MGIC Investment Corp (MTG) 2006 Q2 法說會逐字稿

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

I would now like to introduce your host for today's program, Mr. Mike Zimmerman, Vice President of Investor Relations. Please go ahead, sir.

Thank you. Good morning, and thank you for joining us today on the call and for your interest in MGIC Investment Corporation. Joining me on the call today to discuss the second-quarter results for 2006 are Chairman and CEO, Curt Culver; Executive Vice President and CFO, Mike Lauer; and Executive Vice President of Risk Management, Larry Pierzchalski.

Before we get started this morning, I wanted to remind all participants that our earnings release of this morning, which may be accessed on MGIC's website, which is located at MGIC.com, includes additional information about the Company's quarterly results that we will refer to during the call and includes certain non-GAAP financial measures.

During the course of this call, we may make comments about our expectations of the future. Actual results could differ materially from those contained in these forward-looking statements. Additional information about those factors that could cause actual results to differ materially from those discussed on the call are contained in the quarterly earnings release. If the Company makes any forward-looking statements, we are not undertaking an obligation to update those statements in the future in light of subsequent developments.

Now, to begin this morning's discussion, I would like to turn the call over to Curt Culver.

Thanks, Mike, and good morning. Net income in the second quarter totaled 149.8 million compared with 174.4 million a year ago. Diluted earnings per share was $1.74 versus $1.87 a year ago. New insurance written was 16.1 billion, and that was comprised of 10.1 billion of flow insurance and 6 billion of bulk insurance. As I mentioned last quarter, we continued to see mortgage insurance penetration increase, as evidenced by our SingleFile volume, which comprised 9.9% of our flow new insurance written in the quarter versus 7.5% last quarter and 5.7% in the fourth quarter of 2005.

Persistency ended the quarter slightly higher at 64.1%, with our flow business at 68.2% and bulk business at 48.8%. Our combined quarterly run rate was -- in the quarter was 68.3%. With the increased writings in the quarter and the increase in the persistency rate, we were finally able to grow insurance in force to 169.8 billion, up from 166.9 billion last quarter. In fact, this is the first quarter since year-end 2002 that insurance in force has grown.

Paid claims totaled 162 million in the quarter compared to 135 million last quarter and 158 million a year ago. We had three unusual events impact paids in the quarter. The first and by far the most significant goes back to October of last year when the bankruptcy law was changed. The impact of that change accelerated bankruptcy filings in general. And to MGIC, it clearly accelerated the delinquencies that were bankruptcy related. In fact, we received 4,200 bankruptcy-related delinquencies in October 2005. While over the previous three years, we averaged on a monthly basis approximately 1,500 a month. Further evidence of the acceleration is that over the last eight months since the October filing date, we have averaged only 443 bankruptcy-related delinquencies monthly.

So as you can see, this law change told a large number of people, who would have filed delinquencies related to bankruptcy, forward to the October filing date. And as a result, we accelerated the claims forward also. In fact, we estimate about $12 million in claims was pulled forward from 2007; 8 million of which is in this quarter and approximately 4 million which will be paid in the third quarter.

The second event that impacted paids in this quarter was at the moratorium on filing for closures in the hurricane-impacted states by the GSEs was lifted in March in the counties that were least severely impacted by the hurricanes. As a result, many of the credit-related claims that we would have paid in the fourth quarter last year and the first quarter this year were paid in this quarter, which increased claims in the quarter by about $2 million.

Finally, in Ohio, a state where we have large exposure, we have seen a large backlog of foreclosures build up due to a shortage of magistrates. And to that, there were eight magistrates added in the first quarter. In fact, they went from 5 to 13 to deal with the large backlog. With that increase in magistrates, they started filing foreclosures and dealing with them, and it resulted in higher paids for us in this quarter. In fact, I expect we will see more of that catch-up in that category impacting us in the third and fourth quarters. Then, I think we will see higher claims from the hurricane impacted -- or the credit-related areas in the hurricane-impacted states also now that the moratorium has been lifted in the second half of the year. But the bottom line is that we still feel paid claims will be in the range we spoke of last quarter but closer to the high end at $600 million.

Underwriting expenses of 72.4 million were up 5% from last year, with the majority of the difference related to option and restricted stock expenses. Our joint venture results continued to be very strong. There were up 7% from last quarter and were particularly strong at C-BASS, where the quarterly contribution of $0.34 was up $0.10 from last year.

Finally, we repurchased 1.8 million shares in the quarter and year-to-date 3.2 million shares, with approximately 7.6 million shares remaining under our Board authorization. And with that, we will take questions.

(OPERATOR INSTRUCTIONS). Paul Miller, Friedman, Billings, Ramsey.

Referring to the insurance in force growth, I think for the first time in a long time you talked about you think you could grow in the second half of the year insurance in force growth due to the 80/10/10s or the popularity of the 80/10/10s starting to be marginalized due to the higher rates. Now we have seen it -- it's the second quarter we have seen growth. Are you getting more and more bullish on -- I mean, are you seeing more and more of this stuff coming through? Is the 80/10/10 loans really starting to really die off?

I don't think it's starting to die off, but we are continuing to inch forward relative to competing against us. It has come a little sooner with the rate increases where they have been. And also quite honestly, we had a good quarter for bulk writings also. So that combination of -- if you will -- pulled us forward a little bit faster than we would have thought in looking at this year and what we talked about at year-end. So yes, we are optimistic.

Then with the inventories, your default inventories, I think it is the lowest level in the last five years -- 73,000 units. Does that bode well for paid losses into '07?

Yes, the --

This is Larry talking.

Yes, the inventory certainly -- most of the claims in the short-term are going to come from the inventory. Curt mentioned some of the unusual events here, fast forwarding maybe some of the '07 claims. But the inventory being lower should lead to lower claims provided the claim rate stays where it is. With some of the news out of the auto sector, we will see how that plays out. But the inventory being lower, all things being equal, the claim rate in particular should lead to lower paid.

Edwin Groshans, Fox-Pitt Kelton.

I wanted to touch on credit a little bit also. I think you did a good job running through some of the items there. Could you just relate that to the jump in the primary average severities? Is that something that we should expect going forward?

I think generally, yes -- this is Mike -- because we do continue to see an added loan size increase. And so the exposure is going up -- average exposures. So I think generally, we should see a continued increase in severity just because of the loan size, all things being equal.

The other issue obviously would be claims rate, and that is what Larry just alluded to. But for the most part, all things being equal, subject to any significant changes in mix of delinquencies -- and by that, I mean we continue to be heavily weighted for the Midwest. Severity is up in the Midwest. And so we are subject to that change and have benefited really from the coast, which have not contributed much in delinquencies or claims. So that has been a benefit.

Mike, could you just touch on the claims rate? Has that been holding fairly stable or has it been improving or--?

In general, I would say quarter to quarter that it improved in some areas -- some geographic areas, where we saw some decreases in delinquencies. Then, albeit in the Midwest, the rates are still high and severity is higher. So we are still being penalized, if you will, for our mix of business in the Midwest. The Ohio/Michigan/Indiana states are still high in delinquencies and high in claims. So we continue to monitor that. But I think in the rest of the country, we did see some decrease in delinquencies and rates.

Then just I want to touch on expenses. It looked like they came in really well. Is there anything special going on there? Is it maybe less contract underwriting expenses or any other items?

Right. We have had some reductions. Obviously with respect to the reductions in volumes, we have reduced contract underwriting staffs; we have consolidated some offices. So we continue to work on that as the normal course, and we are getting some benefit there.

Just good management.

Eric Wasserstrom, UBS.

As we think about kind of the revenue growth, obviously, we are starting to see some of the insurance in force and that is a positive trend. But it seemed to be offset by a little bit of compression in the premium. How do you suggest that we think about the drivers of revenue growth from here?

This is Larry. Let me maybe speak to the bulk business. As you can tell, our bulk writings jump around quarter to quarter. And aside from that, the mix within the bulk business can jump around quarter to quarter. The couple of main sources of the bulk business is -- one, the capital markets -- the securitizations. And we provide mortgage insurance on those securitizations to hopefully improve the senior sub credit enhancement structure. That business, dependent upon the transaction and the marketplace, we can cover the over 80s down to 80, over 80s down to 60, over 60s down to 60, so on and so forth. The point is there is a lot of combinations. So for a given amount, a bulk volume for the capital markets dependent upon the coverage level, whether it is shallower down to 80 or deeper, that can influence the premium rate.

The other sources are through the GSEs. The GSEs buy portfolios from time to time, and they may want or need mortgage insurance on some of those portfolios they buy. Some of those may have deductibles and some may not. The presence of the deductible can significantly influence the premium rate. Then, you have portfolio lenders from time to time that would maybe come to the market for capital relief or risk management purposes.

So the point of that is that the volume can jump around and within that the mix and thus the premium rate. I would say as of late, some of the volume we are writing on the bulk side had deductibles, causing the premium rate on the bulk business to come down. Now, you see the impact maybe on the average premium rate, but you don't see the presence of the deductible at this point. You will see it later by the absence of claims. But right now, early on, you'll just see the impact of the lower premium rate pulling the --

On the flow side, I think by and large, the move is towards higher premium rates. We are seeing more and more high LTVs -- over 95 LTVs. We are seeing a good share of alt or reduced stock. The Freddie and Fannies of the world trying to meet their affordable housing goals are doing more in the A- arena. All that lead to, let's say, more -- or higher premium rates on the flow side.

Let me just give you an offset to that on the flow side. With the 10% of the business now SingleFile, SingleFile is in the neighborhood of 25 basis points. We get a lot of that upfront in a 1 point charge generally, but we earn it over four or five years versus our typical premium of about 60 basis points. So that is one of those that actually will lower. So if you are looking for something to really help you figure out those basis points, you can see it is a difficult equation to figure out, depending on the mix on what we do relative to the bulk side and the risks we take -- deductible/nondeductible.

On the flow side, as Larry said, we are -- we continue to have higher LTVs. But within that, we are getting many very high credit scores, which are using the SingleFile premium rates, which again are at a lower level of premium rate. So no easy answer there.

Just on a different question, can you just help us understand maybe what the reserve policy is, given some of these items that you have highlighted from the credit side in contrast to the declining delinquency inventory?

This is Mike. With respect to the decline in delinquencies in the second quarter, they declined about 3,000 units. We reported that about 1,500 of that came out of the hurricane-related inventory that we had talked about at year-end. Recall during that conference call, we said that there was quite a spike in the fourth quarter in delinquencies that related around, we thought, accelerated delinquencies due to the hurricane-related area and the moratorium. So we thought they would not go up at the same rate as normal delinquencies. And in fact, they have not. So we benefited from that, if you will. But there was not much reserve release with respect to that.

In addition, as I mentioned before, although we have seen a decrease in delinquencies in some areas other than the Midwest -- and the Midwest continues to be flat to up slightly depending on what the state is and of course severity is up as well as claim rates in some of the Midwestern states. So it is kind of an offset that we had benefited from some of the decline in delinquencies; some of it came out of the lower reserve hurricane-related areas. The others came in in mostly non-Midwest states. And we still continue to see increases, if you will, in claims severity and some claim rates, depending on the market primarily in the Midwest. So that really kind of covers the explanation quarter to quarter.

Going forward, I would anticipate that incurreds might be up slightly, depending on what happens to delinquencies in the second half of the year. And there again, I would say specifically the Midwest -- we will be more focused on the Midwest rather than the rest of the country. But traditionally, we see some increase in the second half of the year. Albeit if we do that, I would see some increase in incurreds possibly, depending on what -- again, what the mix is. So it will be dependent upon the delinquency levels and specifically what is happening to delinquencies in the Midwest.

I'm sorry. If I could just -- because I think I might be a little bit confused now. The primary driver of the reserve release in the current period was what phenomenon?

Well, the benefit we received from the delinquency reduction -- okay -- from reserves -- delinquencies went down 3,000. So there were 3,000 delinquencies, if you will, that did not go to claim, offset by some improvement in claims rates in the non-Midwest, offset by some increases in the Midwestern states.

I guess with those latter two phenomenon, does that suggest a lesser degree of reserve release from here?

Yes. Because again on a forecast basis, traditionally, we see an increase in delinquencies in the second and third quarter. So we are forecasting obviously --

Third and fourth.

-- third and fourth what would happen. But traditionally, we see delinquency levels increase in the third and fourth quarter. If that would happen, then you could possibly see an increase in incurreds. So it is going to be subject to the level of delinquencies and again primarily the mix of delinquencies.

David Hochstim, Bear Stearns.

Could you tell us what the average claim paid in Ohio is relative to say -- I don't know -- California or Florida or is it very different?

I don't know if we -- do we have that number?

I bet Larry knows it.

We can get it later (multiple speakers) --

We can get that, David.

Then I wonder, can Mike give us some sense of what went on in C-BASS and Sherman this quarter? Should we assume this run rate can continue, which would be a bit higher than we saw in the first quarter?

No, I think as we alluded to I think in the first quarter, we thought they would be a little bit flat year to year. C-BASS has had a very good quarter. They had quite a bit -- a number of transactions they did in this quarter versus the last year -- 2.6 billion versus 1. Their servicing fee is up and lower operating costs. So we benefited from a number of transactions they did this quarter. I think in the out quarters, third and fourth, will be probably somewhat closer to the first quarter. I don't see any significant change in C-BASS relative to where they are today versus the next two quarters.

You mean Sherman?

Sherman, rather. So flat for this quarter's performance.

Has Sherman been buying a lot of portfolios lately or -- because we understand prices are up a lot and the value has kind of deteriorated?

Not more than normal, no.

In expenses this quarter, is there much for international investment?

Nothing significant -- nothing more than the same rate last quarter.

Yes, it wasn't significant, David.

Then just -- I'm sorry if I missed it. But did you say what changed really in the bulk market between the first quarter and the second quarter where you wrote a lot more business?

Well, once again, it is dependent upon those three sources I mentioned. We basically in this quarter got, I would say, more from the GSE-related source than in prior quarters.

On GSE more than the other?

No, the capital markets is still the largest segment.

No, no. One GSE or -- yes, one more than the other, yes.

I'm sorry; one GSE more than the other? Yes.

What about the average claim? Did you find that, or I get that from Mike later?

No, we don't have that on the table here today.

Brad Ball, Citigroup.

Sorry, if I may ask another question on the losses in the guidance. Curt, when you went through the three abnormal issues in the quarter, you described the bankruptcy pull-forward -- that having $8 million impact, the moratorium on the hurricanes having a $2 million impact. Did you say what the impact was from Ohio in the paid claims?

I did not. It is kind of hard to measure. I would say it is in the neighborhood of a couple million dollars just from the preliminary numbers that we have seen on that. So I would say around $2 million in the quarter. I mean, we are going to see I think those same kind of numbers in both quarters also for both the moratorium catch-up in Ohio catch up in the third and fourth also I think in the neighborhood of about $2 million each.

So if we back those out, then paid losses would have come in more around 150 million as opposed to the 162 it sounds like?

Right, right. That's correct.

That 150 would annualize right to the 600 million number, which is the high end of your guidance. You said you would still be within the 575 to 600 range. So does that suggest that the expectation is that underlying credit quality will remain flattish or even improve in the second half?

Well, we said we think we will be at the 600 million. Relative to credit-related, again, it all depends on employment. So we think we are in the range, but you never know what happens. But we did have some claims pull forward from 2007 that are in this number that will help us more in 2007 than this year. But I think we will be at around 600 million.

Are you seeing any reason to believe outside of the macro employment data anything that would suggest that delinquencies would pick up in the second half? I know that you typically have a seasonal pick-up in the third and fourth quarter.

We are just estimating because of the seasonal pick-up that happens in our business year in and in year out in the second half of the year. So there is not -- when Mike talks about thinking it will pick up, it is not related to any condition in the country. It is just related to the fact that it generally always happens.

I know in a response to an earlier question, you said that the reserve releases would probably be -- would not occur in the second half. Would you expect to be adding to the reserve later this year?

That will be subject to delinquencies again. I think generally, we would look for incurreds to be at about the paid level, subject to any changes in delinquencies. If delinquencies are up, then that would be a mitigating factor.

Mike Grasher, Piper Jaffray.

Just a quick question on -- Mike, can you provide us an overview on the paid losses? I know you break it out on the release in terms of bulk versus flow but how about between sub prime, prime, and A-?

(multiple speakers) I don't know if we have that breakdown, so we are scrambling here. It is not quickly apparent, but we will look into that. But do you have another question?

I guess more of a comment just -- it sounds like you are more concerned about the unemployment -- the exposure here in the Midwest than you would be about the affordability issues around the coast -- Northeast or West Coast. Is that fair to say?

That's very fair to say. Again, the affordability -- the need for affordable housing is pretty dramatic, no matter where you go across the country and as you talk about overpriced markets. Again, within those markets, the supply for affordable housing is still at low levels.

So we are not worried about the price of the housing. We are worried about the people working that are living in those housing. You know, our portfolio is -- again, as we have talked about -- is far different than the general market, if you will, with so much of it being owner-occupied -- 93% owner occupied and 4.7% investor related, which is -- to me, that is one of the troublesome areas that will impact the coast that will be more the investor-owned properties, where people are flipping. We have limited condos -- 7.8% condos. The instruments on which we are doing these mortgages or insuring the mortgages -- 75% fixed-rate, 6% IOs, 3.8% option ARMs and less than 1% NegAm ARMs. So we don't have instrument risk either. So our risk is all about people working.

I will go off-line for that other information.

Robert Ryan, Merrill Lynch.

Could you characterize what you saw in terms of the bulk market -- what was available, what you passed on, what you were unsuccessful bidding on, that type of thing?

Sure. This is Larry speaking. With regard to one of the primary sources, the capital market securitizations, spreads did widen out in the last month or so of June. So May/June, spreads widened. BBs probably -- and I focus on the BB because where the bulk MI lien loan level is kind of towards the bottom of the credit structure. So BBs widened out by 100 basis points or so late in the quarter so that is a help to us.

Another change that probably is just in the works -- in July, S&P made some changes to levels -- levels is their risk model. What they did is started factoring in the CLTV a bit more than in the past. So in a lot of these securitizations now, there is a lot of 80/20s. Instead of looking at that 80 as a pure 80, they are now looking at it as more of a -- somewhere in between, let's say an 80 and 100 LTV, which that should do is increase the credit enhancement requirement for those pools.

Now, MGIC has looked at that 80/20 as closer to 100 for a long time. So in a way, S&P is just kind of coming towards the position we have had. So that should make us more competitive, viewing the risks more similarly than in the past.

But in the capital markets, the losses have been low, not many losses to the lower-rated tranches. For most of the quarter, I would say there has been quite a bit of activity in the buying of the BBs and BB-s. A lot of players bidding the price up, which works against our execution.

So a couple of things here as of late -- turn in our favorite -- the S&P change, the spreads widening. On the GSE side and the portfolio lenders' side, it really is a quarter-to-quarter situation, never knowing what appetite they have for the portfolios, how they are going to be competitive against the capital markets and of the portfolios they buy what parts or pieces they want credit enhanced. So that one is hard to peg. That is kind of a quarter to quarter and really tough to get insight into that. But on the capital markets, I would say a little bit improvement favoring the MI execution as of late in terms of the spreads widening and S&P changing their levels model.

Switching over to Mike on the incurred sort of statement for the second half of the year. When you say might be up a bit, do you mean on a year-over-year basis or compared to the second-quarter level (multiple speakers) --

Compared to the second quarter.

Wouldn't that be normal almost in the typical circumstance, almost regardless of what is going on in the economy, the Midwest, or whatever? Usually, your incurreds because of what goes on in the delinquency inventory (multiple speakers) higher in the third and fourth quarters than the second?

That is correct, yes. I guess, again, what I was saying is that traditionally, we would see an increase in delinquencies; that would generate an increase in incurreds at the level of which I think going forward, it will be up slightly from where we are today in the second quarter and again subject to how big the degree would be.

The only caveat would be the Midwest. You know, if there was some further deterioration there and that would be driving delinquencies a little bit higher and severities possibly.

So we should look at the number above 146.5 in the usual circumstance, subject to what might go on in the Midwest and just what normally happens in the delinquency inventory in the second half?

Correct.

Matthew Roswell, Stifel Nicolaus.

Most of my questions have been answered, though with insurance in force beginning to turn positive, normally how long would then that cause the net premium earned to turn? Is it a six month, nine month?

The majority of the earned premium comes from the renewals. So as the book grows, it is going to take probably a couple -- three or four quarters for that growth on there.

Yes, most of the premium -- most of the insurance we write is monthly. So basically, the premium is coming in the door each month. So if you look at it that way, it should start to go up as soon as the in force starts going up, provided the premium rate on the new business is equal to or higher than what is in the force.

Do you have the average age of the book handy? In other words, sort of like what percentage is from '04, what percentage '05?

No, we will get (multiple speakers) --

Mike is looking for it right now.

Did Mike say he would get it to me off-line?

Yes.

Okay, that's fine. Those are all my questions.

Bruce Harting, Lehman Brothers.

What is the persistency rate for the quarter in the latest month? Do you have that?

For the quarter -- this is (indiscernible) quarter -- I would say for the full book, the quarter was 68.3% -- was the quarterly run rate. If you break that down, the flow was 71.4 and the bulk was 58.2. I would not put a whole lot of steps because it's on a monthly, so I didn't calculate those because that could swing around quite a bit.

Then in terms of dispersion of risk, I am just looking at your 10-K and Florida is 9%; California is 7.9 of primary. So you are pretty substantially underweight California?

Significantly.

Yes. (multiple speakers) And so compared to the sort of the '90s when California was the problem in the portfolio for most MI companies and larger average loan size. Right now, it sounds like you are saying Midwest. You know, it seems like a lot of investors are most worried about the coastal areas because we are seeing places like Sacramento and San Diego in their latest monthly or quarterly data for the first time show negative price appreciation after years of glorious, strong home price appreciation. So there is a lot of concern about those markets.

But it sounds like you would say as long as -- employment continues to be strong there, and you are way underweight California. So if there is anything disproportionate going on in your portfolio with sort of 5% Michigan, 5% Illinois and Ohio, it is the area that had the least home price appreciation in the last few years there that are sort of the areas struggling the most just because of employment price. Is that right?

Exactly.

Is that right? So, it's a --

That's right on, Bruce.

So it is a very different profile of credit today than it was, say, in the '90s.

Well, yes. For us, it has kind of always been that way. It is just that the Midwest right now has participated, if you will, because of the auto sector and moreover recession than the rest of the economy. We have always been light California, if you will -- now more so than ever. And that has not been as much by choice as just the realities that the 80/10/10s and the 80/20s in particular were an accepted part of business there much longer than they have been. So I would expect you'll see some of that business coming back to us. But again, we have always been light coast-wise.

Right. And as a percentage of flow risk in force by credit grade and bulk risk by credit grade, would that California weight be any different from flow to bulk? Or would it be pretty representative?

No, on the bulk side, we have done --

We are probably closer to 20% on the bulk California.

Paul Miller, Friedman, Billings, Ramsey.

Curt, I want to approach the issue about the release of the reserves. I mean, your stock -- for some reason, the Street doesn't like everything that's going on out there with this Company, especially the release of the reserves and your credit view. But if you are releasing reserves, in layman turns, that means you are comfortable with credit going forward. Am I correct? Can you just approach that in a layman terms terminology rather than just talking about different regions?

Well, I guess the issue with respect to the release has to do with the number of delinquencies that we have and the amount of which we believe that we reserve for those going forward, subject to reviews obviously and the fact that because -- notwithstanding the fact that we had some higher-paid claims this quarter, as Curt explained. They were fully reserved for. And the remaining reserves, if you will, at the end of this quarter, we reviewed and estimated that the reserves could be taken down.

So in other words, the 73,000 delinquencies that we have at the end of this quarter were down from the previous quarter. And going forward, the rates at which we reserve for we're comfortable with. Therefore, the incurreds were less than the paids. They were significantly less in the first quarter, again, because of a significant decrease in delinquencies.

Going forward, we would anticipate that the incurreds might be slightly higher but net-net not too much off of what paids will be for the year. So I am not uncomfortable at all with respect to where the reserve is, subject to the level of delinquencies we have. And I think if you recall, during '05, we were redundant, if you will. That was over-reserve with respect to the reserves from the previous year, and we disclosed that in the footnote.

So I think it is an adjustment every quarter, subject to the experience we are recording and the results of which we look at each claim at the end of the quarter and the experience rate and going forward, look at rates that we need to be reserved for. Fortunately, going forward, we believe we have got adequate reserves, subject to these delinquencies. So therefore, there was a release of 15 million. But the release really is subject to the level of delinquencies and the fact that they went down 3,000. Though, I think it is a positive development as opposed to a negative development.

The other issue is -- I know you disclose this, and I was just wondering if you can just -- is the -- your book of businesses by vintage -- I would think most of your vintage books is before 2004 and earlier, and I believe something like 20, maybe 25% of your book is 2005 and later. Is that correct?

Well, from a risk in force perspective on the primary risk in force -- this is bulk and flow combined -- for 2005 is 29% of risk in force; 2004, 19.5; and 15% in 2006.

So I mean you are relative -- about 60% of your risk in force is before 2005?

(multiple speakers) Yes, it sounds like 55%.

55%. Right.

My personal view is there is probably very little risk in those type of loans because of the level of home price appreciation that we have experienced in that time period.

Edwin Groshans, Fox-Pitt Kelton.

Thank you for taking a second question here. Just I thought you said earlier, you're still staying with incurreds are going to equal paids. Is that correct, give or take a few dollars one way or the other?

Exactly. Subject to what happens with delinquencies.

Then, throughout this call, there has been a lot of commentary about delinquencies and reserve methodology. Just something for consideration -- I don't know if you ever think about putting up -- maybe giving us some identification of what late-stage delinquencies are, so we can track that a little bit and monitor that relative to reserves. Or maybe give us some idea of pending foreclosures that you see, so we can kind of track some of the later-stage credit issues rather than we see a big drop in delinquencies but a lot of them are short-term, so there is not a big impact on the reserves at the present time. I throw that out there.

Geoff Dunn, Keefe, Bruyette & Woods.

Basically, my questions have been answered. Just judging by where the stock is, just curious on what your buyback capacity is these days?

Well, we have got I think (multiple speakers) --

7.6.

7.6 million, as Curt said, still in the authorization. So we are well-positioned.

How about cash capacity that is available at the holding company?

Well, there is very little there. We continue to still have to special dividend money up each quarter.

Your debt is still limited by that double leverage ratio?

Correct.

Mike Grasher, Piper Jaffray.

Just a quick follow-up. Could you confirm -- reserves per delinquency, are they not at an all-time high at this point?

That's correct.

That correct.

I thought so. Thank you.

Robert Ryan, Merrill Lynch.

Could you just review for us what you have done in terms of special dividends year-to-date as well as the regular $55 million per quarter (multiple speakers) and your status in terms of asking for additional -- at least the regular 55 million?

We had a special of 150 in the first quarter and a special of 100 in the second quarter. And each of those quarters, we had a quarterly 55 approved. Anything in addition to that, we have to ask for going forward.

Any application for something in the third quarter yet?

We haven't met with them yet. We generally wait until we close out the quarter.

David Hochstim, Bear Stearns.

Could you just remind us what the average cure rate has been lately?

The average cure rate?

So if we think about the 71,700 delinquencies at the end of June if we wanted to estimate what percentage of those might go to claim ultimately, can we assume it's --

I don't know if we have disclosed that. But obviously, a high majority of those do. But I don't think that we have disclosed the rate, and I was kind of (multiple speakers) the pricing but a large majority.

Go to claim or cure?

No, cure.

Cure. So some portion of those go to claim.

A small portion of, yes, that is correct.

So basically, one could think you have reserves. So if somebody took the average delinquencies divided by reserves, it looks like 15,000 or $16,000 of reserve for default. But the reality is it is the reserve for ultimate claims, which will be some multiple of that.

That's correct.

So you basically have then -- not that you would estimate, but we could estimate 40,000, $50,000 per possible claim. So that is assuming going to 60% cure rate, which would seem to be pretty conservative if average claim is running, what -- about--?

Well, we disclosed the average claim paid, and it is around -- Mike?

27 point (multiple speakers)

$27,000 a case.

Right, that is the average. There might be some states where it is higher, and some states where it is lower since that's the average.

That is correct.

But I guess I am just having a hard time understanding why people would be panicked that you are inadequately reserved is all.

I think to get back to the question, if you just think mathematically about the reserves and the delinquencies, we have got an average right now per case of about 14,700 compared to about 13,000 at year end and 13,000 a year ago. So as someone pointed out, we are at the highest we have ever been. So I can't say much more than that.

But there hasn't been some big change in the cure rate, or--?

No.

No.

Jim Delisle, Chapel Street Partners.

Looking at your long-term insurance in force and revenue trends, there are people out there who are thinking that the weakness in the second -- the greater than 80% LTV market, the weakness in securitization and the lack of bank bids and maybe S&P's 5.7 was going to give you a little bit more of an opportunity to do more bulk business and possibly increase your premiums. Now in response to a question earlier on the call, I think I heard you saying that the varying levels of deductibles in various pools for various people, it is really hard to track that pricing power very closely.

(multiple speakers) Yes, it is dependent on the risk that we are insuring.

All right. Is there any kind of -- seeing that your IIF is actually starting to move in the direction it seems probably on the margin, most of it coming from business at heretofore would have been handled through other means of securitization of seconds, is there any kind of a metric that you folks can put out there for us to follow as to your pricing power in that? Or could you comment on how much competition you are seeing from your competitors and for that business? Because people I think would like to see some pricing power there.

With regard to the bulk business, I would say that our biggest competitor comes from the non-MI side. I would guess with regard to the capital markets' securitization, the MI industry might be only penetrating the securitizations by around 10% or so. And I say that because I think if you add up all the MI bulk volume, it might be 40, $50 billion or so the last couple years. And I think securitizations have been in the 500 billion area (multiple speakers). So simple math, 50 into 500, it is about 10.

So you can see from that that the MIs largely -- there is a lot of room for growth in the bulk business. It is really a matter of the non-MI bid spreads. There is a lot of money looking for a home in a yield. And so until spreads widen some more and people get maybe a little more risk conscious, it will continue to be a quarter-to-quarter event, I guess.

I would agree. As a follow-up, I would agree. Yes, obviously, the securitization bid widening out as much as it has has presented an opportunity. Is there -- in terms of the premium -- that opportunity, obviously, they move their bids back because they feel that they should be compensated more for taking the risk. Presumably, you as a writer of insurance would have the same concerns on the margin and would evidence it through a higher premium rate. Is there anything we could follow that would indicate -- like if we see BBB- ADX moving from 1.75 to 2.60, we can track that in real-time in understanding it's a quarter-to-quarter event. Is there any kind of series of numbers from your reports we could follow that indicate any increased stability on your part to price the marginal risk attractively to your shareholders?

No.

No.

No. As you see the spreads widening, you will know we will have more opportunities. And that is what you can track. Within an indicator of premium rates, that is very difficult in a competitive world.

Anything in terms of tone as to percentage of business that you are closing on and seeing versus past quarters?

Not really. Again, as spreads widen, we get more opportunities. So nothing out of line.

James Shanahan, Wachovia Securities.

I am sorry to kind of beat this into the ground. But my question about delinquency, I would like to know if you can share with us what the -- how the inventory tracks during the quarter on its way from 76.3 to 73.3? Was it consistently declining throughout the quarter, or was there a trend there that is meaningful?

Yes, this is Larry. Here is the month by month. April was 72.9 thousand; 73.8 in May, and 73.3 in June.

One other quick question -- recall -- I am looking at the delinquencies that were attributed to the hurricanes, and those have been consistently declining for the last couple of quarters since you originally reported that metric. But the question I have is that I -- before you didn't -- I think you talked about not reserving for hurricane-related delinquencies. It turns out now you are experiencing some losses associated with some of those defaults. But my question is --

No, no not that. You are misinterpreting there.

Well thank you; I probably am. Please clarify.

We didn't reserve for where we felt there was hurricane damage. Now within those counties that the moratorium was released, what we are paying claims on are normal, credit-related events that didn't have damage to their property.

Understood.

But counties all had a moratorium where there were no foreclosures regardless of that hurricane activity.

I understand.

Those were released.

Howard Shapiro, Keefe, Bruyette & Woods.

I think I understand the issues at play here, but I am still a little confused. Are you more or less or the same level of concern about the Midwest that you had yesterday or a month ago?

Same.

It's the same, right?

Yes, (multiple speakers) nothing has changed.

Nothing has really changed (multiple speakers). And if we look at loss severity trends in the Midwest going forward, given the slow rate of home price appreciation we have had over the past few years, there is no reason to think that loss severity in the Midwest is going to increase.

I don't think much on severity. I think the only thing would be the level of delinquencies, if they would go up and the claims rate would go up. But severity, not much changed.

But your concern about delinquencies hasn't really changed?

(multiple speakers) No, not at all.

So it is kind of status quo in the Midwest right now?

That's correct.

(OPERATOR INSTRUCTIONS). David Chamberlain, [Dimco].

Just quickly on market -- I know you guys talked about I think in the first quarter about spending a little bit more money to increase the profile of the SingleFile and just MI in general. What is the status of that going in the second half of the year?

Well, we purchased a company, Myers Internet, that has helped us a great deal relative to the broker community, get the message out relative to the MI execution. We have hired a few extra people. We are doing much more advertising than we have done before. So we are spending the money accordingly. Frankly, what I haven't talked about as a trade group, Mike, we are also embarking on an initiative shortly to promote private mortgage insurance, an expense that will be shared amongst all the companies.

So there is going to be a great deal in play relative to the benefits of mortgage insurance, both on a company basis as well as an industry basis. And those are all playing into increased mortgage insurance penetration. So it is a very positive sign.

We are still optimistic relative to the mortgage insurance tax deductibility, which would be a great springing point for us to do a lot of this campaigning. They are trying to tie us to the pension bill, and that has been the problem that the pension bill has a lot of disagreements. But there seems to be a lot of agreement relative to the merits of the mortgage insurance tax deductibility. So I am still optimistic on that.

Just related to that, I mean you've obviously spent some money upfront. At what point do you think in terms of as the book grows, you would start to see some operating leverage in the underwriting expense line?

Well, again, we run the Company pretty lean, as you have seen by the numbers. So I -- it is not as though what we are doing -- there is a lot of reductions we can take out, if you will, maybe 1 million to $2 million.

Thank you. I would like to turn the program back to our hosts.

Again, this is Curt, and I would like to again thank you all for your interest in our Company. Have a great day. Thanks.

Thank you, ladies and gentlemen, for your participation in today's conference. This does conclude the program. You may now disconnect. Good day.