on -- you mentioned the aging of the delinquent inventory is causing that reserve build. Can you give us either the number of loans or the percentage of loans either for the full inventory or just for the 12-month past-due that are -- actually have a foreclosure filing on them?

Matt, first off, I want to correct the [thing] on the aging kind of comment. It's a mix of those. Again, as we set those reserve factors, assuming that we're right on those, the age that's the factor moves from one bucket to another bucket does not influence that.

So aging is not necessarily the appropriate lingo to use there versus the mix. There is a more percentage of that later-stage bucket, and that's what drove up that average reserve for a delinquent loan, a lot of that.

As far as foreclosure, I think of our total inventory -- I don't have it broken down to the late-stage bucket, but total inventory is right around 45%, 50% of the total delinquent inventory that some -- reported to us from the servicers and in some stage of foreclosure.

Okay, so total, 40%, 45%, okay. (Multiple speakers)

45% to 50% (multiple speakers) the total primary delinquent inventory.

Got you. That gets me to my next question there, there is this proposed settlement with the Attorney Generals and the major servicers here, and part of the settlement they're talking about is reducing this dual tracking between someone has a modification and a foreclosure. What can you tell us on that? Can we expect anything? If the settlement does go through, what could be the impact on your delinquent inventory or paid claims going down the road?

We are in part of those discussions. So, we can't comment on what stage of those discussions are and the impact.

From what's been reported publicly, at least, certainly would help our industry's portfolio relative to delinquent loans, but to the manner in which it does, again, without being part of those discussions, I really don't have any idea.

Is there any way to tell how many loans that are in foreclosure actually are in the modification process as well, or is it just that is still spotty on that?

It's still spotty. And all the best -- the reporting that we get is on the HAMP, and that's the 14,300.

Great. Thanks, guys.

I also wanted to clarify the statement on the percentage that's in the foreclosure bucket today, in some stage of foreclosure, is in the 50% to 55% range. A year ago, it was closer to 50%. (Multiple speakers) so just slightly higher than what I had said.

Thanks.

Jack Micenko.

Hi, guys. Most of my questions have been asked and answered, but since I'm in the queue still, just wondered if you could share a little bit -- you talked about providing an alternative with your response and the MICo response. Can you share with us -- you don't want to let the horse out of the barn, but can you share with us some of the general themes on what that response would be or that alternative would be?

Yes. What we have come back with is that under 15-G of the Dodd-Frank proposal, it allows alternatives.

First of all, we think the strongest reason is just to conform the definition of QRM to proposed standards, which we are going to -- indicate what those standards would be within our response, and within that, we would say a 90 -- a 5% down payment.

Mortgage insured should be the standard, and then we are giving a bunch of underwriting criteria along with that that we think in total makes it a strong credit decision. Or if that does not move the regulators, we also are talking about allowing an exemption from risk retention requirements for loans complying with these proposed standards that we are going to put forward.

And also, another alternative is that we will discuss is allowing a zero risk retention for low credit risk category of loans, again those complying with the proposed standard. And then, another alternative which we will discuss, the ability for regulators to get privately insured loans as part of this and private capital back into the equation is a treatment of private mortgage insured as a [long] form of risk retention. I mean, they talk about 5% risk retention, and the reality is mortgage insurers take anywhere from two to seven times more risk retention than that level, and as such we should be deemed a risk retainer.

So we have a number of alternatives that we explore within our response to the NPR, and from my standpoint and looking at the studies that have been done by Genworth, Promontory, and also Milliman for our industry, I think there are strong reasons why you should -- that the regulators have to conform the definition to our proposed standard.

Great. Thank you.

Mike Grasher.

Thank you. My questions were addressed. Thanks.

John Helmer.

Thanks for taking my question. I wanted to telescope out for a second and ask you how you as management think about intrinsic value?

I have been as a long-term value investor involved in the name for a few years, and should we use -- should we be thinking about book value as an indication? When we've done what we think are conservative burndown estimates, we get something on the order of $12 a share in residual value, and that does not give any credit to an ongoing franchise. And I've obviously seen management buy stock above seven of -- and again, just below six, and I think it's very important, given the specifics of this industry, for you to be able to communicate to long-term investors how you think about intrinsic value. And I've got one other question.

Well, relative to the value, that's every individual investor's decision on how they determine that.

Clearly, we went through an exercise as part of the capital raise whereby we went through the exercise that you described relative to the runoff value of the Company, and then you add in the quality business that we are writing since that point in time and you get the higher levels on what the stock is priced today in book value. So looking at the -- on that basis, you have to make your own decisions on what value is.

Let me put it a different way, then. You did give us a sense per burndown value on the capital raise, and would you say that your assumptions for future losses have changed materially? Obviously, the market -- the point here, obviously, is they have not. (Multiple speakers)

We can't hear your question. You're breaking.

I'm sorry. I appreciate -- that's very helpful because obviously when those assumptions were made, the stock was much higher, the market -- Mr. Market is taking a different view, and it's very helpful for you to reaffirm that those original assumptions would still hold in your view. None of this current noise going through the system has changed that perspective. (Multiple speakers). So thank you.

Donna Halverstadt.

Thank you. I got onto your call very late, so apologies if you already talked about this. There's two things I wanted to ask about. The first thing is -- relates to lenders' allocation of business to all the various MIs. And to the extent that we exclude market share related to things such as your and Countrywide's ongoing debate, if we exclude that sort of stuff, have you noticed any market-share shift based on lenders' or GSEs' perceptions of the counterparty credit risk of the various MIs?

We really haven't, Donna. We're holding at a -- we've increased a little bit this year, I think from 22% to 24%. On our estimate. Although I must tell you, it's hard to estimate because United guarantee numbers aren't in MICo, nor Essent numbers. But given the best we can do with that, we would estimate ours has gone up somewhat.

I don't see lenders making that decision relative to counterparties, or the GSEs changing their opinion because of counterparties.

Okay, great. The other thing I wanted to ask about, I hope, is just a clarification for those of us who don't live and breathe legalese. But in your disclosure related to the lawsuit with Countrywide, you talk about the panel hearing on the 24 bellwether loans. And at some point in the past, that was postponed from October 2011 to May 2012. Does -- can the lawsuit itself continue between now and May 2012, or is the lawsuit held up until that panel hearing on the bellwether loans takes place?

It gets -- overstatement here, extremely -- or understatement, extremely complicated when -- in trying to reconcile arbitration hearings and legal court issues that both state and federal are involved. So I know this is not a specific answer, we can do a little more research and try and give you a little bit better clarity on it, but suffice it to say that it's extremely complicated and a lot of inner workings between state and federal courts, as well as the arbitration.

Okay, great. Thank you.

Nat Otis.

Good morning. Just a couple of quick questions. Just going back on that foreclosure settlement, you said you were a part of the negotiations. And my guess is you probably can't comment on it. But what -- any color on whether -- if there are discussions on principal write-down, that they would -- that the banks would have talked to you about the possibility of taking part in any type of principal write-down?

First, Nat, just to clarify, we are not part of the negotiation discussion on the Attorney Generals or any other settlements that have been announced. We are certainly -- that could be affected, but we don't know how. But we are not a party to those discussions.

But has anyone come to you in any way and given you a hypothetical, or you've just been left on your own?

No one has come to us with, here is this and what would you do. So we've had none of that.

Okay. Fair enough. And then, just -- there is a little commentary in your press release on a disagreement with Freddie Mac on pool insurance. Is there any way to give just a little bit of color on that?

The color, unfortunately, is what you have in there, is that we responded almost a year ago with positions and in response, [of course], that they made to us, and we are waiting to hear.

Has that (multiple speakers)

(Multiple speakers) is whether or not -- it's interpretation of coverage on the policy and whether things sunset or don't sunset.

Has that had any negative impact on your reserving going forward as you deal with them?

No. As we pointed out, we are continuing to pay claims under that policy. But relative to our interpretation, we have not booked incurreds this quarter of about $50 million, I think. And we'll continue to footnote that.

So to that point, we've not booked $51 million, I believe, on future contingencies. So we are paying under the policy, but relative to future increased losses under certain of the policies, we have not accrued for those and we highlighted that.

Can you give -- so the disagreement is with Freddie. Can you give any commentary on the Fannie side, whether -- has Fannie -- are you in agreement with contractually how you been going about things with Fannie on that side?

We don't have any issues, similar issues, with Fannie.

Fair enough. Thank you.

[Ryan Zakaria].

Thanks for taking my question. Mine is just on the claim rate. It would appear that that percentage of new notices that you ultimately expect to pay a claim on has increased substantially. We can't get a clear determination of what it is without knowing what the split between adverse development and redundancy and the provision on new notices is. Can you just provide maybe some color until the Q comes out on how that trended vis-a-vis last quarter and last year?

Claims rate have been increased in this quarter. I think they were increased in the second quarter. Severity, no modest -- or maybe just a modest change, but probably relatively in the last three quarters pretty flat. So increasing claim rates each of the last, I would say, four quarters, based again on the mix of inventory that we talked about and the aging.

Can you provide an order of magnitude that might allow us to glean how much of the provision was on new notices or the current year versus prior years?

I guess we don't break it up that way, but I would indicate that on an order of magnitude, the rates that we would have been providing increased, as I said, on a quarterly basis, the amount of which would probably be, on a mix basis, somewhere around $100 million probably when you gather all the notices and by vintage for the given quarter.

Okay, thanks.

(Operator Instructions). Steve Stelmach.

Just real quick, I want to follow up on the insider purchaser question. You were buying stock at 776. You continue to guide towards GAAP losses going forward. What is your investment thesis here? Why do you think the stock is worth more than 776?

The opportunity -- this too will pass, what we are going through. And the franchise value of the Company in looking at the value of the business going forward and the opportunity that we have to regain market share from the FHA is tremendous.

So looking at both the existing value of the book that we have, but then also, more importantly, the value of the business going forward, I think is a terrific opportunity. We have the opportunity with the FHA to gain back tons of business, and I think with the new standards that will be coming out as part of what's going on on the regulatory front relative to underwriting standards, it's a return to old relative to credit quality. So I feel really good about the credit quality of business going forward. I think we have a tremendous opportunity to recapture business from the government.

I also think there's opportunities for MGIC to consolidate our own industry and take advantage of opportunities there where some may not look at the business the same that we do. So on many fronts going forward, I think it's a wonderful opportunity.

That's very helpful. And then, just maybe follow-up to that is, a lot of pushback investors have or fear is that perhaps, given the risk to capital number, they could face a capital raise or dilutive capital raise. Buying stock at 776 probably precludes that, at least for the time being. How do you get comfort that's not an eventuality?

What happens in the future, who knows? But in looking at the opportunity, I think if those capital raises happen, it would be ones that were based on an offensive basis of either consolidation of the industry or FHA opportunities to write new business.

But there's no regulatory pressure at this point in time, is there? Not that I see, at least.

Not with MGIC.

Understood. Curt, thanks very much.

I'm showing no further questions at this time.

With that, thank you all for your interest in the Company, and have a wonderful day. Thank you.

Ladies and gentlemen, that does conclude today's conference. You may all disconnect and have a wonderful day.