MGIC Investment Corp (MTG) 2007 Q1 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the MGIC first quarter earnings conference call. [OPERATOR INSTRUCTIONS]

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

Thanks, Ryan. 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 first 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. Also joining us on the call today to discuss C-BASS results is Bruce Williams, the co-Founder and Chief Executive Officer; and Peter [Serwyn], Manager, Director, and head of Investor Relations for C-BASS.

Before we get started this morning, there are a couple of housekeeping issues that I wanted to cover. First, I wanted to point out we made some changes to the additional information disclosures this quarter that are contained in the press release and we've provided that information in the new format back to the fourth quarter of 2005. We hope you find this new format useful. Next, Curt will begin with his opening remarks, which will be followed by comments from Bruce and then we'll open up the lines for questions. The call this morning is scheduled for one hour and given the number of participants on the call, I, and I'm sure they would appreciate if you could limit yourselves to one question and a follow-up and then return to the queue.

Finally, I want to remind all participants that our earnings release of this morning, which may be 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 contains certain non-GAAP financial measures. During the course of this call, we may make comments about our expectations of the future. Actual results may 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 statement, we are not undertaking an obligation to update those statements in the future in light of subsequent developments. Now I'll turn the call over to Curt.

Thank you, Mike, and good morning. Net income in the first quarter was $92.4 million compared with $163.5 million a year ago. Diluted earnings per share was $1.12 versus $1.87 a year ago. The first quarter financial results, while disappointing, mask numerous positive changes in our business fundamentals. But before I address those positives, I wanted to discuss the drivers of our earnings shortfall of which there are two, C-BASS's results and the losses incurred.

Regarding C-BASS, we're joined on the call today by Bruce Williams, C-BASS's co-Founder and CEO. We've had so many inquiries this quarter regarding C-BASS and the turmoil in the subprime markets, that we felt it would be appropriate to have Bruce participate on our call. In a minute I'll turn it over to Bruce to discuss C-BASS's results.

The second item was that we took loss reserves up in the quarter by 15 million over paids, even though delinquencies fell in the quarter by 2500 to just over 76,000. The reason we did this is that the newer delinquencies we are receiving are from higher-dollar states and as a result the loans have significantly higher loan balances. As we discussed last quarter, the average loan size on our bulk business has grown from 147,000 in 2003 to 236,000 in 2006 and the flow has increased from 144,000 to 161,000. At our year end call we felt paids for the year would be up by 10%, even though the actual number of claims we paid should be relatively flat year over year.

In addition, we did not repurchase any shares within the quarter due to our transaction with Radian. While in discussions with them in January, we were unable to buy any of our shares and after the transaction was announced, we were limited by SEC rules to an immaterial level of purchases. After the transaction is voted on by Radian's and our shareholders, Radian's meeting is May 9, and ours May 10, we'll be at normal repurchase limits.

Now regarding the positive business fundamentals we are seeing, as we discussed last quarter, we expected to see an increase in the penetration rate for our flow product, and that indeed is what is happening. Even though the general mortgage market continues to be down, MGIC's flow new insurance written of $10.4 billion was up 32% over last year. Interest rate levels, slowing property appreciation, and MI tax deductability are all playing a role in the return of our flow product, although the single largest factor, I think, is a renewed interest in credit quality protection.

Bulk writing, while written at higher premium rate levels in this quarter over last area's levels have not yet rebounded as the interest rates charged to borrowers have not yet been sufficient to include our cost. While lenders are now building in such rate increases, it probably won't be reflected in increased bulk writings for MGIC until the second half of the year. Persistency at 70.3% also continues to improve, up slightly from 69.6% last quarter and 62% a year ago. Flow persistency was 74.6%, up from 73.5 last quarter and bulk persistency was basically flat at 56%. Now the quarterly run rate for the combined portfolio was 75.2%, up from 71.5% last quarter.

As a result of the improvements in penetration and persistency, insurance in force grew to 178.3 billion, up 1% from last quarter and 6.8% from a year ago. The average premium yields held slightly to 67.4 basis points from 67.9, reflecting the much lower levels of bulk writing in the quarter, even though as I mentioned, the average premium rate on the bulk was significantly higher than last quarter and last year. The other strong influence was a continued runoff of the higher-priced '02, '03, and '04 books of business. Paid losses in the quarter totalled $166 million up from $157 million last quarter. While some are expecting a sky is falling scenario for paids for the year, we continue to see a year much as we thought, with paids up approximately 10% from last year.

A caveat would be California, however, where we have seen a marginal uptick in delinquencies. California has a 120-day disclosure period compared to most other states which they have a 12 to 18-month period. As a result, you can get claims much quicker there. Also, the average paid claim on California loans is two times higher than the average paid claim from the remainder of the country. So if the majority of these delinquencies don't cure in accordance with our claim rate assumption, you could have additional paids. We have not seen that in our development and we'll know more at the end of the second quarter and update you accordingly.

The subprime business continues to be a hot topic, particularly the 2006 book and rightly so given the many credit practices that were abused last year. Our industry and our company did not participate a great deal in these practices, as most subprime loans were originated with 80/20 piggybacks that avoided our product. MGIC did insure approximately $4 billion of subprime business last year and by subprime, I'm talking about all loans which have FICO stores below 620 and about 7% of our 2006 insurance writings, we have subprime exposure and relative to our entire book, we have 4% of our risk enforce. However, regarding the 2006 writings, which have generated so much publicity, it will not be one of our better books and its loss development should closely match our 2000 book. However, even with the poor loss performance, through the combination of higher premium rates and loss deductibles, we still expect the 2006 subprime book to perform at a combined loss and expense ratio of 80% and provide a 20% margin to MGIC.

Regarding our international expansion, we, after recent positive meetings with the Australian regulators, we expect to receive our insurance license by the end of April. We also are in final stages of discussion with the rating agencies and don't see any obstacles to securing our necessary ratings. We continue to have numerous positive customer discussions and expect to hit the ground running once our license is received in April. In Canada we are still targeting the fourth quarter to obtain our license and start writing business. We hope to have our full management team in place in Canada within the next 30 days, and again, like Australia, we have had many positive discussions with customers.

Finally, regarding our merger with Radian, we're pleased that we did not receive a second question under HartScottRodino. So the waiting period before we can close under the Federal Antitrust law has expired. The principal remaining legal hurdles to closing are twofold. The first is approval of each of the Company's shareholders, the proxy materials were mailed earlier this week and as I mentioned, the votes are scheduled May 9, for Radian and May 10, for MGIC. The second hurdle is approval of various state insurance departments. We expect these approvals to be completed late in the third quarter or early in the fourth quarter. In the meantime, we've had numerous discussion ongoing with coworkers from both companies regarding the integration with a great deal of work expected to be completed before our closing.

Now before we take questions, let me turn the call over to Bruce Williams, C-BASS's co-Founder and CEO. Bruce and his fellow co-Founder, Saul Sanders have created the premier company in their space, one that quarter over quarter has turned out returns of over 20%. Unfortunately, the events in the first quarter broke that string. However, longer term, these same events with C-BASS's strength, I think, will benefit C-BASS and MGIC. Bruce?

Thanks, Curt. Good morning. This past quarter the subprime market experienced a substantial repricing and increase in credit risk premium. This is primarily due to the well-publicized issues with subprime loans, especially the 2006 vintage to include poor underwriting, increasing delinquencies, and foreclosures. That has created liquidity issues for all subprime participants and caused the following, a dramatic drop in origination volume, accelerated subprime originator consolidation and bankruptcy, increased margin calls from all lenders, and significant tiering for subprime asset-backed issuers, plus a lack of access for certain issuers.

Demonstrating the strength of our franchise, we were able to continue to access the market, pricing four whole loan deals, a CBO, and a second lean deal in the first quarter. In anticipation of the changes in the environment, we tightened our purchase requirements and increased our expected losses in late 2005. This led to a reduction of our loan purchases of 2006 originated loans by approximately 50% versus 2005 originated loans. In addition, our early pay defaults are running well below the industry average due to these changes. However, like every participant in the market, we were surprised at the speed and over reaction of this past quarter's adjustment and have not been immune to these market events and responded accordingly.

From a liquidity perspective, we have reduced our whole loan position/exposure from $3.9 billion at the beginning of the quarter to $2.6 billion as of March 31. Total assets declined $1.9 billion from $8.8 billion to $6.9 billion at March 31. We began the quarter with approximately $300 million in cash resources. After absorbing margin costs of approximately $200 million, we ended the quarter with approximately $200 million in cash resources.

Due to these significant changes in the market, our first quarter results and full-year forecast has been negatively impacted. For the quarter, we reported a pretax loss of approximately $15 million versus an expected pretax profit of approximately $60 million. The major reasons for this shortfall are as follows--the write-down of approximately $34 million in our existing securities portfolio. This is due to both spread widening and an increase in our loss assumptions for our 2006 book. We realized securitization losses of $11 million for deals issued in the first quarter due to wider spreads. We had a mark-to-market adjustment of $34 million and our home loan portfolio to reflect current market conditions. In addition, we wrote off $21 million in outstanding claims to certain impaired counterparties.

These results were offset by an operating profit, and written loan servicing, interest income on our portfolio, and reduced compensation expenses. A substantial portion of the first quarter shortfall occurred during the last week of February and in March as the market continued to deteriorate due to announcements relating to New Century, Fremont, and others. Results during this period also led us to revise our 2007 forecast. We expect a return to profitability in the second quarter of 2007 and expect a pretax return for the year of 15 to 20%, which translates into approximately 150 to $200 million in pretax earnings for the full year. This assumes that our credit portfolio continues to perform within our current expectations, our whole loan volume remains at depressed levels, and spreads do not widen significantly from here.

Lastly, additionally in the quarter, we entered into a definitive agreement to acquire Fieldstone Investment Corporation. We view Fieldstone as a long-term strategic asset. Fieldstone filed a preliminary proxy statement with the SEC on March 27, 2007. Therefore, at this time, I am unable to comment on this transaction other than to reiterate information contained in the publicly-available documents.

Thanks, Bruce. With that, operator, we'll take questions for either MGIC team or Bruce.

Thank you. [OPERATOR INSTRUCTIONS] Our first question comes from Geoffrey Dunn of KBW.

Curt, so I heard right, are you reiterating your 10% paid loss guidance for the year?

Yes.

Okay. Is there any way you can give us any parameters around the provisioning? I know it's a hard number to project, but obviously that's the real wildcard in the MI business. How should we be thinking about how that number trends from this quarter? Can you put any kind of range around that to help us firm numbers?

This is Mike. I would say at a minimum, we'd be at this level going forward. It will be subject to the level of delinquencies. As Curt talked about, one of the things that we're seeing in the underlying delinquencies is that there's been a big shift, if you will, in the pre '03 books are declining rapidly, but being offset by growth in '04 and '05 and '06 and '06, or really '05 and '06 books. And the average severity there is significant. The underlying loan coverage, as Curt talked about.

Another thing that happened, as Curt mentioned, about California. A year ago, California delinquencies were about 1900. Now they're up to 3400. So we're seeing an increase in California delinquencies. As a matter of fact, even though we had a reduction in delinquencies overall quarter to quarter, about 2500, we had an increase in some areas like California, Arizona, Florida, Mass., New England area of about 1500. And those delinquencies obviously carrying much higher loan averages. That's what's happening with the underlying coverages.

As a result -- even though we didn't have much of a change in severity, that is delinquencies going to claim, only in a couple of small areas, the major increase in reserve factors were changes again in average severity because of the average loan size. Now, going forward, I would say that we'd be running at the current level of 181 million quarter to quarter, subject to changes in delinquencies if they, for example, they tend to decrease again in the second quarter, historically, they have, anyway. And then some increase in the second half of the year. So subject to that change and the mix again, but, again, we'll have to look at that as Curt mentioned in the second quarter.

Okay, thanks.

Our next question comes from Steve Stelmach with FBR.

Good morning, gentlemen.

Good morning.

With your stocks trading so close to book value, I guess investor's concerns are, how safe is book value? And outside of C-BASS and Sherman, anything on balance sheet that you guys can see that may impact book value going forward? And then also, looks like you delevered a little bit in the quarter, any--?

No, that was just the fact that -- if you remember, we had a couple hundred million coming due in March and we took the opportunity and raised the money earlier in the fourth quarter and that was just the planned reduction of that five-year debt that was coming due in March. That's the only difference in leverage. On book value, I'd say, no, there's no significant changes. The balance sheet is stronger than ever. Investment portfolio is strong as ever. As Curt mentioned, I think the fundamental business looked very good. We need to get through the bulge, if you will, in the '06 book, and '05 delinquencies, but no impairment on the book value.

This is really a strange quarter and maybe a year for the company in that I feel really good about the business fundamentals. Now, we're going through some issues here relative to subprime and its impact on C-BASS, and our business also to a limited extent. But relative to what's going on fundamentally behind the business, things are looking very good, but it doesn't show up, obviously, in the financial results, which is why it makes this call so difficult.

And then just one follow-up, I guess it's for Bruce. If I'm calculating my numbers right, it looks like C-BASS tends to run leverage around 9 times after you sold some whole loans, looks like you're running around 7 times. As better underwriting works its way through the system, do you foresee getting back to that 9 times? Do you have the capacity to get back to 9 times in this environment, or is the margin call just going to prevent you from getting there?

Well, I guess our view for this year is that origination volume for the whole industry is going to be down substantially, and we bought probably 10 to 11 billion of home loans last year, we expect that and that's in the forecast that I stated to be down about 50%. So, no, I don't think we're going to lever back up this year.

Thank you.

Operator?

Our next question comes from [Brian Hennessey] with [Tribeca] Management.

Question for the MGIC team. Could you shed some light on your due diligence into Radian's subprime exposure and its financial guarantee business? We know that there's 567 million of subprime RBS exposure and about 769 million of exposure to CDOs with subprime collateral. Just wanted to get a little better picture on how you've gotten comfortable on this exposure?

Yes, this is Larry Pierzchalski. With regard to the due diligence, we received loan level information on their total insurance portfolio. We ran it through the pricing models we have here and we're comfortable with the outcome. Quite frankly, it's not too different from some of the business we did, although the portions may be a little different. So anyhow, that's what the due diligence process was. Loan level information running through our pricing models, and we also had reviewed some of their NIM transactions and credit default swap transactions.

Thank you.

Our next question comes from Mark Patterson with NWQ Investment Management.

Hey, there. Question on the book value. Just in the roll forward, the book value at the end of the quarter was much higher than I would have expected it to be given the earnings level and the fact that FAS 115 would have been a reduction, it looks like and your $0.25 dividend. It looks like the book value is more than $1 higher than a normal roll forward would be. Does that have something to do with the Sherman dividend or what's really driving that?

I can't -- there's nothing significant. FIN 48 wasn't a big issue.

Well, book value is $1.76 higher than year end.

FIN 48 might have been $0.50, $0.52 or something, not that much more. So no significant factors.

Okay, I'll check back on that. One other thing. On C-BASS, since Bruce is on the line, did I hear what your -- your appetite, obviously you've been active, the Fieldstone purchase looks like it's a good return on investment, but do you have other opportunities you can take advantage of over the next two, two and a half quarters, or are you saying--?

No, no, no. Maybe I wasn't as clear as I should have been. We're seeing lots of opportunities. I was just stating the fact that it's our belief that the amount of product that will be produced this year versus last year is going to be substantially reduced--.

You're probably looking at a subprime business that was $600 billion last year, down to 3 to $400 billion this year.

But we are -- there have been some well-publicized distressed sales. We had seen those and to date we have not participated because we didn't like the pricing in terms of where they traded, but we expect that will continue.

Okay. So just one clarification, then. There are some distressed subprime properties, some portfolios that could be available, but you're concentrated on what you've done with Fieldstone and your own opportunities, even though the market is coming down in terms of volume?

Correct.

Thank you.

Thanks, Mark.

Our next question comes from [Adam Gileskie] of Goldman Sachs.

Thank you. Can you talk a little bit more about claims management? And in particular, I've heard from other sources that if a loan has been conveyed fraudulently, then the insurance is not necessarily valid? How does that impact your expected approach to claims management as this year progresses?

Well, this is Larry Pierzchalski. With regard to fraud or other reps and warrants, in particular on the bulk business side, when we do a transaction, we receive loan level information, so one of the reps and warrants is the accuracy of that information and aside from that, there are usually additional terms in terms of what's included or specifically excluded and fraud, certainly first-party fraud and generally third-party fraud is generally excluded. Typically, we would, at time of delinquency really check out all the rep and warrant items, contract items, and check for fraud, in particular on early delinquencies, where fraud tends to be more of the mix than later in life. So that's the process. Once again, we're looking for material misrepresentations. We've been doing that through all the policy years, it's not something we've started and I guess we're of the opinion that although we rescind from time to time due to various items, we try hard to clear those up with the insured as we see them and hopefully make that less of an issue on a go-forward basis. The last thing I'd say is some of the rep and warrant items that are in the marketplace between buyers and sellers of loan may not be the same as the rep and warrants in items we have.

Yes. What I would like to say, Adam, is we've always been -- MGIC's always been very consistent relative to our approach on -- we've had a team here that investigates the business and long-term have done that. It's not something we gear up to when we have tough times or something like that. All our customers appreciate the fact that we're consistent. If the item's in dispute, we did not cause a change in the underwriting decision, if you will, those are things you let go. But if they are things that were material to the risk and fraud was present, then those we enforce.

So recognizing that the process hasn't changed, nevertheless, the marketplace has or allegedly has. So can you comment at least qualitatively on how that -- how much more or how more often are you seeing fraud in your current process than you were in previous implementations of the same process?

I would say that the percentage hasn't changed, but you get more opportunity to the degree there's more delinquencies, and obviously with delinquencies up, largely as a result of the slowdown California, Florida housing market, obviously there's more delinquencies. And if you look at more delinquencies, even if you're raters of fraud is the same, you'll discover a few more cases.

Again, I don't think a lot of the business that we did that you read about in the subprime space is business that was insured by our industry. So it's not as though these are perpetrated on MGIC or our industry, those were done with piggybacks, no income, no asset, or stated income-type programs that avoided any check by anybody. So those just weren't things that came to our industry.

Thank you.

Our next question comes from David Hochstim with Bear Stearns. David, your line is open.

Thanks. I wonder if you can just talk a little more in detail on the uptick in delinquencies on the reduced stock loans, and are you seeing anything strange or unusual there, everything else performs relatively normally from a seasonal standpoint in terms of delinquencies, but that portion of the book didn't.

With regard to the uptick in delinquencies, whether it be alt-A or subprime, in my mind it has a lot to do with the slowdown in California and Florida. Those markets were very hot. Appreciation, 20% per year. Appreciation cures a lot of ills. On the bulk business and in the capital markets, Florida and California were a big piece of the mix. In our bulk book, recent writings probably are 35% or so. So to the degree those markets cool down from the 20% per year appreciation to let's say more normal in-line with the rest of the country, the true nature of the risk shows up. If you remove the impact of California and Florida in some of these mixes, alt-A, subprime, I don't see, at this point, much performance difference to date.

Another thing, David, is just the mathematics of that. We've written more business within that space. I think on a percentage basis, it's not out of line of where it's been the last couple of years, relative to 8 or 9%. So I think it's just that we did a little more business in there, we had a couple large originators that we trust their processes and we did more business in that space with them.

Is it possible that you could provide us with the historical presentation of bulk versus flow delinquencies in addition to this new disclosure, so we could get a sense of how--?

That's, actually, David, in the text of the press release, we give you that bulk and flow delinquency rate, really doesn't impute the actual numbers, but.

I think it goes back to 2005, doesn't it.. Do you want longer than that, David? Is that what you're asking?

Well, we have the old stuff.

The bulk and flow before this change here. But in the text, there, we give you the delinquency rates, David, so we'd have to impute what the number would be to get there.

Okay. Can I just ask Bruce to clarify what he said on the mark-to-market adjustments and how much of that -- he said there's a $34 million spread widening and increased loss assumptions and how much was each piece? And if you could just review the other mark-to-market adjustments that drove down earnings that weren't cash flow-related or earnings-related?

Yes. You can see the $34 million, about half was spread widening and half was an increase in our loss assumptions. I mentioned the $11 million in realized securitization losses. I mentioned a number of deals. We issued over 3 billion in the quarter. If you take that 11 million against that 3 billion, that's a pretty small difference in terms of what happened in the market. Our home loan portfolio, as I said, we reduced, but home loan prices dropped dramatically and most of that is in anticipation of where we can securitize things out, which is now at wider spreads. If you look at each of our three deals over -- our deals over the quarter, our home loan deals, we priced exceptionally well in January. The last week of February, we did price very well and widened out very substantially in the third deal in March. A lot of that $34 million is due to what are current spread levels in the market and then the $21 million is -- we have a very robust claim process been claiming on our origination counterparties for the past decade. Obviously, there were some surprises in the last three to four months in terms of some major originators going out of business, not the least of which was a big surprise, which was New Century. That was a fairly dramatic change in a very short period of time.

Is there a possibility of some recovery from that, or that's really gone?

Subject to where they liquidate out and what their book value is, we're hoping to have potentially some recovery. We don't expect to get 100 cents back on the dollar, but prudently we decided to write all of that off in this quarter.

Okay, thank you.

Thanks, David.

Our next question comes from Kenneth Posner of Morgan Stanley.

Thank you. Curt, I wanted to ask, if I could, one question about magic and one question about C-BASS.

Sure.

I think people in the market must be very concerned about losses, for all mortgage companies and one of the things they may have forgotten at the mortgage insurers is that a significant portion of your book is covered by captives. And we don't know the exact attachment points, but I'm wondering if you can give us a sense of at what level of chaos in the housing market would you start to see, at least some of your losses limited by those captives?

Well, with regard to the captive, kind of on the flow side, there's two forms. The two forms being excessive loss and the other form being the quota share. The majority of the excess loss transactions probably have attachment points roughly equivalent with foreclosure rates of about 4 to 5 per 100 out of the original population. Past books didn't come close to that '05, '06, potentially it could, we'll have to wait and see. The other form, quota share, obviously they're sharing proportionately on claims starting with the first claim.

Right. And so can you -- I don't know if you -- I don't know if you guys disclosed where the cumulative foreclosure rates are running, but can you give us a rough sense today where we are versus the 4 to 5% attachment rates that you have on some of these captives?

The early 2000 books probably half that.

Yes, I was going to say 2. Maybe these books -- Ken, I don't think -- to me it's a statement about how we feel about housing, I still don't think those excessive loss treaties are going to be touched.

Really?

And if they're touched -- which says a lot about what we think about housing. Half of our flow business is under those contracts, so what you're saying is a wonderful safety net for our company if things deteriorated to that level. I just don't see that.

Right. So it's got to -- and is there -- do the captives -- the captives don't start to repay until the deals have seasoned enough that you -- because you wouldn't get a 4 to 5% foreclosure rate immediately, it would take several years, right?

Exactly. It's our business, running -- you have a couple years before delinquencies hit and then claims in that third to fourth year.

Can I just clarify on your guidance for the 10% increase incurred this year. If I was reading the press release and doing the math right, in the first quarter the paids were up 20% year over year, and so--?

Well, the first quarter last year was an unusual -- I'm sorry, go ahead with your question.

Well, I just -- actually, I should clarify, is the guidance of 10%, is that the swap?

That's the annual.

That's the annual. Is that the paid or the expense?

Paid, paid.

Okay. So we should see the paid stabilize at this level and not go jumping up in the next few quarters?

Yes. As we talked last quarter, we thought paids would be an area of 670 to 680. We still feel that. The caveat, as I mentioned, is we are seeing that a slight uptick in delinquencies, which Mike mentioned in California, now those -- and they have a faster foreclosure rate there. So that's the only caveat.

That's right. And then if I could just ask one more question on the MI side. The net premiums earned were flat year over year. We were looking for a little bit faster growth than that. Now, there is better persistency, finally, after many years of waiting for it, and the insurance of course was up 5%. What are the -- I think you mentioned some things in your comments about some of the pressures on the premium yields, wat's your take on the net premium ARMs growth? Is that going to be indicative -- is that a blip, or are there still pricing pressures that have to be contended with in the business?

This is Mike. You'll start to see some quarter to quarter growth as we get into second quarter. So as we continue to move out you'll start to see the benefit on the persistency. The unique thing on earned premium this quarter, as Curt mentioned, was the low bulk business we did, but I think as we continue to move through the quarter, you'll continue to see a modest growth quarter to quarter on premiums earned.

Written was up about 1.5%, but earned was, as you mentioned, was flat. But it will continue to grow as we go through the rest of the year.

Okay, great. Then just my one C-BASS question. The $11 million write-down of securitization, I guess is tiny. It's 40 basis points of $3 billion in loans, if that's the right way to look at it. I just wondered if I could get the denominator for the $30 million write-down and just understand that as a percentage.

Of the securities -- you mean -- there's two different. There's the whole loans and then there's the securities.

Right. There's a $34 million write-down, I just--.

In the whole loan portfolio.

Well, the $34 -- our securities portfolio as of the end of March is $2.3 billion.

$2.3 billion. So the $34 million is actually a fairly small percentage that's a 1.5% markdown of that portfolio.

Yes. Most of the spread widening came in sort of the well-documented BBB sector.

Right, right. So for the overall asset collateral, it's a small amount. If I can just ask, there's been some deals, I guess, where people have bought portfolios at $0.95 on the dollar with some of us thinking that the collateral is worth $0.98 to $0.99. The assumption is that some extremely opportunistic buyers are stepping in and getting very good deals buying collateral from companies that are on the verge of going out of business, unfortunately. Are you saying that C-BASS doesn't find that $0.95 attractive, or is it just given that given the need for you to securitize loans and given the finite size of your balance sheet, you're just going to be cautious and careful and not to be looking to opportunistic buyers this year?

It's hard to answer that question. We have seen those transactions, and I think as I stated, we would probably have paid a different price, to be perfectly blunt. But that's not to say -- that's what makes this market really different.

Well, you all know what you're doing, so I'll leave it at that then. Thanks very much for answering my questions.

Welcome to the call.

Our next question comes from Terry Shu with JPMorgan.

Since everyone asked two, if I could ask two also. One on C-BASS. One more clarification. When you gave us the first quarter original expectation of over $60 million and said the $15 million pretax loss, there are all these whole host of items that you've discussed in very good detail. But just in a very top level general sense, how much of it would be because of reduced volume, how much of it is real economic loss write-off, and how much do you think is temporary kind of mark-to-market impact? Then you gave us the full-year number, 150 to 200. I think the previous expectation was probably 250 plus, I don't remember. And again the same question, how much is each component. Because mark-to market, if the loans mature, you get it back if there's no true impairment. So if you could just broadly speaking size it, how much is because of reduced origination, reduced volume? How much write-offs?

Let's just -- as I said, if you look at top level--.

Kind of big picture type stuff.

Okay. The top level numbers I gave you added up to about 100 million.

Right.

I would say probably 25 to 35% of that potentially comes back from spread widening or improved credit. In terms of the full year when we look at the 150 to 200, we're being -- we're not assuming that spreads tighten from here. We're assuming that the market -- these things take a long time. Those of us that have been in the market for more than the past five years, this is not the first time this has happened. It obviously happened in '98, a lot of people went out of business. If you think back to '98, and that context, it took the market a good 12 to 24 months to come back. So what historically happens with spreads is, they grind tighter, they stay very tight for a period of time, and then there is an event, that is the mortgage market. The event, we hope it's the only event, happens in this quarter and spreads blew out. It's very tough to avoid that.

Right. But assuming that the environment there are more bad news and such, it could widen some more, but this -- and it could have another hit. But longer term, I gather, the volume comes back. So of the maybe $100 million reduction, a big chunk of it is this volume coming down, some write-offs, and then finally just the current quarters' market to market. Is that right, sort of big picture?

Yes, big picture, right. We have reduced our anticipated volume for the year, certainly. We have--.

By a lot. Right. Because you said the whole loan by 50%, is that correct?

That was a year-to-year comparison. We did not anticipate in our original numbers buying anywhere near what we bought last year. We thought that the market--.

Okay.

-- was going to be down in terms of origination this year. We didn't think, and I don't think anybody in the industry thought it would be down 50%.

You see, questions one gets is that are there big write-offs yet to come in C-BASS? And I think you're kind of indicating confidence that you may have reduced earning power, but no material write-offs. Am I reading that right?

I would say that is correct.

Okay. Then another quick question on just the Magic. The loss reserve and paid trend, Curt, you had said that you -- other than maybe worsening experience in California, you're still confident of the 10% pay growth, however, reserving now is at a higher level. I gather that translates into still higher paid growth into '08, then. Because reserves then flows into paid losses. Is that correct, with a lag? So overall there is worse than expected experience because of higher severity?

I think that's accurate, Terry.

Okay, so when one goes forward, one kind of reduces one's earnings expectation on that basis; correct?

Into 2008.

Into 2008.

Right. Yes.

Thank you.

Thank you, Terry.

Our next question comes from Eric Wasserstrom with UBS.

Thanks. Curt, I was wondering if you could just repeat the comments you made about the bulk coupons? I'm not sure I totally understood that.

The bulk coupons. I'm not sure--?

Or the pricing, the customers building in--.

Oh, yes. Larry -- I'll let Larry take that. It's his area of expertise, one of many.

Yes. I guess the issue is that even though spreads have widened out, which makes our product more valuable, because without us, they would have more in the lower traunches and wider spreads, all that becomes more costly. Our pricing is up akin to the spreads in the marketplace because the appreciation in California and Florida in particular not being 20% anymore, but something relatively flat. So that impacts the risk, in our eyes. That increases our premium rates. And even though relative to the market, the spreads being wide, not much is being done because the premium has to come from somewhere, that somewhere being the interest rate to the borrower, the originators at the front end have not had time to put that higher interest rate in the note to the borrower up front, have those loans get originated close and put before the marketplace. So we think it will be maybe another quarter here before that process takes place and there's enough money, vis-a-vis the interest rate to the borrower, to pay for whether it be our premium rate or the higher spreads in the marketplace.

Yes, Eric. The originators felt they were going to do these deals with senior subordinated structures which unable to do and our product is one that allows them to do, but they didn't build it in to the right to the borrower. And those applications probably lag us by 60 days from when they can build it in to when we can see it, to where they can accumulate a pool that we would insure. So that's why we say, we don't think we'll see it until early second half of the year.

Just to translate then, that to the benefit, the pressures and drivers of premium margin, can you just sort of parse out what those may be?

You mean as far as size in the second half of the year on bulk writings?

For example, it seems like there could be some improvement there, that will increase bulk, which would buy us up the premium margin, but is there still an impact from '03, '04 run off or from more quota share business? That kind of thing.

The premium yield, both in the bulk and the flow is flat or up. So the yield in the flow is flat to up, the yield in the bulk is certainly up. The total depends upon how much bulk versus how much flow in force. And because the writings on the bulk first quarter and probably second quarter will be about the same, flow is a bigger percentage and that's why you see the overall premium yield down a tad because -- not because the yield in each channel is softening. In fact, it's the other way. But because the mix is shifting more to the flow than the bulk because of the lower level of new bulk writings. But you should see the runoff of the '03, '04 coming to a closure in the next couple quarters. So that will be very positive to the yield anyway, but not positive to our financials maybe.

And just finally, the year on year impact of quota share business?

Not much change.

Yes. We had a large originator that switched to quota share, but it's not a meaningful impact.

Great. Thanks a lot.

Our next question comes from Rob Ryan with Merrill Lynch.

Good morning.

Good morning, Rob.

I want to explore the area of percentage severity as opposed to the dollar amount of claims which you often refer to as severity. So if we considered the trend of what's been happening on claim sizes compared to the relevant risk enforce on the loan that actually went bad, what has the trend been there? What's been happening with your loss mitigation opportunities? That's my first question.

Let me try to answer that. Larry Pierzchalski here. I would say the loss mitigation opportunities in the current quarter compared to the past few quarters is relatively flat. We're seeing the same amount of acquisition opportunity, the same amount of presell activity, and so on and so forth. Another way to come at it is, the dollars we pay as a percent, factoring into the UPB and the coverage so the severity relative to coverage is basically also held the same. The reason the average paid is up is because of the balances moving through, but the mitigation opportunity isn't any less. Does that make sense?

Yes. So it's an issue of the more recent vintages that are going to claim, having higher average loan sizes?

Correct.

Okay. This whole issue of future percentage severity has something to do with certainly home price appreciation in the future. What do you consider to be a reasonable assumption for the OFHEO home price index at the end of '07?

I think you'll be probably in the 3% area. In the fourth quarter, it was 4.4% -- I mean 1.1, 4.4 annual. So I think it will probably be for next year, 2 to 3%.

And could you contrast that with what people are saying about National Association of Realtors and median home price?

Well, that median home price is -- the repeat home indicator is much better relative to, at least my perspective. Median just -- it depends on what price homes are selling. So if high-priced homes are selling, the median it goes up, and if lower-priced homes are sell, it goes down. Probably for our business, because we're lower priced homes, the less the median goes up, the better, because it means lower-priced homes are selling. The realtor's number doesn't do a lot for me or our company. We take much more stock in the repeat home sales on what a home sold for and now the repeat sale of it, what happened to it over that interim. That's a true indication of value. Not the median.

Very good. Thank you.

You bet.

Our next question comes from Wayne Smith with Touchtone Investment.

First question I have is on C-BASS. You mentioned the loan portfolio was $2.3 billion and I'm just looking at the presentation you guys gave for Radian's conference call in November and on that you had to break out as of September 30, 2006, the amount of paper of the securities you had in the balance sheet, the -- 90% of that was BB-rated or lower, 45% of it was unrated residuals, BBB was 16% and I'm just trying to get my arms around the markdown you put on that. It seems like $75 million on a $2.3 billion loan portfolio, the bulk of which is lower-end residuals, it just doesn't seem to make any sense given what the market has done. Can you just give me color on why it's so low.

I don't think it's so low. It's $100 million.

Yes, but I mean, BBB paper is down 30% for the year-to-date.

Let me give you the categorizations. As I said, the securities portfolio is $2.3 billion. The BBB part of that portfolio is 18%. The A, AA, and AAA bucket is 4.5%. The BB is 30%, the B is 3.1%, the nonrated is 44.5%.

So it's about the same as it was?

It is about the same, and again, as I said, we priced $3 billion worth of issuance. We have a very good standard in terms of the spread we executed at in terms of our BBB portfolio. So we are marking it based on where it clears the market. We are not -- we are not a generic issuer. We are pricing and trading well inside of that due to the franchise value that people associate with how we look at our credit and how our credit has performed.

To what extent are you affected or would markdowns would not happen until the rating agencies downgraded the paper. Because there's been a lot of talk about most of this stuff will not get mark to market until the rating agencies actually downgrade it. Is your portfolios impacted by that or did you mark everything to what you thought the clearing price was right now?

We marked to the clearing price. The rating agencies are irrelevant to us.

So if there was no further deterioration on the quarter prices out there in the marketplace, you should not have any further markdowns?

Subject to the credit performance inside the portfolio meeting our expectations, that's correct. We don't think -- yes. Spread widening, we're through the fall of '98. In my mind, and again I have a pious view. I think if I look at spread product in the mortgage business and where BBBs are trading versus corporates or high yields, I like BBBs at some of these spreads for mortgages.

Okay. And then just -- and this is for both guys, for C-BASS and for MTG. The assumption of -- well, I guess it is just for C-BASS, because the assumption for MTG, is 2 to 3% HPA this year. Is that similar to your expectation of C-BASS?

No, ours is less. We assume flat.

You're assuming flat?

Nationally -- but again, it's very specific by geography. And we bias our purchases in terms of where we think the best housing price appreciation may happen.

Okay. And then just one--.

Very, very specific.

I think that's a wonderful point. The national number isn't that meaningful either. It's where you have concentrations of risks and where you take that risk of who you are really going after. We don't pay much attention either to that national number.

Okay. Then just one last question. On NFI, for the MTG guys, NFI has been having well-publicized problems. You guys wrote all the deep coverage mortgage insurance for NFI it looks like over the last two years of probably 9 of 11 or 12 of the securitizations they have done. Can you just talk about what level of exposure you have to those guys? Is that the bulk of what you talked about of subprime, because it sounded like your subprime was pretty low.

Yes, our subprime is fairly low. I think Larry's got the list of percentages.

NovaStar has been and continues to be a good customer of ours. We're happy, satisfied with the performance of their business and I think some of the issues there dealing with today really isn't so much about the performance of their loans, the collateral.

It's more a liquidity crunch and Larry and I and others of the team have met with their team on a couple of occasions and that business with MGIC relative to peers has been very good. Now, if there are issues there on how they move forward, you have the right to pull servicers and things of that so that the existing loans are serviced properly. But our experience there has been very good.

So you're saying that their problems aren't related to the facts that they're underlaying credits or people are saying that they think they are going to go bad? I'm confused?

Well, I think this is a lack of liquidity in the marketplace for warehouse lines is making it very difficult for people to continue to do business regardless of how their credit is performing.

All right. So you guys, the fact that they -- it seems they took out 30 to 40% coverage insurance on the bulk of their securities?

We need to go to new questions.

Wayne, this is Mike Lauer. We can talk about this, but we've covered in the past how we've priced bulk, we go through selection process and so on, so to say we've covered 100% of NovaStar product is an incorrect statement and we can go through that in more detail, but we're not going to--.

All right, I'll catch you guys offline. I understand.

We're not going to make comments about whether any entity is going to stay in business or not stay in business.

I understand. We'll talk about it offline. Thanks, guys.

Our next question comes from Andrew Brill with Goldman Sachs.

Just wanted to run through this question again just to make sure. You guys didn't make any changes to your severity assumptions, so you're still running losses about 80% of max exposures? And you didn't make any changes to claims rates, so you're still seeing the same percentage of delinquencies? And you take the higher losses, lower prices have run their course, so really on a go-ahead basis, we should expect that the average reserves per loan are in-line with first quarter levels?

You said a number of things there, so Larry and I will start. But we have made changes to severity, significantly in the quarter. That was the biggest piece of the adjustment as I talked about because of the average loan size. The point that we made was there aren't, as someone else talked about earlier, there aren't a higher percentage of loans going to claim. There's still about the same number. It's just that as they're going to claim, they've got higher underwriting loan balance, higher exposures if you will, and we're changing factors for that. With respect to your question about, did we make any changes to claim rates, I would say modest rates, not significant, the significant aspect of the adjustments had to do with the average severity.

Are you still seeing losses at about 87% of your max exposure? I think those are numbers you've given out in the past.

I think we were at 83%.

And we're using that calculation from the numbers we disclosed versus earlier we talked about the call, the exact delinquencies when Rob was asking that question.

And then there was one other thing you mentioned there, something along the lines that the higher loan balances have run their course. That is not true. These newer books are ramping up through the claim curve. Curt mentioned the--.

The California, Florida books.

And in particular, California and Florida now given them becoming more normal than where they were. That's yet to play through. We're just seeing the beginning of that.

So going forward, then, we should expect that the average reserves per loan continue to rise?

Yes.

Okay. I'm just having a little bit trouble reconciling that with your comments that we should expect $181 million of incurred losses subject to delinquencies.

No. I think the question was, what are incurreds going to be, and what I indicated was, at this rate, you could look at this rate subject to the change in the level of actual delinquencies and subject to these changes we talked about relative to average loan size. For example, this quarter, some analysts may have thought that the incurreds would have been less than paids, okay? The point is no, not, because of these factors.

Is there any way to think about the average reserves per loan on a go-forward basis?

No, because of the mix changes.

No, changes on the mix -- again, it gets back to risk is very localized. As we have improvements in the Midwest, now you're seeing, as I said, marginal upticks in the West and the South relative to delinquencies. It's too hard -- and based on the concentration of a mortgage insurer in those markets, that makes all the difference.

Just finally, just given what we're seeing on the home price front, why shouldn't we be more concerned that at some point in the future you do actually have to make a change to your assumptions for claim rates or severity as a function of total expected losses?

Yes, well, we could. At some time in the future, claim rates are either going to improve or deteriorate and we'll make changes accordingly.

Thank you.

Our next question comes from [Jonathan Grassy] with Piper Jaffray.

A question regarding the low dock loans. I see that they were released this quarter. I hasn't seem them in the previous quarters. Is there any particular rationalization as to why those released now? And following that, can we expect or do you think any -- there's going to be any additional provisioning for these style of loans, or do you think they've troughed? What is the possibility of seeing some additional provisioning?

Jonathan, this is Mike Zimmerman. The reason we changed the disclosure is that for the most part, most folks that I talked to are looking at the market irrespective of the channel of acquisitions. So whether it's prime, subprime, A minus or reduced dock, that's how the market is analyzing the performance of both mortgages and mortgage insurance. So we thought it would be more appropriate to reflect that as a category of lending activity, if you will, reduced dock versus the full documented loans.

Given the guidance on originating, reduced dock loans, and I think even the salability of those loans in the secondary market, you would think you'd see less of that business going forward given the complications in originating and marketing those loans.

Do you think the trends you're seeing in those low dock loans are different than the industry, higher, lower, about on par?

And what trends are you talking about?

Just the delinquencies that you're seeing for low dock loans.

I really don't know why they'd be different than the industry.

What we insure probably has different profile and come from different originators than the general market. We look at our performance, how it compares to the general market, offhand, I can't tell you.

Our average loan to value is lower and I can't necessarily compare it to the overall market. I don't have that statistic, but on average our loan to value for reduced documentation loans is around 80%.

Yes, I think -- because Mike and I were asked this, I believe, where we did run the appraisals on those, and I think our effective LTV was 69%, so both the subprime and lower dock loans. So maybe we have a lower LTV because we did a lot of adds through our bulk channel.

Okay.

Thanks.

Thank you.

Our next question comes from Brad Ball of Citigroup.

Thanks. Bruce, just a clarification on your guidance for 15 to 20% ROEs this year. Are you basically saying that you're not expecting either an expansion of spreads or a tightening of spreads, you're basically assuming the current environment? And did I understand you to say that the sale trends in the latter part of March were worse? And if that's the case, then how do you have confidence that things won't deteriorate further?

In answer to your first, you're correct, we don't expect -- we're assuming not a significant spread widening or spread tightening. If you look at spreads over the last three weeks, we priced the deal the third week in March. Things have generally calmed down and are more stable. The market -- there is a period from the last week of February through basically the first three weeks of March where the market was very unstable and it seems to have calmed down. Spreads seemed to have found a level. It's not to say that they can't go wider from here, but these are pretty much the historical ways.

Okay. When you say--.

There's a limit. We have like 15 people in queue. I really need--.

I've got one for you real quick, Curt. Would you care to comment on the prospects for your plan to sell down your exposure to C-BASS, any changes in that positioning?

No. What was presented on the call that we did is still how we're anticipating the move forward.

Okay, thank you.

Our next question comes from Josh Smith with TIAA CREF.

Thanks. Just quickly a comment on the book value. I think you only earned $1.12, you added a $0.25 dividend and the book value went up by $1.70 so there's something missing there. I'll follow up later, but there's clearly something missing.

That was $0.85 I think I said $0.55, it was $0.85.

What is 1048 again?

FIN 48?

What is FIN 48?

Tax change. Tax disclosure. Foot note.

Is that a one-time issue?

It's a book adjustment.

One time.

One time, great. A question for the C-BASS gentleman. On Fieldstone, initially, you talked about, it was disclosed that you were offering $7 plus and it ends up being round $4 at this point. Was there any thought to waiting for Fieldstone to go much lower? Didn't seem like there was a lot of other people bidding for this and we're seeing all these companies go bankrupt. Why not just wait, especially with respect to how it might impact the larger merger of Radian with the focus on subprime?

I'm not really at liberty to talk about it, because we're in a quiet period.

He's not able to talk about it.

Right. Okay.

Thanks.

If I could just remind everybody. Again, we still have a number of people in the queue. We want to accommodate everybody. If you could limit it to one short question, it would be helpful.

Our next question comes from [Ross Webin] with Arbiter Partners.

Hi, guys, if we're looking at industry-wide default rates, it seems as though a quarter where in the aggregate they were down for you guys is probably better than most people were doing. I know that there's been sort of some shift in defaults towards some of the higher priced areas that you mentioned, which I guess implies not only that you guys are doing better than the industry in terms of aggregate defaults, but that there must have been areas not in the high-priced areas where there was considerable declines in defaults?

Right.

If you could help us understand how you guys are walking through the raindrops, as it were?

Well, I mean, I think it's just a -- I mentioned we had a 2500 decrease for the quarter and we really had an increase of about 1500 in some of the other states that I mentioned, so everything across the board, primarily even quite surprisingly, Michigan, Ohio, Texas, Illinois, Georgia, Indiana, North Carolina all had some pretty healthy decreases in the quarter. So it's seasonal. I would remind you again that we talked about the fact that this book is seasonal. We normally do see a decrease, but it was a good decrease in Texas and Michigan and Ohio. So that was pleasant. Whether or not that continues is going to be another issue. Primarily it's seasonal.

One other question on the claim rate. You guys mention that materially you're keeping it essentially the same in terms of your assumptions for loss in the event, or when you have defaults reported. I guess that seems a little bit unusual given what's been going on with -- in terms of housing price appreciation, and sort of what we hear in terms of increases in foreclosure rates. At what point do you start to see a pickup in claims as a percentage of defaults and at what point does that start to sort of work its way into your accounting ahead of the actual cash flows out?

Let me clarify. The reason I said -- talked about the claim rate not moving much and we have to go back in time, and that is that in '05 and '04, we saw significant increases in claim rates, primarily in the Midwest. Whereby, most of the reserve increases that we had were being generated not as much by the level of delinquencies, but rather by the percent of which loans in the Midwest were going to claim. What I said earlier was we didn't have much of a change in claim rates. There was some modest increases in some markets, but not significant. So I guess what I'm saying again is that overall the major change in the reserve has to do with the underlying loan factors. There were some markets where we did increase the claim rate. Obviously, California as we talked about and Florida. But I would say in the Midwest, only a modest change, which was pleasant. We hadn't seen that and we've seen that for the last few quarters, although we do have high delinquencies in the Midwest, the amount at which they are going to claim and the claim rate by which they are going to claim has not changed materially quarter to quarter.

Do you expect the same for 2008 or later 2007?

No. I think as this book changes and let's -- I think about where the new markets are, Florida and California, if they continue to, if you will, send off delinquencies, then they will begin to go on at higher rates and much higher dollars. So they could have a material change because of the size of those markets.

You also have to understand, as Mike said, we had a huge concentration in the Midwest which was running at high claim rates. As you have improvement there, that can offset increases in claim rates elsewhere. So others didn't have quite that exposure to the Midwest that we did. So we're coming at it from a little different perspective.

Thanks. One last question then I'll drop off.

Can we get you back in the queue.

Sure.

Thank you.

Our next question comes from Samuel Crawford of Stone Harbor.

Good morning. Thank you very much for taking the question. You've made it clear that you look forward with a good degree of confidence at the housing market generally. I'm wondering if I could get you to drop back and think about home price depreciation in 2007, perhaps just pull it forward as the first half of 2008; but not at a national level. National statistics to me are a little vague. Can we talk about the two big markets, California and Florida and in your comments would you mind indicating what kind of expectations you might have for the mix of foreclosures versus new home sales that might have a flow through affect on home price appreciation.

California. This is Larry Pierzchalski. California, just to give you a few of the things we are seeing in California, a month's supply in January of '07 is roughly about nine months' supply versus I think six a year ago, so that's, in my mind, a leading indicator. Things are doing better in the North. The North would be four to five months and the South, closer to ten months' supply. So the North is, let's say, in-line with a more normal market, four months' supply and goes back to some of the earlier comments. It's not just the national housing market, there's all sorts of submarkets and the North is doing much better than the South and California. The upper end in California, 500,000 to 1 million is doing much better than the low end. Single family detach is doing much better than condos. So pick your market, but -- so if we ensure more in the North than in the South, obviously we'll be probably anticipating a little better home price appreciation than maybe in the South. But I'd say to try to answer your question more directly, I'd say California and Florida, if you look at Moody's.com and OFHEO and this and that, probably flat to down slightly, once again depending upon the submarket within those states.

All right. Thank you very much for dealing with all the questions.

Thank you.

Our next question comes from Matthew Roswell of Stifel.

Yes, Stifel Nicolaus. I will keep this very, very short. Turning to Sherman, given that credit card performance have been quite good recently, should we have any concerns that Sherman could go through similar issues as did C-BASS?

No, we've not seen that, it's performing well. I guess in our guidance for the most part is flat, if you will, to the last year. They had a very good year last year and we haven't seen anything change and the credit card performance has done well. There has been -- this quarter, there was a little bit less purchase activity because of competition, but the offset was the credit card bank, but nothing significant.

Okay. I'll follow-up with my other questions with Mike.

Thank you.

We've got time for one more.

Okay. Our last question comes from George Sacco of JPMorgan.

Hi. One number actually, and then I have a question. The number is, if you could just tell us what the SingleFile was as a percentage of your new insurance written? The question is, the last couple of quarters, we've seen a pretty big increase in the 95 and higher LTVs as a percent of your new insurance written. Just wondering -- if something has changed there pricing wise or what's really driving that increase, especially given that house prices seem to be clearly showing signs of weakening at this point. I would think that some of those higher LTV products would be, at least appear to be more risky at this point.

SingleFile was 10.4% of our writings -- of our flow writings in the quarter, I think it was 9.9 for the year last year.

I think it was 7.5 last first quarter, if I'm correct.

Yes, year over year, but I'm talking about for the whole year. As far as the pricing, nothing as changed. What's happening is a lot of that business that was done with 8010s and 8015s and 8020s is now flowing back to our industry as 95s or 100s. So you're just seeing the change back given our product is a much more favorable product relative to those, because they're not available.

Okay. Okay. So it's not, it's more just an assumption of the market, not necessarily your risk appetite or pricing?

Not at all. It's all related to the fact that we're getting a return in penetration from borrowers that we didn't have a chance to get in the last few years, anyway.

Yes. Okay.

Okay, thank you, George.

Thanks.

With that, I think we've had a full morning of questions. I do appreciate your interest in your company. As I said to start this, we were disappointed with the financial performance, but again, and also what happened at C-BASS, but I think C-BASS will be better coming out of this with the opportunities that it has and I know the fundamentals behind MGIC's business are ones that I've been looking forward to for a long time. I'm still very optimistic relative to the Company and our business. Thank you.

Ladies and gentlemen, this concludes today's program. Thank you all for participating, and have a wonderful day. You may now disconnect.