MGIC Investment Corp (MTG) 2007 Q2 法說會逐字稿

Good day ladies and gentlemen and welcome to the MGIC second quarter earnings conference call. At this time all participants are in a listen-only mode. Later we'll conduct a question-and-answer session and instructions will follow at this time. (OPERATOR INSTRUCTIONS) As a reminder this conference call is being recorded.

I would now like to introduce your host for today's conference Mr. Mike Zimmerman. Sir, you may begin.

Thank you. 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 second quarter of 2007 results are Chairman and CEO, Curt Culver, Executive Vice President and CFO, Mike Lauer, and Executive Vice President of Risk Management, Larry Pierzchalski. We also have two senior executives from C-BASS, Bruce Williams John Draghi joining us via telephone to discuss their quarterly results. Before we get started this morning, I again would like to ask all participants to try to limit themselves to one question and a follow-up and return to the queue to provide as many people as possible the opportunity to ask a question. Also, back by popular demand, we've added back the sub bulk and flow statistics for the additional information disclosure we provide and that are contained in this morning's press release. Next, Curt will begin with his opening remarks which will be followed by comments from Bruce and John and then we'll open up the lines for discussion.

Finally, I want to remind all participants that our earnings release of this morning, which maybe accessed on MGIC's website which is located at www.mgic.com includes additional information about the Company's quarterly results that we will refer to during the call and includes certain non-GAAP financial measures. During the course of this call we may 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 on 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 developments. Now I would like to turn the call over to Curt.

Thanks, Mike and good morning. Net income in the first quarter was $76.7 million compared with $149.8 million a year ago. Diluted earnings per share were $0.93 versus $1.74 a year ago. The quarter was a difficult one as evidenced by our financial results. The results do reinforce, however, the franchise issue that has concerned me the most over the past five years and that was a continued growth of customers avoiding our product due to accelerating home prices and therefore perceived lack of risk in the marketplace. That perceived lack of risk in the marketplace also led to significant abuses of underwriting guidelines in the subprime sector but also the prime sector, a fact that many of us connected to the residential mortgage market are dealing with today.

Before I discuss a loss side of the business, let me take a minute to discuss the long-term positives we're creating on the revenue side. Flow new insurance in the quarter totaled $17.3 billion which was up 71% from a year ago and 66% sequentially. Year-to-date flow was up 53% from a year ago, clearly signifying the growth of mortgage insurance penetration nationally. Persistency also improved during the quarter to 72%. Flow was at 75.9% and bulk at 59.6% and those were versus 70% last quarter. So a nice increase there also. With the growth in flow new insurance written and the continued increase in persistency we were able to grow our insurance in force to $186.1 billion, which is a 10% increase from a year ago. The impact in long-term revenues for the Company from this growth is significant.

On the loss side, we are experiencing a run-up of delinquencies and paid losses in the high dollar state of Florida and California and continued weakness in the Midwest, particularly Michigan. As a result, paid losses in the quarter totaled $188 million and losses incurred totaled $235 million. I'll discuss our loss expectations for the remainder of the year in a minute. The second quarter was also a difficult one for C-BASS but one in which they returned to profitability and, as Mike mentioned, Bruce and John are with us by telephone and following my remarks I'll turn it over to them to discuss their results and their outlook. Internationally I was pleased that we insured our first policies in Australia during June, which sets us on a path for long-term positive international results. We also hired a CEO and Chief Operating Officer in Canada and established an office in Toronto with hopes of writing policies in Canada by the first quarter of 2008.

Regarding the integration of our transaction with Radian, we continue to make significant progress and are on track to close in early October. We have received all state approvals required but two, and expect to receive them to meet our expected closing date. Regarding the disposition of our interest in our joint venture partner Sherman Financial and C-BASS, we also have made progress. In the case of Sherman, we are in the process of documenting a transaction. Regarding C-BASS, even though there is much noise in their space we're pleased with the number of investors who are participating in the process which demonstrates the strength of the C-BASS franchise and its long-term opportunities. Management is in the midst of making presentations to interested parties and again, we're encouraged by the list of potential partners.

Regarding the remainder of the year, it will be a difficult one financially. While we had hope paid losses would be in the neighborhood of $680 million this year, up 10% from a year ago, it now appears they'll be in the range of $750 to $800 million. And while we wish we could narrow that range for you, it truly is dependent on the speed in which certain delinquencies result in claims and our potential to mitigate certain of these claims. While some of the bulk securitizations we insure put some limit on our ability to work with borrows to eliminate or mitigate or limit their losses, we are exploring a number of strategies that fit within the securitization restrictions that could be meaningful for us in savings. On the revenue side, I couldn't be more pleased with the long-term rebirth of our product brought on in large part by today's losses. Our flow NIW penetration should continue to expand as should persistency to help to us grow insurance in force at a meaningful pace. And we expect to retain renewal premiums for a longer time resulting from the slowdown in real estate values. I also think our future books of business will stronger from a credit perspective given the return of underwriting discipline to the marketplace. Hopefully we'll also see additional bulk volume this year, which would also add to our growth.

Finally, a very important part of our long-term financials are also our joint venture results. As shown by our financial results, Sherman had another strong quarter. C-BASS on the other hand had an extremely challenging quarter but one that also highlights the strength of their franchise as evidenced by what they accomplished in the quarter. With that, let me turn the call over to Bruce and John who are joining us as I said telephonically from New York. Bruce?

Okay. Thanks, Curt. The repricing and increased credit rate premiums for subprime paper that typified the first quarter continued through the second quarter. Bankruptcies and consolidations continued poor performance at the 2006 vintage and uncertainties surrounding possible government regulation all combined to keep market pricing at distressed levels. After a brief recovery in May, the news of certain hedge fund problems and corollary investor redemptions caused spreads in the cash markets to widen back to first quarter levels and cause index spreads to widen beyond first quarter levels. Investors became even more keenly attuned to servicing issues on subprime paper and most recently, the rating industries took industry-wide rating actions that deepened stress in the market. Liquidity remains a primary issue for subprime market participants as illustrated by continued suppressed origination volumes, increased margin calls from all lenders, significant tiering for subprime asset-backed issuers and limited access to the CBO and RMBS securitization markets.

We continued to update our credit and prepayment models to ensure that we are comfortable with the valuation of our portfolio. Our disciplined buying practices during the course of 2006 and 2007 have been effective and the delinquency performance of our 2006 paper is well below the industry average. In fact, the July rating agency actions affected only two C-BASS issued first lien RMBS bonds. Only eight underlying bonds across four C-BASS CBOs are on the rating agency list. No C-BASS CBO was downgraded or put on credit watch and CBOs that were subject to rating actions are not in any of our CBOs. In addition our CBO portfolio has no exposure any CDO squared transactions. Our bond valuation method has not changed. Our book value approximates market value, assuming orderly transactions that are not distressed sales or liquidations. We verify this by a three-way comparison of observed market prices, financing marks from our lenders and our internal calculations during our ongoing securities review. We evaluate all significant differences between book value and market value among these three valuation methods and take appropriate action as warranted. With that, I'll turn it over to John to talk about the second quarter.

Thanks, Bruce. Our focus for the second quarter was to return to profitability, which we achieved by reporting a pretax profit of approximately $50 million. We were negatively impacted during the quarter by a change to our loss projections for second lien collateral and a change to our prepayment assumptions for alt-A collateral. These results include an operating profit at Litton Loan Servicing, interest income on our portfolio and reduced compensation expenses. Demonstrating the strength of our franchise, we were able to continue to access the market. We priced four whole loan deals and one CBO during the quarter, all within our spread expectations and at the tight end of the market. In fact, on the day that Moody's and S&P released sweeping ratings action for subprime paper, we were able to successfully price 2007 CB6 and place bonds with investors. From a liquidity perspective, we decreased our whole loan position from $2.6 billion at the beginning of the quarter to $2.4 billion as of June 30. Total assets declined $300 million from $6.9 billion to $6.6 billion at June 30. As compared to year-end, whole loans are down from $3.9 billion and total assets are down from $8.8 billion. This is in line with industry-wide delevering.

We have today on the order of $150 million in cash resources. We will remain conservative with our cash through the rest of the year, adjusting as needed to reflect market dynamics. We closed the Fieldstone transaction on Tuesday, July 17th at the merger price of $4 per share. The acquisition provided us with strategic assets, securities loans and servicing. We continue to believe our ability to align a front-end originator with our analytics and servicing expertise will complement our business model. The recent market has been challenging for subprime mortgage originators, but as a response, we've been actively involved with implementing the appropriate changes at Fieldstone to include downsizing initiatives and revise production guidelines.

For the full year, we expect pretax earnings to be between $125 and $175 million, due to adjustments in the second liens and all day collateral we already mentioned. This assumes one, our credit portfolio continues to perform within our current expectations. Two, our whole loan purchase volume remains at levels and three, spreads do not widen significantly from here. With that, I'll turn it back to Mike Zimmerman.

Great. Operator we can open up the lines for questions. I would invite all participants if we can one question and a follow-up.

(OPERATOR INSTRUCTIONS) Our first question comes from Steve Stelmach of FBR.

Good morning,.

Good morning.

Curt, thanks for the pay claims guidance. Could you just give us a little bit more detail around the incurreds. I know it's a little more difficult to do, but incurreds outpaced claims this quarter by about $15 million. How do we think about that? Are you guys anticipating peak losses coming sooner then you otherwise anticipated or does it mean the magnitude of loss is going be much greater then you otherwise anticipated and then going forward off that paid number, it's a $50 million gross up for incurreds about a right run rate.

Yeah. This is Mike. The run rate would be correct on the reserve build of about 50. And I think what you have going on is two things. One is, again the increased severity and Larry is going to talk about that later about what markets that came from and then secondly, the increase in the pure delinquency itself for the quarter. We did have an increase in notices for the second quarter and traditionally, we would have seen a flattening or decrease. So it was a significant jump and we would anticipate further increases, maybe modestly in delinquencies but more importantly increase in severity. I would say from a benchmark standpoint at least a 50 over the paids going in on a forecast basis for the third and fourth quarter. That would be subject to again any significant changes in claim rates, et cetera, but clearly we would be building reserve in this environment as we anticipate higher claims and severity.

Okay. And then when it comes to the pace or loss development, is the accelerated loss development in '07 sort of running any losses into '07 or is it simply just a matter of higher losses in '08 as well?

There will be higher losses in '08 also. I think they'll be up marginally where we'll be today or at year-end, but they will be higher in '08. I don't think you'll see improvement until '09.

Thank you very much.

Our next question comes from Geoffrey Dunn of KBW.

Thanks, good morning,.

Good morning.

I want to kind of follow up on that line of questioning. Do you have an idea of how far into the '05 seasoning you and how far into the '06 seasoning and with respect to '06 has it surprised you how quickly ' it's ramped up. I think before you thought the seasoning would be back on a more traditional pattern.

From my perspective and Larry's is looking up a couple of the seasoning questions, I think on '06 it clearly has surprised us relative to the speed in which some of the higher dollar states, Florida and California, have progressed. And again, it's not as those these states are in trauma, it's just a return to some sense of normalcy from where they were coming from which was so low. But the acceleration there of, if you will, delinquencies to paids has been a little quicker then we would have anticipated. Larry?

Relative to your question on where are we in the life cycle for the '05 book, I would say for the '05 book from a delinquency inventory standpoint peak, we're close to peak. Maybe another quarter or two and then relative to paids on the '05 book. That would generally follow another three quarters or so after from the default peaks. So a quarter or two on the default peak. Four to five for the claim peak on the '05 book.

Did that answer the question or?

Any feel on how far we're into '06 we are?

'06.

Or maybe to give you a way out. A lot of people are citing end of '08 kind of peak delinquency period for '06 vintage. Would you agree with that?

I just gave you the '05. I would say add another four quarters onto the '05 so '06 default would probably peak in about six quarters.

Okay and then just to understand Curt, what you said about '08 paids being marginally higher and I know it's hard to look out that far. Why would they only be marginally higher if the brunt of '05 paids and '06 paids have yet to hit.

I think a lot of '05s are already hitting us and again, the early development in '06 has been not good. We've had step ups relative to delinquency and paid performance there outside of our boundaries anyway.

And then on top of that. The bulk writings here are very minimal, so they will not be adding and moving up that default or claim curve. So that the -- just the size, the smaller size of the new bulk book in particular won't be feeding claims in '09.

Okay, thank you.

Our next question is from John Barlow of Equity Strategies. (OPERATOR INSTRUCTIONS)

Maybe we can come back to John.

Our next question comes from David Hochstim of Bear Stearns.

Hi, David.

Hi. I wonder if you could, maybe Larry could talk a little bit more about, provide some more color on the claims and kind of what's happening in California and Florida specifically. What kinds of loans you're seeing generate claims and what you have learned from that experience in terms of underwriting and then wonder if you could just give a sense of how high industry penetration could get in terms of new insurance written over the next year or so.

Let me take a shot at that. California. It's not much of an event for the flow book of business simply because our California exposure in the flow book is only a few percent. Whereas on the bulk side of the business, it's 21, 22% of the exposure, so as California moves, it impacts the bulk much more than the flow. Florida kind of impacts both of the flow and the bulk. Florida flow, I think it's about 8% of the in force and 14, 15% of the bulk. And those two states have moved from very low levels to at least normal, maybe above normal default levels rather quickly. We anticipated Florida and California slowing but it moved much quicker and not as gradual as we anticipated, thus the forecast being off and in particular California.

There's a short time frame from delinquency to claim just because of the foreclosure time periods there. So, California moved quicker and not as gradual and because of the foreclosure time frames, results and claim activity much quicker than the rest of the country. And then on top of that, we thought the Midwest, the auto states, had stabilized and Michigan in particular is a tad worse in terms of the delinquency rate on a new books versus just the prior books and just looking at some of the employment numbers now, I guess '06, Michigan employment was down 1.1% and now the estimate for '07 is down almost another 1%. And so the auto industry is still working through some of their issues there.

Are there any -- is there a possibility that some of these claims that are being generated from the bulk business in California are a result of, I guess, flawed underwriting or underwriting that didn't live up to the standards that were, I guess promised?

Well, when you have 20% home price appreciation going on in California or any place, there isn't a bad loan being originated because to a degree there's an issue of appreciation bails. Now that is gone. Home price appreciation is rather flat. So that impacts all types of loans, even a bit in the prime space. So, some of these subprime people by definition that have trouble managing credit as well as the alt-A stated income loans. When that appreciation goes away, you find out what you really have.

And David, the subprime space where the borrowers with lower credit scores rely heavily on being able to refinance their debt annually almost and with the appreciation that was going on people were able to do that, particularly in California. This past year they weren't able to do that, so their outlet was taken away. The subprime spaces or bulk space is more driven probably by ability to refinance than employment, which drives the flow side of the equations and that's what's hurt the performance in that space. It's just an inability to refinance because of lack of appreciation in their market.

You referenced to a spillover for underwriting and subprime into the prime. What are some examples of that?

There's no question that's been going on. We've talked about it for a long time. With alt-A where you had stated income and the NINA no income, no asset documentation patterns where we saw far higher claim rates. Five times higher claim rates on similarly FICO scores and in fact, that the highest 700 pluses there was a seven times higher claims rate documented versus undocumented. So that runs through the system and that did become a mechanism to originate mortgage loans on a growing percentage over the past couple of years and that is running through people's books today.

Okay. Thanks.

You bet.

Our next question comes from Michael Grasher from Piper Jaffray.

Good morning. I wanted to talk for a bit about the new insurance written in the quarter. Looks like 45% of the new insurance written was greater than 95% LTV, which seems rather high, or maybe another way to think about that is just we're looking at a risk transfer away from the originators into the MI's. So I've got a couple of questions. I guess your thoughts that. Why so much? Is there a back wash of loans that have moved away and are simply being pushed onto you? I want to get some perspective on that.

A couple of things if you look at last year's originations of first time home buyers, the average down payment for those first time home buyers was 2% and 43% of those home buyers put nothing down. So, that's usually the general market in which we deal. We've had a growing percentage of plus 95% loan since 2000 when that started coming into vogue, if you will. And I think with the return of our product penetration against 80/20's, you're seeing some of that now reflected in mortgage insurers' numbers, that they're insuring those borrows that would have not put a down payment on the other transaction doing the same under the mortgage insurance transaction. So we feel we're adequately priced. Our claim performance on those loans has been within -- far within our expectations, but I think it's more of a reflection of the marketplace and also our increasing penetration of the marketplace against alternative executions.

I guess another way to think about that is the 80/10/10. Where has that gone, will it ever return?

Well, the 80/10/10 pretty much fell out of favor a couple of years ago to the 80/20.

Understood.

And I'm not here to say that those will go away forever because I've been in the business 30 some years and 80/10/10's have been part of the business for all that period of time. It's just that it used to be a very small percentage that grew to in excess of 50% of the business. I think we're trending the other way back to the good old days of where it should be, that there is real risk in the marketplace and credit enhancement is far superior and you're seeing that on a lot of lenders' and banks' financials today in their home equity lines, where that's where their losses is showing up. So again, this is one of the things that we're paying the price today but we're building our franchise for tomorrow.

Okay. And then in terms of the risk in force that you put on in this quarter, is it fair to say that number correlates or -- relative to the 45%, I guess, what percent of these policies would the risk in force be.

Did you ask what percentage of risk in force is hundreds?

Yeah.

Okay. Just wanted to restate that?

Do you know that Mike?

Yeah, on the flow book. It's approximately 30% of the flow.

And that's the new policies in the quarter?

No, that's total risk in force. Flow risk in force is 30% in the quarter. That's the number that we disclosed in the release, which is a 45%.

Not of risk in force, though.

Yeah, it's not as -- or at least that doesn't seem to imply risk in force.

30% of the flow risk in force is over 95.

I thought she was asking what percent this was of our total book. If we answered your question, so be it. I misunderstood it then.

I'll come back to you offline. Thank you.

Our next question comes from Mark Giambrone of Sparrow Hanley.

Good morning, fellows.

Good morning.

Two quick questions, one is I see you did buy back some stock in the quarter, included a substantial buy back coming after the Radian transaction? Could you just tell me a little bit about your appetite and/or ability to buy back stock prior to the -- including the transaction and then again how quickly things can be done post the transaction.

Post the transaction, as we outlined in February, is going to be subject to the deals. Obviously, you recall we had proceeds from the sale of investment in C-BASS and Sherman as well as some additional leverage. So all of that is still on track just as we stated for the most part in February and there's really nothing has changed in our assumptions with respect to that. Relative to the third quarter we have plans for probably getting another dividend upstream. So very similar maybe to the second quarter.

And lastly I apologize to make you repeat this I missed the paid claims guidance number.

750 to 800.

For this year.

For this a year.

All right, okay, thank you.

Yep.

Our next question comes from Matthew Roswell of Stifel Nicolaus.

Yes, two questions, first, in terms of the default book, are you willing to break out the percentages sort of by cohort '05, '06 or pre '05, '05, '06 and by state?

We have the '05/'06 comprises approximately 38% of the delinquency inventory and we'll have to look to see if we have the state breakdown by that available right here on the call, but '05/'06 comprises approximately 38% of the unit.

Okay and a quick follow-up question. The C-BASS profit guidance of 125 to 175, is that pretax?

Yes.

Okay. That's what I thought. Thank you.

Our next question comes from Robert Ryan of Merrill Lynch.

Good morning.

Good morning,.

I noticed a fairly dramatic slowdown in the year-over-year change in sort of the average loan size? So they're still getting larger but at a much slower pace. What implication does that have and at what point in the future for claim payments?

In terms of severity?

Yeah.

Well, for the next year or so severity should be moving up, driven by the bulk business, in particular the '05/'06 books, which had on the bulk side probably $120,000 to $140,000 average loan amount for '05/'06 compared to 170 to 180 bulk books '03/'04. So as those '03/'04 move past peak and are replaced by '05/'06, you're seeing a pretty good jump in average loan balance and on top of that now we're getting activity from high cost areas from Florida and California, so the bulk severity should move up pretty strongly here for the next year-and-a-half before it starts slowing. You got the same thing going on the flow side, but book to book to book the loan balances were probably up about $5,000- $6,000 year-to-year. So that 25/30 cover should translate into average claim on the flow side going up about $2,000. Average balance is going up $6,000- $7,000 grand per and annum. 25, 30 cover should translate into a couple of thousand. Maybe a tad more that because of the high cost areas but on the flow side we don't have the California issue but we have the Florida issue. So short-term, next 18 months, pretty good increase quarter-to-quarter on average severity.

Okay, so all told, would you say there's a decent likelihood that paid claims might be able to peak in 2008 and then stabilize or actually go down in '09.

Yes. We hope so. Sure.

Great, thank you.

You bet.

Our next question comes from Andrew Brill of Goldman Sachs.

Thanks, I just wanted to understand a little bit better what gives you confidence that there will be improvement in losses in 2009. And I guess specifically, what are you assuming for home prices that gives you comfort that the '07 book is going to fare a lot better then the '05 and '06 book?

We see the early delinquency pass of the '07, late '06/'07, and it doesn't seem to be any further deterioration at this point. Most of it was from beginning of '05 into the second half of '05 and '06 books and we're not seeing further into '07. But going back to an earlier comment, probably one of the biggest reasons is the bulk books have been $20 billion or so, and we've only written four this year and I want to say in the short-term we would see much improvement in that, so we'll probably go from $20 billion book size in the bulk to less than 10. And in '09 that will be pretty much driving towards peak default and claim, so that book will be much smaller and the bulk book tends to be more riskier and so a much smaller book in '09 will offset some of the other things you're seeing.

A lot of the '06 bulk book which was small, I think it was even less then 20, like 16 or so, was agency transactions with deductibles involved and profits. So it won't run through the numbers as significantly as the larger bulk books did that we had done in the past.

Just a separate question. If, by some chance you can't get the pricing you're hoping for on C-BASS, do you have any flexibility to postpone that sale at all, potentially even into next year?

Well, right now we don't see that to be an issue, but I guess what you're asking for is in the event that we were unable to sell it at a price that was acceptable to us, would we have another form of alternative. Yeah. We would have to do something. We certainly wouldn't take a significant hit on it just because of a unique time in the market.

All right, thank you.

Thank you.

Just to clarify, I think it was Mike with the question on the hundreds. Greater than 95 is a percentage of total risk in force. Bulk in flow combined is 25%.

Our next question comes from [Al Cupercino] of [Maydolf].

Hi. Thanks very much. I know my question is focusing just on one factor on the loss incurred ratio and there's obviously several factors that impact that, but I'm just looking at the paid to incurred was -- looks very conservative the first half of this year, just eyeballing at 80 or 90%. I believe the paid to incurred in the first half of '06, again eyeballing it and hopefully my math is right, 110, 115%, somewhere around that. So that's become much more conservative. There's obviously in a normal year a seasonal increase across the quarters in the loss incurred ratio. Given the increase -- given the reserve buildup you've done the loan paid to incurred, should we -- could you help us think about how the seasonal increase in the loss incurred ratio maybe different this year than in past years?

Yeah. That's the comment I made earlier about the second quarter was a little bit unique in the sense that normally, traditionally, you would see a decrease in delinquencies in the first and second quarter. We did have a decrease of delinquencies about 2500 in the first quarter of this year and last year, in the second quarter, we had a decrease of about 3000. This year, we had an increase of about 4500. So the June or the second quarter increase was significant. We -- probably higher then we would have thought even in these times. So that was the first change. And then secondly, we would anticipate some increase in delinquencies historically in third and fourth quarters. So to your point, the incurreds generally are greater then the paids in the third and fourth quarter.

Fair enough, thank you.

You bet.

Our next question comes from Brad Ball of Citigroup.

Thanks. Curt, in your prepared remarks you talked about exploring workout strategy during the securitizations. I wonder if you could give us a little color on that and what you would expect in terms of timing of those efforts mitigating losses.

Yeah, I know by my saying that would bring in that question which is somewhat competitive situation on what we're doing relative to our competitors. Suffice it to say that we have strategies that we're working on today that we think are very workable in the marketplace and that will meet within the restrictions and covenants of those securitizations that, again, could be meaningful and I can't get into the details of that again just for competitive reasons. But --

Let me try to add maybe a little more color. There's a couple of three components. One is certainly REO and we're trying to be -- seize every REO opportunity that we can. A second component is presales. Try to dispose of the property earlier in the game rather then let it go through the full cycle and add to the loss and then lastly, and maybe hopefully more importantly, it has to do with loan modifications, particularly helping people deal with the pay shock.

And that's the area in particular that we have some workout strategies that we're working with servicers on.

Got it. Okay. And separately, given your revised guidance for paids and your comments about reserve billing for the balance of the year, it looks like we're talking about at least a couple hundred million higher losses incurreds, but nothing that seems life-threatening. Could you envision a scenario where your book value is at risk?

No.

No.

That's right.

Which we have talked about a lot and that certainly seems to be the sentiment, but we don't envision that scenario. Again, I've addressed that many times.

If there's something that will catch -- that would require to you revise guidance again that would catch you by surprise in the second half, what is that going to be? Is that going to be a further spike in delinquencies?

I can't imagine a scenario that would change our guidance in the second half of the year. We've been -- I mean -- because of the fact that if you will that we put the guidance out at 680 and I felt terrible about the fact that we're coming in at a higher guidance given how fast things move, so we think that as a very conservative number that we're dealing with and I can't imagine a situation that would change guidance.

Okay. One last question. You said there are two states that still haven't granted approval for the Radian merger, which are those and what are their issues?

Texas and New York. And there's no issue, they just take longer to do their work.

So they're not dragging their heels. It's just --

No, no, there's no issues they've expressed with us at all.

I got it. Thanks.

Our next question comes from Eric Wasserstrom of UBS.

Thanks. Just to follow up on that last topic. Heading into the merger and you mentioned a little bit that things are going as you expected, are you still comfortable with the cost savings expectations and the revenue synergies and the timing and all of those kinds of things relative to when you first discussed it?

On the cost savings. Again, this is me speaking but I think we'll do better than the numbers that were shown there but I think it will take longer to do than what we expected or put forth at the time we announced the transaction. So, I think the cost in summary will be better savings, but we won't implement them as quickly as we had hoped. But I think that may go out maybe six months more relative to those savings. As far as --

The revenue synergies -- we did not field any revenue synergies in.

Maybe the market share.

Yeah, sorry. I meant the market share issue, yeah.

Yeah, we've -- and that's something that can be addressed with Radian next week, but we've seen or had no customer feedback relative to curtailing opportunity because of the proposed transaction. So, if anything, we're growing market share now on the flow side.

And what is it that you think is going to cause that extra period of time to realize the cost savings?

I think the integration is just so difficult relative to the technology. And that's something we didn't -- you can't get your hands on until you really down getting dirty with it and we've been doing that for a while now and we fully understand the complexity of the project and I think that will take us a little longer then we estimated, that we'll have to keep redundant facilities in Philadelphia for longer then we expected, things of that sort.

And then just finally on C-BASS, I didn't quite understand whether the valuation process was a more of a mark-to-model process or mark-to-market process and I guess I'm just trying to reconcile that with the magnitude of spread widening that occurred in the second quarter relative to the first quarter and the difference that played through in terms of the financial results.

Bruce or John, you want to --

Well, I think as I indicated, Curt, our book value approximates market value. We did take some widening and write-downs in the second quarter. As an added point, June 30th sort of marks the end of our fiscal year and Deloit performed a review of our valuation methodology as of 6/30 and no adjustments were made so we're pretty comfortable with where we've got our portfolio marked.

Thank you, but is that a market mark or a model mark?

The reality of it is they're very few things trading in the market so the (inaudible) spread is wide as it has ever been. It's very difficult to tell you what the market is. We have no intention of selling our portfolio and we expect it's going to realize what we expect in terms of the yield where we've got it marked.

All right, thanks very much.

Our next question comes from [Ross Levin] of [Arbiter Partners].

My question is regarding the cure rate assumptions for defaults as they're reported to you. You guys have sort of impugned a level of cure at various times from between default and an actual foreclosure, have you started to modify the probability of cure at various time periods to reflect changes in what you're seeing recently, or are you still using a sort of an older model and if so, at what point do start to have to modify your cure rate assumptions.

We adjust those every month based on experience. Every delinquency is tracked by market by product. And as the experience changes, an example would be Michigan as Larry talked about earlier. This quarter in the last couple of months we saw an increase, if you will, in claims rate or decrease in cure rate so we could adjust it. It's modified if you will every month based on experience.

There's no sort of trailing effect where it's the average of the last six months but you're incorporating the last month's data. You're taking the last month's experience as of a given sort of portion of the life cycle of a default, if you will and applying it to every incremental default.

No, you wouldn't do that because you would have current -- you would have a delinquency. We're talking about different delinquencies now in Michigan for example, one that's been there for 24 months versus one that's been there for a month. So the month, the current month delinquency obviously would have I different claims rate assumption then one that's been there for three months, four months, five months. But in overall for the aggregate for Michigan we took the claims rate up, based on the experience.

Thank you.

Our next question comes from [Mike Grandall] of Key Colony Fund.

You talked about a special dividend in the third quarter. Is that your regular quarterly dividend or is this going to be a larger dividend?

What we're talking about here is dividend from the writing company up to the holding company, not a common stock dividend.

Right, but what I mean is that the regular $55 million quarterly dividend?

No. That would be a special dividend.

And how big do you think that will be?

We'll report on that next quarter.

Okay.

Our next question comes from Geoffrey Dunn of KBW.

Thanks. There seems to be a lot of speculation that you won't get the value you want for C-BASS or not even get the sale off. But I have to imagine that the people you're talking to and I'm assuming its more strategic then financial buyers are understanding this is a pretty unique environment and looking at the longer term value. Is that what you're encountering or are you encountering pushback from people who think that this environment is perpetual and hence affects their valuation.

No. We can't talk about that. We're in negotiations with numerous buyers and they're doing their due diligence. And there's no reason for us to anticipate any change then from what we said in February.

Okay. Thanks.

Yeah.

Our next question comes from Nandu Narayan of Trident Investments.

I was hoping to get more clarification on some of the assumptions you're making in -- when you're taking the reserves, because obviously in terms of the housing market, we appear to be in uncharted territory. Because I think for the first time since the great depression we're seeing a national decline in home prices, leverages at all-time highs, foreclosures are rising dramatically, not to mention, of course that we're now technically in a recession yet. So, what is the scenario you're building in terms of the reserves you're setting up right now because obviously, I think in Brad Ball's earlier question you responded that you couldn't even see a scenario under which things could get much worse than what you'd reserved right now? What is the scenario you constructed?

It was under paids.

Paids. We were talking about paid loss guidance there. We've got existing delinquencies there.

I understand. Okay. But what is the scenario you're projecting in terms of your loss reserves. Because if you had to sort of look at the various scenarios that might transpire, we seem to be uncharted territory in the housing market.

Let me just clarify again. Understand that the reserving relates around the existing delinquencies. Okay? So if you're asking to speculate in '09 what the delinquency level would be under certain scenarios that would be a different thing but the current financials reflect the existing delinquencies and the rates at which we believe they'll go to claim.

Okay. So in terms of your loss reserves. Are you making any projections as to what things might look like or what would be your central scenario on the basis of which you've made your reserve assessments?

As I stated, they're based on the existing delinquencies. You're saying in a different economic scenario if the delinquencies were higher in '08 or '09 what impact would that have on the financials? It would have impact in '08 and '09. Do you follow? We don't have -- I guess what you're saying is can it change that fast for delinquencies to be that much higher in the second half of the year, is that what you're asking?

What I'm concerned about perhaps is just the fact that in an environment like this when we seem to be uncharted territory in the housing market, part of the issue that might arise is that delinquencies might be dramatically higher.

They could be next year, yes.

And given also the fact that in this environment buying back stock and doing things like that would necessarily make it much more difficult for you to weather that possibility. So, I'm just wondering how likely you all think that the environment might get dramatically worse?

Again, I think we've addressed that from the aspect of what we felt paids will be and that they'll be up somewhat next year after this and a return to some semblance of normalcy in 2009 and within that it has those factors that you described that are impacting the housing market. It's not as though we're in uncharted waters if you look through our history or at least my history in the business. The real estate business is very cyclical and we've been in this area before. So based on what we see in the marketplace and our expectations of it, we've given you what we think will happen.

Okay. All right. Thank you very much.

Our next question comes from Howard Shapiro of SPK.

Hi, two questions, if I could. Moody's in their conference call last week mentioned that one of the areas where loan performance has varied versus their model is just the amount of absolute fraud that they're seeing. Can you just remind us what your options are or what your protections are if there is fraud in the underwriting and when you are now looking at all of the increase in delinquencies, how much effort are you placing on identifying fraud?

Well, first off, let me say that the level of fraud that we're seeing is relatively the same as it has been. Part of that, I mean fraud in a large way is tied to who you're doing business with and we try to do business with the right players and if we spot issues, then we try to not do business with those entities anymore. But quite frankly, if we see fraud, the material fraud, misrepresentation of the loan information, we have the ability to rescind. But our rescission rates haven't changed much. When we look for a fraud, we concentrate on delinquencies in the first year or so. That is when fraud, if it is it present, generally shows its ugly head. So we take a hard look at what we call early payment defaults. And that includes also the valuation side, which we have the ability to rescind on, but rescissions -- our stance is rescissions aren't any good for us in the long run. So we have the right to rescind but we try to get at the root cause and eliminate it so we don't have that issue down the road.

We've had that group that has been checking that for many, many years and so our performance within that is pretty consist year-over-year on what we rescind. And as Larry said, the key is knowing who you do business with and again, given our history we have a pretty good sense of who we're doing business with.

Okay and then just in terms of follow-up, can you tell us what your assumptions are for the next two years for, I guess I'll call it home price depreciation and if there's a significant variance for California, the Midwest and Florida can you tell us what those are specifically?

By and large, we kind of adopt or are pretty much in line with the economy.com, Moody's economy.com, Mark Zandy. Everybody's wrong. But they're in the business of doing that and we kind of review that and that kind of sets the stage, but we do make our own call but we generally work off of Moody's economy.com.

I think in the flow side you'll see less, if you will, depreciation in value. I think it will be insignificant relative to depreciation and you may see slight appreciation in that side of the business over the next couple of years which is what everyone is concerned about. It's the bulk side that has more significance because that steps out into higher loan amounts and there, as Larry said, we track closely with what Mark Zandy's group does.

Thanks.

You can look up his expectations.

Sure.

Operator, we have time for just one more question, please.

Our next question comes from James Shanahan of Wachovia.

Terrific, I'm glad I got in.

We are too, Jim.

I would like to ask a question about the bulk business. Can you describe how recent bulk channel volume say in the first half of '07 would compare to '06 or '05 in terms of the structures or the types of risks that you're taking, maybe the underlying collateral? I'm just interested in hearing how the bulk channel business maybe changing or evolving.

Yeah, the bulk business that we're writing as of late the last few quarters has been largely through Freddie Fannie, the GSE's, whereas historically, a good part of it, share of it, was not only GSE but capital market securitization oriented. Given what's happening with the securitization market as of late we're really not doing much there. It's generally for the purpose of the GSE's. And I guess our pricing is stronger now than last year by quite a fair margin, but that's kind of held us back in our writings because the market wasn't prepared to pay us what we thought the risk demanded.

How are those GSE deals being structured these days?

Well, of all sorts, some of them have deductibles. Some of them have deductibles and stop losses, some of them are straight per loan.

Which again was a big part of our '06 writings in the bulk space also.

Okay and a question about C-BASS and Sherman. When you've sold the majority interest in these joint venture businesses, and it's finalized, will there be an announcement of that or will we hear about it first sometime in October, early October, when you announce the closing of the Radian transaction?

Probably in conjunction with the closing of the transaction. That would be the plan.

Okay. And does C-BASS, by the way, hold any residual securities the balance sheet anymore or have those been written down significantly or what is the dollar value of residuals for subprime securitizations?

Yeah. Hi, this is John. We do hold residual interests on our balance sheet, just bare with me one second. The number's been in line with where we've traditionally been. It's somewhere on the vicinity of 35 to 40% of the total securities' portfolio that we have.

How does the securities portfolio compare to the total size of the balance sheet?

Securities portfolio is approximately $2 billion and the total balance sheet as of 6/30 reportings were approximately 6. So about a third.

Okay and do you have the range of lifetime loss assumptions that you would have in place for those residual securities?

It's a difficult question because it's pretty granular and it varies by deal and by vintage and also by whether it's a first lien or second lien transaction. So certainly the 2006 vintage first lien deals has been noticeably higher and the good news there is we anticipated that going into the transactions. So, those are noticeably higher then we were in 2004 and 2005.

And this is a portfolio that has been built up over a long period of time. So it does not reflect, as John said just recent 2006 production.

Actually well over 50% is prior to 2006.

Okay. Thank you very much.

Operator, with that, I know there are other questions online. We see that, but we've taken over an hour of time answering it and again, as we talked about it has been a difficult quarter, but truly it is one that reinforces why we're in the insurance business and is driving the revenue line for us long-term and I'm extremely pleased about that aspect of the business and the losses will work through the system over the next couple of years. So with that, I would like to thank you for your time and your interest in our Company and have a good day.

Ladies and gentlemen, thank you for your participation in today's conference. This concludes the program. You may now disconnect. Thank you and have a great day.