Radian Group Inc (RDN) 2006 Q3 法說會逐字稿

Thank you for standing by and welcome to Radian's third quarter 2006 earnings conference call. At this time all participants are in a listen-only mode. Later there will be an opportunity for your questions and comments. Instructions will be given at that time. (OPERATOR INSTRUCTIONS) As a reminder, this conference is being recorded. I would now like to turn the conference over to our host, Senior Vice President of Investor Relations and corporate communication, Mona Zeehandelaar. Please go ahead.

Thank you, operator, and thank you for joining us today. With me on the call are S.A. Ibrabim, Chief Executive Officer of Radian; Bob Quint, Chief Financial Officer; Mark Casale, President of our domestic mortgage business; and Steve Cooke, President of our financial guaranty business. As I do every quarter, let me remind you that today's conference call will contain statements that are forward-looking. As you know, these statements are based on current expectations that are subject to risk and uncertainty and Radian's actual results may differ materially from those expressed or projected in these forward-looking statements. Factors that could cause Radian's actual results to be materially different than those in the forward-looking statements are described in the Safe Harbor statement that is included with our webcast and in our periodic reports including item 1A of part one of our annual report on form 10-K for the 2005 fiscal year, and in item 1 of part two of our quarterly report on form 10-Q for the quarter ended June 30, 2006.

Today we will follow our normal format. S.A. will begin, followed by Bob who will review key quarterly financial metrics, and then we will take your questions. For those logged onto the webcast at www.radian.biz, slides are provided as background and should complement our remarks today. Included in this slide is a breakdown of our MI loss reserve increase for the quarter, slide 15, and a brief primer on smart home. We will beholding an investor day on November 9th in Philadelphia at the National Constitution Center. Presentations will be made by Radian, C-BASS and Sherman management. Please contact me for information on registering. Now it is my pleasure to introduce you to S.A.

Thank you, Mona, and thank you all for joining us today. I assume you have seen the news release and the additions to our MI loss reserves. Bob will go through these numbers in detail shortly. Clearly our results for the quarter were not what we wanted them to be due to the impact of the additions to the reserves. Importantly, though, other aspects of the quarter were positive. We grew book value again and I am pleased with the strong performance we've achieved in the first nine months of the year which included earnings per share growth of 10% year over year. Before I comment further on the quarter I would like to highlight three key points.

First, Radian is a diversified company with a portfolio of synergistic businesses. This diversification provides us with alternative sources of revenue and earnings. Second, we place strong emphasis on risk management. Our risk is diversified across asset classes, products and geographies and also distributed into the secondary market when appropriate. Third, we are focused on creating long-term value and are committed to prudent balance sheet capital management. We have a seasoned management team with broad and deep experience at Radian, as well as at other leading financial institutions.

Now I would like to give you my overview of the third quarter. The quarter's results were largely impacted by the MI loss provision. The primary drivers of our addition to reserves, which are shown on slide 15 of the webcast slides, include one transaction that is a first loss, second lien quarter share deal. Also new delinquencies and aging of existing delinquencies and an increase in severity resulting primarily from a larger average home loan [side]. Bob will provide more detail. The key takeaway is that our loss reserves are strong and appropriate.

In the quarter we saw several positive MI trends, including strong MI flow volume and increased persistency. The financial guaranty and financial services businesses performed well and as expected. Although third quarter net income of $112 million and diluted net income per share of $1.36 both trailed the year earlier quarter, net income for the nine months of 2006 was ahead of 2005 first nine months. And importantly, diluted EPS was up 10%. We continue to accomplish our goal of building book value and maintaining our focus on ROE. While ROE fell slightly below our target this quarter, we remain well within our 12 to 15% range for the first nine months of the year. And this is important.

At September 30, 2006 our book value per share of $49.11 was 14% greater than a year ago and adjusted book value per share of $67.68 was ahead by 11.4%. Adjusted book value is a non-GAAP financial measure and a reconciliation to book value is included with the webcast slides available at www.radian.biz.

As you know, share repurchasing has been an important complement of our capital management strategy. In the third quarter we repurchased an additional one million shares at a total cost of $60.6 million bringing the year-to-date total to 3 million shares under the existing 4 million share repurchase plan. Since the end of 2004 Radian has repurchased 13.8 million shares or 14.9% of the shares that were outstanding at that time.

Now turning to our businesses, beginning with mortgage insurance, we continued to experience good flow MI volume which is benefiting from positive cyclical trends, stable interest rates and increased MI penetration. We did less structured business this quarter. This reflects prevailing market conditions with deep MI execution not being attractive and a tight credit spread environment. As we've said before, this business is by its very nature, somewhat lumpy based on risk and pricing, and we will continue to pursue this business with a disciplined focus.

Primary insurance in force at September 30, 2006 was down slightly from last quarter was $115.3 billion, up from $109.7 billion at year end 2005, and our risk in force remains steady. Reflecting the stable interest rate environment, persistency continued to increase on a twelve-month lag basis for the quarter and for the year. For the quarter persistency stood at 69.2%, up from 66.8% for the second quarter. For the twelve months ended September 30, 2006 it was 65.7% up from 62.8% a year earlier.

Commenting further on MI credit while our claims experience has been moderate over the past several quarters, we have significantly intensified our efforts with regard to claims management and loss [litigation] particularly for nonprime and [seconds] businesses. As a result, we have been adding more resources and enhancing our processes. While our process speeds up payment of easy claims we subject other claims to greater scrutiny, even if it means taking longer and consequently having a higher inventory of pending claims that can drive up reserves.

The point I would like to make about reserves is that our balance sheet loss reserve remains very strong. It is above the midpoint of our modeled range which is consistent with the second quarter of 2006 because of the continued uncertainty in the housing market. You will see enhanced disclosure in our 10-Q this quarter that depicts our modeled range of loss reserves for the quarter and for the comparable quarter. Our current loss reserve stands at approximately $24 million above the modeled midpoint.

Even though the unemployment outlook may have further stabilized as the result of recent positive numbers on employment and the pressure on borrowers cash flows with respect to high energy costs have significantly abated, there are still many conflicting views regarding the economic outlook. So we remain cautious. Going forward we are confident about the outlook for our MI business. Our focus is clear, to continue to position ourselves for economic uncertainty, to benefit from the cyclical upturn in the mortgage insurance business, to leverage our unique capital markets expertise and to effectively manage our credit risk exposure.

Turning now to financial guaranty, our financial guaranty business continues to perform well in a difficult environment with production managing our strong first quarter levels. You will note that net premiums for the nine months were up 20% to $186.5 million versus $155.7 million for the same period last year. Our structured finance business performed well this quarter despite the tight credit spread environment and with an increase in competitive pressures demonstrating again the viability of our AA franchise.

As I've mentioned before, the structured finance synthetic market is growing faster than the funded market, allowing us to provide the AA credit enhancement solutions which have been at the core of our success. Although public finance direct production declined during the third quarter, our nine-month year-to-date results are only slightly below our results for the same period of 2005. Traditionally there has often been an increase in activity during the fourth quarter. However, I should note that the current municipal finance market is characterized by the persistence of challenging market conditions, including declines in overall market issuance, in short penetration and refunding. We have held our own under these conditions and all the while continued to emphasize pricing and credit discipline.

We are pleased with the progress and prospects of our financial guaranty business and are continuing to invest in both production and risk management resources. Financial guaranty provides Radian with important balance in its corporate portfolio mix, especially in an uncertain economic environment and in particular because the business is not directly driven by the housing market. As in the case of MI, we have clear priorities for our financial guaranty business. Production, breaking stability, operational efficiency and risk management.

Finally, turning to our financial services segment, both C-BASS and Sherman continue to be important and steady contributors to Radian's results. We are pleased with the additional ownership we picked up in Sherman during the quarter. The senior management teams of each company will be featured at our investor day on November 9th, and we know you will be interested in hearing what they have to say. We will also be discussing Radian's international growth strategy during our investor day and providing more details on how we continue to leverage our mortgage risk solutions and form strategic relationships with global business partners.

Before handing off to Bob who will walk you more thoroughly through the financials, I would like to mention our recent appointment of Teresa Bryce as Executive Vice President, General Counsel and Secretary. Teresa comes to Radian from Nexstar Financial. Previously, she held senior positions at Banc of America mortgage, P.N.C. Mortgage and Prudential Home Mortgage. Teresa brings a strong and impressive background to Radian including her experience on the board of the Mortgage Bankers Association. Her significant legal and mortgage industry experience will be a great benefit to the company.

Bob will now provide additional information on our results for the second quarter. Following his remarks, we will take your questions and close with a few closing remarks.

Thank you, S.A. As always I will be providing some detail about the third quarter financial results. On the MI premiums side premiums are down from last quarter because of the runoff in the NIMs and second book from a relatively lower amount of cancellations this quarter and single premium deals, and from an increase in ceded premiums from Smart Home. We would expect a continued challenge to grow on premium in the near future in light of the recent low volumes of structure business written and the difficult market for higher premium bulk MI.

However, keep in mind that Radian has continued to build a very strong unearned premium reserve that is currently up to $246 million. Persistency has trended up and will give us a better opportunity to grow our flow insurance in force. The structured persistency was 56% and the flow persistency 69% for the twelve months ended 9/30/06. In examining the structured primary transactions written during the quarter, not only was the volume low but there continues to be a lower average premium rate on this business then our traditional rate because substantially all of it had a deductible in front of our risk position which should translate into lower losses on those transactions. In turn the capital associated with these transactions is lower, and the returns are within our hurdles.

Compared to the second quarter, the reserve for losses on the MI side went up by $59 million on a gross basis. A reconciliation of this increase is presented on Webcast slide number 15 which lists the five reasons for the reserve increase and our estimate of the amounts associated with each (inaudible). The reasons are a gross up of loss recoverable reserves on a second lien transaction, Radian's share of loss reserves for that same transaction, new first lien delinquencies for the quarter, a severity increase in our loss reserve model and an aging of first lien delinquencies including an increase in pending claims. As we noted last quarter, included in gross loss reserves are the reserves associated with a quota share reinsurance arrangement on the second lien transaction which began in 2005. This is essentially a gross up where the reserve is included in loss reserves but a corresponding asset is booked as a recoverable, and there is no P&L impact to Radian.

This gross up comprises $14 million of the quarter's reserve increase. Radian's 50% share of the same transaction comprises another 14. Up until the third quarter, this deal had been performing as expected. However, we received an increase in default and claim activity during the quarter. As a result of this activity our loss management group is thoroughly examining every delinquency, and will approve every claim before it is paid.

The deal is currently close to breakeven when taking into account premium, claims and reserves, and it is our expectation that it will ultimately continue to perform in that range. The balance of our second business is performing within our expected range. Another reason for the reserve increase is the new first lien delinquencies, which required an additional $11 million in reserves. Another $7 million was due to an increase in severity assumption within our loss reserve model, primarily due to larger loan balances. And the remaining $13 million represents an aging of first lien delinquencies which require a greater loss reserve per delinquency because they are closer to foreclosure and/or claim payment.

During the quarter there was an increase in claims received on both first and second lien. We were also making a deliberate effort to more closely examine claims, especially in the nonprime and second areas. These two factors have resulted in an increase in pending claims which have a full reserve amount booked and contributed to both the reduction in claims paid for the quarter and the higher reserve. Overall, the increase in more severely delinquent loans and pending claims will likely lead to an increase in claims paid over the next few quarters in the range of 12 to 15% above the third quarter level of $74.4 million.

As expected, our operating expenses rose a little bit this quarter as compared to the second quarter for a variety of reasons including severance and consultancies. We still expect that overall operating expenses could increase a little bit over the balance of the year as we look to add people in the international business and the structured products area in both production and risk management.

Financial guaranty had a good quarter of written premium considering the tough environment from municipal bond issuance and the tight spread environment. The trade credit runoff which impacts both premiums written and earned was offset this quarter by an unusually large amount of refunding in the public finance reinsurance products which accelerated some earned premiums. We also had a reserve reversal in the trade credit line of about $5.4 million, and that business runs off and we eliminate some IBNR reserves supporting exposures that are now off the books.

Excluding refunding and trade credit earned premiums are expected to be steady over the rest of the year as the books slowly grow and excluding the trade credit release of reserves with stable loss and expense activity we experienced this quarter is expected to remain in the current range in the fourth quarter, although expenses could be up a little bit. Our Conseco claims paid and trade credit claims paid represents substantially all of the financial guaranty paid claims in the third quarter.

Financial services again performed very well during the third quarter. For C-BASS which hedges its interest rate risk, the interest rate rally at the end of the quarter produced some mark to market hedge losses which are offset economically by gains on their loan portfolio. However, these gains won't be recognized until the loans are securitized which will be during the fourth quarter and into 2007.

For Sherman collections on their charge-off portfolios remain strong, and the origination business again did well. We expect solid results for C-BASS and Sherman in the fourth quarter of the year, steady in the case of Sherman and for C-BASS somewhere in between the current quarter and the first-half results. Please note that these businesses have built up a nice portion of their earnings from recurring sources and it is the goal of both companies to keep building upon that.

The mark to market activity for the quarter was very minor and like last quarter all of the movement related to spread changes, not credit issues. With respect to stock repurchase we expect to complete the current $4 million share repurchase authorization during the fourth quarter. If nothing changes with regard to the business opportunities and credit spread environment, we continue to expect to receive modest dividends from our operating subsidiaries in 2007 as we still believe we are in a modest excess capital position for both businesses.

Finally, our tax rate is a little bit higher than normal at this quarter due to a true up of 2005 taxes of about $1 million. We would now like to turn the call over for questions.

(OPERATOR INSTRUCTIONS) Michael Grasher, Piper Jaffray.

Good morning. Wanted to come back to the reserves and just a question first of all is around the severity increase. It seems prudent, how does this compare with previous quarters? Is this an adjustment to catch up? How should we think about this on a go forward basis, and I'll follow up.

Generally the severity has been increasing kind of slowly over time as our loan amounts have gotten larger. So that is consistent with the business and loan amounts getting larger. The other part of the severity would be the amount of claim paid as a percentage of the amount that we could have paid, and that again that is going to drive the severity model. So generally most of the increase is going to be related to the fact that loan balances have been increasing over time, and we would expect that would continue as, which is very consistent with just the mortgage business.

Okay, understood. And then on the aging of delinquencies, the plus $13 million here, is there some concern on your part, or should we be concerned as there is definitely a changing dynamic in the marketplace in terms of the rate in which delinquencies cure?

That is not really the reason for the increase in reserves. Much of that increase was driven by our proactive measures regarding loss management and our decision to hold more claims and look at them more closely. So that was something that we did proactively. We did see an increase in claims received during the quarter. That was a smaller part of the increase in the pending claims that the proactive approach that Radian is taking.

And then you mentioned an increase in claims paid 12 to 15%, I think. What was the reason behind your expectation for that, or what is the reason?

That is going to be driven clearly by the increase in the pending claims. So with more pending claims more will be paid. But look at I think you have to look at also the claims paid for the quarter were pretty low. So the 12 to 15% increase above that number, it is an increase, but I think if you look at that increase we believe it is fairly moderate and consistent with our expectations for the business.

I look at delinquencies I see little change and I see the paids are under control. I appreciate you giving the detail that you've given here in terms of breaking out these loss reserves.

Andrew Brill, Goldman Sachs.

Just a few questions here. Firstly, could you give a little bit more commentary on the reserve additions? Were they localized to a given book of loans geographically or was it widespread? Any additional color you can give there regarding the locality there would be helpful.

Andrew, other than the one transaction which we've isolated very specifically, the rest of the increases were really not concentrated in [Vintage] or any geographic area; really spread kind of throughout the book. And so there's no concentrated position in any area other than that one transaction.

I guess just talking about the rise in delinquencies, is there anything market related that is driving that or you think it might just be a third quarter aberration?

The increase in the amount of delinquencies?

Yes.

No, it is really a very normal in our minds, a very normal increase. There is some seasonality to it. If you think about the first two quarters where delinquencies went the other way and again, we look at a lot of nine-month activity as opposed to quarter by quarter. The number of delinquencies is actually down from year-end.

And just going back to the aging of the delinquencies, should we expect the proactivity you've been referring to to continue into the fourth quarter and potentially lead to further reserve increases?

I think the message around the reserves on the first lien side and to the second lien side really is with the aging in delinquencies, the pending claims. And as S.A. has mentioned publicly through the course of the year, we have strengthened our loss management area and that has resulted really in the increase in pending claims for the quarter. We do not expect that number to grow because that will work through the inventory through the next couple quarters.

And can you just talk to what impact you might be seeing from the slowing housing market? Has that had any impact on cure rates in the quarter? Or losses in the quarter?

We really haven't seen that impact.

And then just my last question with regards to the subprime book. How does the experience compare there and at what point do you Smart Home kick in?

We do not. We still -- the four deals that we've done todate we're not even close to the expected range. So we still, to remind everyone around Smart Home, Smart Home is used as a hedge around unexpected loss and is relatively inexpensive hedge and also is used to manage our capital, both on an economic capital basis and a rating agency capital basis. So in terms of the subprime book we don't see anything out of the ordinary that we've seen in the past few quarters and from last year.

Paul Miller, Friedman, Billings, Ramsey.

I hate to keep on coming back to this reserve issue, but going forward, I mean I guess, Bob, what really changed this quarter? Because you guys were much more upbeat in the second quarter about flat paid losses and relatively flat reserves. And in this quarter it really jumped, so can you go through what really was the change? And secondly how should we model this out? Should we be remodeling $30, $40 million increase reserves or have reserves matching relative to the paid losses?

Paul, you know we give guidance on paid claims because that's really where we have the most visibility. So that is what we are able to do, and that is what we do. With regard to changes in reserves we have much less visibility regarding that. But I think we've laid out the reasons for the increase; a large part of the increase was due to the one second transaction both the gross up and obviously Radian's share of that, as well. We believe that is isolated to that transaction. The rest of it is I think fairly normal increases in the number of delinquencies. The severity which is an ongoing thing and then this increase in older buckets and pending claims part of which is our doing. Now if you remember the way we reserve, it is very mechanical. It is very objective. We have a loss reserve model that basically plugs in each delinquency the stage of delinquency, and then tells us what the reserve needs to be on that. So it is very much objective and very, very, very little subjective.

So if you look at the aging of the delinquencies and we break down the bucket by each 30 days, the overall delinquencies in the existing book at 9/30 is older so it requires a greater reserve. The model tells us to put a higher reserve against that. And then the pending claims clearly there is a full reserve against that. And the reserve is an estimate. Ultimately what we pay out will really determine what the actual losses will be. And whenever the amount paid out is different from the amount you have reserved, there will be an adjustment made at that point.

I guess in my mind you know the aging in the portfolio and the loan size was coming. I mean that is something that you see everyday so I was just wondering why the adjustment is made in this quarter and why wasn't it made over time as the portfolio got older. I guess that's where I am confused.

Remember the way we reserve is based on existing delinquencies at a given point in time, and that is the reserve on the balance sheet. So if you believe that your defaults will age more over time, you can believe that but you can't book it until it occurs. So that is just the mechanics of the way the model works and the way loss reserving is required for our business.

And then the other question I had -- and this is really just housekeeping -- your persistency ratio is 65.7%. Did I hear S.A. talk about a 69.2% at the end of the quarter, or was it that I miss hear that?

That was for the quarter annualized. So the information was the most recent information, the most recent quarter. We always give -- when we say persistency is X, that is for the 12 months ended. The numbers S.A. gave you were the 69 were the point is that it is trending up, and the last quarter was significantly higher.

Okay, so that means you're still trending up. And then the last question is on the one deal that you reserved $14 million for. Was there any -- why is that particular deal -- is it performing weaker than everybody else? I know you guys don't insure against fraud but is there something else in there that is special that is having that act more poorly than the rest of your deals?

The transaction that Bob referred to was structured as a 50-50 quota share where our risk in force is approximately $75 million. This deal, as he said, was performing well within our expectations until our counterparties entered into an agreement to sell the underlying loan servicing platform during the quarter. Consistent with the time of the sale we saw significant increase in filed claims related to this transaction. As is our policy, we reserve 100% for all filed claims regardless of their default rates, but reemphasize what Bob said and to make it clear for everyone both on first and second liens, we are reviewing every one of these claims, and that is a big part of the pending increase that you saw in this area.

So they change -- can you go back -- they change servicers and once they change servicers the pool started acting very poorly; is that what happened?

They did not change servicers. They announced a sale of the servicing platform to a third party. (multiple speakers) At the same time of the sale we saw a significant increase in the file claims.

Okay, increasing file claims. And then you guys don't have any say where this goes. Can you go in there and pull that out of there?

We have consent rights around the sale.

Okay, thank you very much.

Hope that clears it up.

Yes, it does clear it up.

Geoffrey Dunn, KBW.

First on the seconds, I think this is the third time in five years where we've had a major disruption. The second time around we asked this same question, is it time to rethink the strategy of doing that, especially when you've seen other competitors get out of that business?

I would our returns on capital on the second lien business have not been consistent with our overall return hurdles for Radian. I think that is obvious. As a result, in the beginning of 2005 we did modify our strategy to do only subprime second liens either as a structured transaction and with a large deductible in front of us or as a quota share transaction like the one mentioned above.

The only second-lien transaction we did in first loss position was the quota share deal that I mentioned. We've also -- but we have done six transactions where we have significant deductibles, and our performance to date is well above our expectations.

And I want to go back to the reserves because I am still very confused. It doesn't seem to add up. On the severity factor, your severity sequentially improved. So I'm confused -- and it's not a surprise that severity is going to go up over the next year. So I'm confused about the message you're sending about how much '07 severity is going to ramp.

In terms of seasoning delinquencies, I don't think it is a surprise that frequency is going to pick up. So again, confused about the message you're sending about '07 about how quickly that's going to pick up. Is this more of a provisioning ahead of '07 expectations and the '07 credit outlook remains benign, or are you sending us the message that '07 is shaping up much worse than we would have expected prior to today?

Jeff, we are describing exactly why the reserve went up. And regarding severity, the severity assumptions in our loss loan reserve model were increased because the percentage of the ultimate claim that we are paying is going up. That is a fact, and that has to be built into the model because the model is continuously updated for the history. So that is very mechanical.

You look at our claim paid, and our claim paid did go down, but the paid claim as a percentage of the loan balance which is really what the severity is did go up. So if you apply that percentage to larger loan balances, and clearly larger loan balances are a part of our portfolio and are a part of the delinquent loans, that is what happened with severity.

Okay, I guess I'll have to follow up. It doesn't seem to add up because I don't think any of this is new information, and I think Paul hit it earlier. It seems like you are kind of accounting for this all at once rather than accruing for it as it occurs over the next 18 months. And so I just don't understand is this more padding ahead of time? Because the trends in the quarter don't support such a big change. And I think it is leaving kind of the credit trends in the quarter and your outlook very ambiguous, and I think that is what is translating into how your stock is opening this morning. So for my part I still don't understand how such a big development occurred so quickly.

The reserve that we have is our best estimate of what we are going to pay out on existing delinquencies, and it is almost entirely objective based on how the loss reserve model which tells us based on our history for the most part, what we expect to pay out on the current delinquencies. That is the way we have to reserve, and that is what we've reflected, and then we try to lay out exactly why the increase in the reserve occurred. So the current reserve is our best estimate of what we're going to pay out in the future based on current delinquencies.

Rob Ryan, Merrill Lynch.

Just for clarification because reinsurance recoverable is a relatively new thing on your balance sheet. To look at things on an apples-to-apples basis the mortgage insurance gross reserves should be reduced, I believe, by $5 million at June 30, and by $19 million as of September 30th. Is that correct?

That's correct.

And that for most prior periods essentially no reduction?

None whatsoever. First period was second quarter.

Okay, so an apples-to-apples percentage increase or something like that, that is the way we should look at it?

Yes.

The second thing is it seems like a long time ago at this point, but I would be interested in an update as to the default inventory as well as the size of the reserve related to the hurricanes.

The reserve on hurricanes and the number of delinquencies which we have really been -- we've been giving you the exact numbers each quarter, they went down slightly from the second quarter. So if we look at the numbers that we gave you last quarter it was total delinquencies of 4,104 with a reserve of $37.4 million and that included 1,302 severely delinquent with a $12.4 million reserve. That is now down to 3,810 delinquencies, $34.7 million so its down a little bit, and the severely delinquent ones are 1,028 with a reserve of $9.8 million. So that has come down fairly steadily and continues to come down over time.

Thank you.

Brian Horey, Aurelian.

A follow-up on the change in behavior on the quarter share deal with the second lien loans. The fact that you saw the increase in claims activity with the sale of the platform, the servicing platform, seems like an odd coincidence. What does that tell you about the general state of quality of the second lien marketplace, or what does that suggest to you is going on out there?

I don't read anything into the -- remember it was just an announced sale, so it was the same underlying as it was the same servicing platform. They just announced the sale and then started filing the claims. I wouldn't read anything into the performance of overall second lien activity in the market. That's why we're looking at every single claim as thoroughly as we can.

And then in the risk in force in exhibit K you had a big jump in credit default swaps year to year. Is there a change in strategy as it relates to those?

I think that year to year that is the growth in the international credit defaults, while we did a large transaction in the fourth quarter.

This is the German and Danish transactions that we [disposed] fourth quarter.

Okay, and then in schedule M you had said earlier that you didn't see the defaults other than the second lien transaction were pretty evenly spread, but there seems to be a big jump in the A Minus and below default number for the structured loans. Is that all reflect the second problem with second loans, or is there something else there going on?

That would be first liens only, so what that is going to reflect is the delinquencies on the A Minus loans and the stage of delinquency they are at. So I would say more of that change is going to be driven by the stage of delinquency. They are older, and they are certainly more pending claims on those as well.

Thank you.

Howard Shapiro, KBW Asset Management.

Just a few follow-ups on this whole credit issue. S.A., maybe you can just help us a little bit here. What is your assessment of poor credit quality trends going into 2007?

As we said, absent the one deal that we talked about, the second deal, everything else is normal. We have described all those factors to you in terms of the aging of the inventory, the only other thing that stands out other than the transaction that we highlighted is our conscious decision to hold onto claims. Absent that, we expect the credit to behave in a normal fashion, and based on delinquency trends which we cannot predict, it can go up or go down and our reserves will go up and go down commensurately based on what we see at that point in time.

Maybe just going back even a little ways in time to your time at GreenPoint, there is frequently an increase in delinquencies when servicing is transferred. Is that correct? You probably have seen that in the industry in the past when servicing is transferred from one party to another. Doesn't mean it is necessarily a decline in the deterioration in the underlying credit metrics. Would you agree?

What I can say to that point is I agree with you and that is why we're going to look at every particular claim. But we do not want to presume anything at this point. We're going to carefully evaluate every claim and in fact, we have been setting up a process to look at claims thoroughly for the last year or so anyways and this allows us to take advantage of the process and the resources that we have broadened to look at these particular claims in a very diligent, rigorous manner.

Okay, and one final question Bob, maybe you can help us, how much have loan balances increased over time?

We're trying to find that for you, Howard. Certainly they've increased over time. Mark has (inaudible).

Where the real increase has been we've seen average loan size on new insurance written in 2003 was approximately 164,000; in 2006 it was 184,000, and that is actually consistent both in the bulk and the flow market. And the other thing really to respond to Jeff's question from before, the average loan amount in default has grown from 132,000 in 2003 to 181,000 in 2006.

And if I can just be permitted one last follow-up, how do we reconcile your increases in your severity assumption with the data you have disclosed to us on average claim size, which is actually flat year-over-year and down from Q4?

You really need the other piece, Howard, which is -- we gave you the amount paid, but severity is the amount paid divided by the amount we could have paid. So or the loan balance. So that is the other piece that you don't have, and what we are saying is that amount, that percentage has gone up a little bit to be reflected in the model that required an increase in the reserves.

Thank you so much.

[Chris Ariano], [BBI].

Question on the one second lien quota share transaction. As I understand it, it is $28 million gross increase with $14 billion net. And someone made the comment that the deal is close to breakeven. Do you just explain how this transaction can be close to breakeven and what the premiums that you received associated with it are?

You can see we have a -- we Radian -- so I will forget about the piece that is the gross up, have a $19 million loss reserve against it. So that is obviously the big part on the expense side. So our premiums would be close to that. And then the other part would be claims paid. There has been a little bit of claims paid on it, not very much, so the claims paid plus the reserve, which is the total losses have been offset essentially by premium. That is where we got our breakeven.

So what are the premiums associated with it?

I think it is roughly 20 million, a little bit more than 20 million which equals the reserves which are 19 plus a little bit of claims.

So what is the size of the transaction than the aggregate primary in force?

The risk in force on this transaction for us is $75 million.

That is the maximum exposure we have on it.

Good. Thank you very much.

Mark Devries, Lehman Brothers.

Just want to clarify a few points. Bob, were you saying effectively that the increase in the subprime delinquencies is due to the aging of the book? And just normal season?

I think with everything, not just subprime.

But with the Alt A you actually had a decline year-over-year. Is that also seasoning where you've actually seasoned through the peak years on average? Or there are some positive trends going on there?

It is going to be related to all other factors; its going to be the aging, its going to be the loan size and its going to be the number of delinquencies, so all of those factors are going to figure into the reserves.

Okay, and then for the first line of the change of reserve I just want to make sure I understand that. Was your point that actually created an asset that you expect to hopefully from the quarter share to collect on later. So I think you said it was kind of a net no impact on earnings, but is it a negative impact today and then in a future period where you actually collect from your counterparty that's when it becomes a net neutral to earnings?

No, it's a net neutral to earnings from day one as long as we believe we can collect on the asset, which we do because we wouldn't be able to book it as an asset unless we did.

Okay. And then finally, I just want to understand the last -- make sure I understand the last line of the reserve breakdown. Are you effectively saying that we saw an increase in the aging of the delinquencies because you've decided to more aggressively try and negate your losses and hold onto losses so they are actually delinquencies that have carried over that might have otherwise been paid claims?

Partially, yes, absolutely.

And then I guess presumably your thought is that as a result of that you always fail to reduce in net what you might have otherwise ended up paying out had you just quickly gone ahead and wrote the check.

That's right.

All right, that's it. Thanks.

Ed Groshans, Fox-Pitt, Kelton.

Thank you, gentlemen, for taking my call. I guess I just wanted to talk about the opening comments it was commentary about the munies and the expectation of it picking up in the fourth quarter. Just was this a normal seasonal decline you saw in the third quarter, or was there some other factor embedded in there? And then Bob, in your opening statements you talked about the earned premiums and you talked about reduction in NIMs, seconds and single premium. And I was wondering are you just seeing market factors out there that you are pulling back a little bit right now or was it just a lull and you expect that to pick up going forward also? Thank you.

With respect to the municipal volume I think I wouldn't describe the results of the third quarter as being seasonal. Remember we had an extraordinarily good second quarter this year. I think we do expected to have traditions in the case, some uptick in the fourth quarter. I think the results when you look at it on a nine-month basis I'm given the challenges that I think have been in the municipal market as a whole, overall decrease in market issuance, refundings down, insured penetration down, and short par down. Are such that I think our results are reflected, reflect the fact that we held our own I think done well and we expect that to continue for the balance of the year. I wouldn't read too much into just the results for the third quarter itself.

I was going to handle the premium question. I think on the NIMs one of the things that Bob references some of the older book is starting to run off, hence the lower premiums over the next couple of quarters. However, if you note in our stat supplement we had our best quarter of NIMs risk written in the last two years. So we saw a couple of nice opportunities in the third quarter which we took advantage of, and as we start to shift our strategy more towards this barbell approach that S.A. had talked about, we are working probably even more closely with the issuers. And are seeing a lot more opportunity to transfer credit risk that is already on the balance sheet. So we're seeing some nice opportunities in that area and that's also when S.A. talks about structured products, we were down in the quarter for structured primary, but actually on a risk written basis we are slightly above where we were in the second quarter. And I think that is what sets us apart a little bit. We have a number of different ways in which we can put our structured risk on the books.

Jerry Bruni, J.V. Bruni and Company.

During the increase in reserves related to the quota share transaction, are you saying that increase is or is not related to the sale of the servicing platform to a third party?

We are saying it was coincident with the sale and for that reason we are examining every single claim before it would get paid. So we don't know at this point, however, we are looking at every claim to make sure that everything is appropriate before we pay it.

Okay, then would it be fair to say that obviously it is coincident, but it would be a pretty extraordinary coincidence if they were not related?

We cannot make any such comments. All we can do is focus on looking at each claim we receive in a very diligent and thorough manner. And do what is right from our point of view and from our customers' point of view.

[Terry Shu], JPMorgan.

I think the reserve question has been answered many times. Just one last go around. The second lien, that was a big item, and you talked about it, and it was worse than expected. But the other part, formula driven part was as you said objective, not subjective. It was because of the -- you are holding on to claims longer to study them and therefore it drove the reserve model, reserving model up. Overall you don't see any adverse trends versus your expectations. Is that sort of it?

We do not see anything unusual other than the fact that we pointed out in terms of the first deal and the claims activity.

Okay, and then just in terms of holding onto the claims longer, you are doing it because you see that the environment remains tough and you just want to be extra careful. Is that the rationale? Was that what you said? I'm not sure.

I think that is precisely what our intention is. That is very consistent with a lot of second liens and nonprime business where you read in the news a lot of heightened repurchase activity. So it is very much standard practice for that part of the business.

So really what seems to be the surprise on the part of investors is that as you said, we can't see the whole formula, that it really, the formula is not that we can't see it, I guess it is the loan, the claim and then the paid relationship which is the severity and therefore it drove the model. But there is nothing else ominous about it basically, right? It should not be a big surprise because it is model driven, and we couldn't -- it wasn't visible before or it wasn't visible enough to be modeled. Is that something like that? I just wanted to get a feel whether or not it should be a surprise and it doesn't sound like there was a surprise other than the second lien. Correct?

That's right; the model is based on history and you can read all the history into it and it literally tells us the reserve.

Right, and earlier the question from Geoff Dunn on the second lien is to why you want to stay in it even though there have been a couple of issues, can you go through that one more time, why is it attractive?

Sure, I think what we've done since the beginning of 2005 is we've modified our strategy so we have on second lien in first loss position related to subprime business, we would only do as a 50-50 quota share which we did one deal. The rest of the stuff that we are doing is all with subprime second lien, but it's all with significant deductibles in front of us as a structured transaction.

A key point that we need to point out here is that we have had a very dynamic strategy with respect to our business. Two things that are relevant to your question, are first in terms of the seconds responding to your question and Geoff's question, we have learned from our experience and we've incorporated into our strategy and as Mark said for the most part other than this deal everything else in our second business is performing well. The other point to be made is we consciously in terms of our strategy said to our investors that we were positioning the company to deal with the uncertainty in the housing market. And that had several compliments to it in terms of the deals we look at, how disciplined we are in terms of the deals we write and underwrite and also operationally we indicated that we separated the loss mitigation and claims management, complement of operations, brought in a senior person reporting directly to our President who manages that area and we have put resources and enhanced our processes in that area. Hence we are holding onto claims longer.

Okay, thank you.

There are no further questions in queue so we will turn the conference back over to S.A. Ibrahim.

Since there are no questions I would like to take a moment to reiterate to you that we continue to believe that we have the right strategy to achieve the long-term results that we believe,, and we believe our investors desire. I would also like to remind you of why I think Radian is unique. First, we are a diversified company with a portfolio of synergistic businesses. This diversification provides us with alternative sources of revenue and earnings. Second, our strong emphasis on risk management diversity and discipline positions us for the long-term. And finally we are focused on creating long-term value and believe the business will continue to generate returns of 12 to 15% ROE.

We are poised to benefit from the near-term cyclical upside in the MI penetration based on a unique capital markets and traditional MI capabilities we are positioned to benefit from the structural changes taking place in the mortgage industry. Financial guaranty business has demonstrated that it has a viable double-A franchise and has demonstrated success both in the public finance and the structured finance area. Our financial services complements performed well. We have a strong balance sheet with appropriate and strong reserves, and we have a strong and seasoned management team. And with that I would like to thank you all for having participated in today's call.

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