MGIC Investment Corp (MTG) 2003 Q3 法說會逐字稿

Good day ladies and gentlemen, and welcome to the third 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. If anyone should require assistance during the conference, please press star then 0 on your touch-tone phone. I'll now introduce Mike Zimmerman. Sir, you may begin.

Thank you and good morning, and thank you for joining us this morning and for your interest in MGIC Investment Corporation. Joining me today to discuss the third quarter results are Curt Culver, President and CEO, Mike Lauer, Executive Vice President and CFO, Larry Porcelski (sp?), Executive Vice President of Risk Management, and John Fisk, Executive Vice President of Strategic Planning.

Our earnings release of this morning includes additional information about the company's quarterly results that we will refer to during the call and include certain non-GAAP financial measures. The release and this additional information may be accessed on MGIC's home page, which is located at www.mgic.com. During the course of this call we -- excuse me. During the course of this call we will make comments about our expectation about 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 are contained in the quarterly earnings release. If the company makes forward-looking statements, we are not undertaking an obligation to update those statements in the future in light of subsequent developments. At this time I'd like to introduce Curt Culver.

Thanks, Mike. Our third quarter was one of our busiest in almost every aspect of our business. New insurance written was a record $28 billion with a record $21 billion of flow and $7 billion of bulk. Year to date, we've booked a record $78 billion. While the rise in interest rates during the third quarter will lead to persistency growth in the longer term, in the short term it led to strong refinance, loan closings and a decline in our persistency rate to a historic low of 45%. By way of comparison it was 50% a quarter ago and 59% a year ago. The good news for us is that I think persistency should start rebounding in the fourth quarter and trend upward throughout 2004.

However, as a result of the low persistency rate insurance in force declined to $191 billion from $194 billion quarter ago and $197 billion a year ago. However, even with the decline in insurance in force premiums earned totaled $347 million, and that was up 16% from a year ago, reflecting higher average premiums from the maturation of our bulk and A minus business. Regarding credit losses, the inventory of delinquent loans continued to increase, reflecting the maturation of our larger new books of business, but also reflecting the weakness of the labor markets, particularly the manufacturing sector.

As many of you are aware, while MGIC's market share strength is well diverse -- well dispersed, being number one in 38 states and number 2 in nine states, in particular our strength lies in the Midwest. And while long term that is where I want our strength to continue to be, in the short term it is participating more strongly than other regions in the loss of employment and as a result, higher delinquency inventory growth in our flow business than we would have expected. This increased delinquency activity, coupled with higher loan amounts, as well as fewer loss mitigation opportunities, caused us to increase reserves by $96 million. Underwriting expenses in the third quarter totaled $78 million, down from $80 million a quarter ago. The primary driver of this reduction was contract underwriting volume, which fell 23% in the quarter.

Regarding next quarter, we would expect to see lower volumes of flow new insurance written but a continued strong appetite for our bulk business. Persistency should start turning upward and expenses should continue to decline. Our delinquency inventory should continue to increase, and paid losses should continue to be at third quarter levels or somewhat higher, reflecting the higher delinquency levels. Next year, particularly in the first half of the year, we'll continue to be challenged by the credit loss trends I discussed earlier, with continued growth in delinquencies as well as paid losses.

On the positive side, we expect to see a year dominated by purchase transactions, which is a positive for insurance in force growth as well as expenses. Longer term, I'm extremely optimistic relative to our business for a number of reasons. Two other major insurers have made moves similar to MGIC's captive reinsurance position, and obviously this will help us with our market share. The discussions in Washington relative to the GSE's also should be beneficial to our industry in minimizing the GSE intrusion on our business. Finally, the support of the U.S. banking regulators regarding the capital credit through the use of private mortgage insurance under Basil 2, are coupled with the strong bipartisan support in Congress for MI tax deductibility, which hopefully will lead to its passage early next year, will go a long way in reducing the success of 80.10.10, 80.15.5's and 80.20 structured transactions.

As I've discussed earlier, or at many other conferences, these structured transactions have cost our industry approximately 25% of our market share. So in summary, we have some short term challenges as we deal with a weak economy. That's the nature of our business, and has been so for over the last 20 years that I've been part of it. But longer term and more importantly, major franchise issues are being dealt with that will be very beneficial to our long-term financial success. With that, let's take questions.

Ladies and gentlemen, if you have a question at this time, please press the 1 key on your touch-tone telephone. If your question has been answered or you wish to remove yourself from the queue, please press the pound key. Again, if you have a question or comment please press the 1 key. One moment for our first questions, which comes from A.J. Rowell. You may begin.

Yeah. Could you give us some color as to how much of your insurance in force is in the Midwest region relative to the rest of the country, and also your tax rate was down quite a bit, your effective tax rate. Could you give us some color on that and what to expect going forward? Thank you.

Let me first talk about the tax rate. If you remember, on the last call, I commented on the same thing, and that is that underwriting income as a percentage of total income is trending down. That is, even though premiums are up, incurred losses are up significantly. So the mix between underwriting income as a percent of total income is declining, and yet our muni income is remaining flat to up, and that's the significant driver of that rate change, okay? So, and I would say that going into the fourth quarter, we'll probably see tax rates at about the same level or maybe just a tick down for the same trends.

As far as the Midwest percent of our flow business, Michigan, Illinois, Ohio, the top states there, a little over 5% each, so those three states in total, about 16% of our flow in force. Wisconsin and others, less than 2% or so. So I would say roughly about 20% for the five, six states of the Midwest.

And how would that compare to the industry?

It would be higher. Our market share within those states, you know, Michigan we're in the 40's, market share, and each one of these states we're higher than our national market share of 22%. We probably average around 30% across the mid western states.

And just to follow up that, what kind of default and delinquency rates are you seeing in these states relative to your national average, the other states?

Well, those states have historically been very good, very low, and they have moved the most to this point here now. The point being is that the delinquency rates were very, very low, and as Larry said, the factors of increase have been significantly higher within those states. I think Ohio and Michigan alone are about 15% of our delinquencies for the total of the 85,000 that we have. So, while they're not out of line with, let's say, Florida and Texas, relative to an absolute rate, the run-up within those has been significant year over year.

Nationally we're about 3, 6, 7, as we disclosed on the flow side. In the Midwest we're probably mid-4.4, 4.5% area.

Our next question or comment comes from Paul Miller. You may begin.

Thank you. The $220 million in incurred losses that you reported this quarter, is that a good run rate to go forward for the next three or four quarters, or is this somewhat of a trough level and we should see some improvement going forward?

Well, I think, relative to the next two to three-quarters, I would say yes, it is, because we are anticipating an increase in paids the next three-quarters, as well as an increase in delinquency inventory. So, all things being equal, I think this number at this rate or slightly higher would be a good trend. As well, I would say that relative to the reserve buildup, again in the fourth quarter I would trend to see something similar to the third quarter, or maybe higher again, depending on notice activity. As we get into the next year, it's a little softer on the forecast. I think certainly in the first two quarters we still see a continued increase, if you will, in paids because of the carry-over, and also anticipate higher notice inventory.

If, on the other hand, the economy starts to pick up and we start to get some pure rates improving, that could maybe mitigate some of that in the second half of the year. So, I mean, relative to your question about the next three-quarters, I would say yes at this number going slightly higher. and then mitigating, if you will, in the second half of the year, if the economy starts to pick up and we see some declines in pure notice inventory. I think we'll see several things, if it starts to pick up. We will see a flattening, if you will, or a decline, in notice inventory. We also might see at the same time an improvement in pure rates. Both of those impact positively to incurreds, and then ultimately we will start to see some decline in paids.

Some of the data came out, like the unemployment report last week came out, and showed that we grew 57,000 jobs. I guess there was some positivie data today. Are you seeing that at all through some of these notices? Is any of that good news flowing through to you yet?

No.

No. Only in some states that are lesser, and as Curt mentioned, relative to the Midwest we have not seen it, and again we had an increase, if you will, in notice activity in the Midwest, and no improvement with respect to claim rates. It would be too early, I think, for the data that you're seeing

And it would be somewhere between a 3-to-5-month lag before you would expect to see improvement with some better macro data that we're seeing?

I don't know how to put a time frame on that, Paul.

Okay. And a real quick question. Just going back to the tax rate, we can expect about a 24% tax rate going forward? Is that what you were saying before?

Yeah, it might even be down to 23. Again, we're talking about the relative position of operating income and what percent the muni income is, you follow? And it's been declining. Underwriting income has been declining.

Okay. Thank you very much, gentlemen.

Our next question or comment comes from Makiko Cokely. You may begin.

Your guidance last quarter for claims paid this quarter is up 15 to 25, and you actually report up to 27. Did something happen in this quarter that was much worse than you were expecting?

Makiko, could you repeat the question?

Yes. In the second quarter conference call, you said that third quarter claims paid would be between $15 to $25 million.

Right. Increase.

Increase, right. And you actually showed $27 million increase. So I'm just wondering if something happened in this quarter that was much worse?

No, it's up. I guess you could say we were $96 million in the second quarter and $124 million in the third. Okay? So we're up $28 million versus $15-25 million, so that's pretty close.

This is a forecast, Makiko. We're giving ranges here based on what we anticipate. We've got a relatively reasonable fix, but again it has to do with the economy, claims mitigation efforts, et cetera. And I would say, relative to the fourth quarter, we would look at this number as a base case in the third quarter and look to an increase of maybe 10 to 15, anywhere from 5 to 15 increase.

For fourth quarter.

Yes.

I see. So, in terms of a credit, there is nothing that is happening in your portfolio that is worse than expected?

Just the economy. If you were to ask us to go back a year ago, would we have anticipated this level, I'd say no.

I'm talking about from second quarter to now. I think the economy is kind of strong in the third quarter, and then yet your data is not showing any improvement.

I think we just kind of discussed that. The economy that you're seeing some immediate impacts right now with respect to what you're reading in data, but it's lagging. The people that are defaulting on these loans have defaulted a year ago and are working out, if you will, the loans and they're going to claim, and they're not going to be impacted by what's happening in job growth today.

Okay. And in terms of your premium, average on premium rates, it's up again another 3 basis points again or so this quarter. Would you say that this is a result of you doing less captive insurance, or is this the mix of product?

It was a mix change, not -- I wouldn't say the captive influence has really run through yet.

I see. And then where is this number going?

Well, I think it's being driven a lot by the level in bulk we did in the quarter. If that continues at that level, and the mix remains about the same, we could be at about the same level in the fourth quarter. It's kind of a quarter-to-quarter forecast with respect to that, because of the level of business.

Bulk pricing has been very strong, and we've taken that opportunity to do a number of transactions.

So you can continue to trend higher?

Well, I would say, looking at the fourth quarter, I think so.

Okay. Great. Thank you.

Thanks.

Our next question or comment comes from Brad Baugh. You may begin. Thanks.

Mike, I wonder if you could comment on the fact that did you not buy back any shares in the quarter. Could you comment sort of broadly on the capital management and sort of whether there are any implications for the up-streaming of dividends?

Yeah, what we did last year, you recall, we were somewhat stronger because we did a combination of things to repurchase stock. We dividend up about $150 million from the writing company and also took on some short-term debt. This year we're working down the short-term debt position and trying to repurchase basically through up-streaming dividends. And we're in the process of still doing that, so relative to the rest of the year, will be our ability to up-stream some dividends in the rest of the year as well as looking at other things. But for the most part, our program going forward would be to utilize existing cash upstreaming dividends with respect to the stock repurchase and minimizing the amount of debt for that strategy.

So you would expect to be back buying shares this quarter?

Yes.

Yes. And you said utilizing excess cash.

Yeah, to the extent that we dividend up cash and have it available at the holding company, relative to everything else we're doing, if the opportunity present itself and we deem it appropriate to buy back some shares, we'd do that.

Okay. And separately, you've touched briefly on your cure rates. Could you just talk about what you sell on the way of cure rates in the quarter and what your outlook would be, and maybe if you could talk about sort of the monthly progression through the quarter, were cure rates better or worse at the end of the quarter than at the beginning?

The last couple of months the cure rates have been comparable to a year ago, but a year ago was down from the prior year. So I guess the slippage occurred early in '02 and has been flat from late '02 to current. We haven't seen any further slippage as of late.

Let me just say that that's an overall. With respect to some of the select markets, some of the larger markets, there has been deterioration with respect to the Midwest. So quarter to quarter, I would say a couple of the states that have larger notice inventory have deteriorated. Not significantly, but they've not improved. That's the key.

Okay. Thanks.

Our next question comes from Richard diamond. You may begin.

Yes. Good morning.

Good morning.

As I model forward, should I, you know, continue to increase loss reserves unless I see, you know, job creation, let's say in the 150,000 job range?

I think what we tried to say earlier, with respect to the loss development, was that on a short-term basis, certainly the next three-quarters, we would anticipate seeing higher notice inventory, higher paids, and, therefore, higher incurreds in some reserve buildup. To the extent that the economy improves and we start to see a leveling off of notice activity in the second half of the year, that's when that could take place, but relative -- I mean, that's a forecast, and you'd have to use some factors as you think yourself, but we're not going out that far at this time. It would be too early to forecast anyway.

So if I --

I think the point is, the point we're trying to make today is that we have not seen any significant improvement that would lead us to think that the notice level has changed in the short term here, the next three-quarters.

Next question, you know, although mortgage refi rates in the M.B.A. index have come in, historically we're still, you know, we're still in a refi boom. What type of coupon would we have to see in a conventional 30-year mortgage, to see this refi boom come to a halt, let's say a sub-1,000 index number?

This is Mike. I think, relative to trying to forecast that M.B.A. refi index, I think, is a trick unto itself. But current market conditions are about 25, 30%, originations, are running refinancing. So current coupon rates would have to fall fairly considerably back to the June levels in order to respark another refinance wave, given how much has come through. And at the rate we're at right now, kind of at 6.25 to 6.5%, no-point, no-cost origination loan, seems to be leveling off pretty well at that level. Today the 10-year's up a little bit as well.

If rates hold flat or increase at all, you will get refinances back to the more normalized, I think 15% of the market, 15 to 20%. And that's always going to be inbred in the marketplace as I see it going forward, given the information in the marketplace, as well as the prevalence of wholesale lending whereby brokers are calling borrowers any time rates dip just for a day or something of that sort. So I think you've always got 15 to 20% of the market refinancing.

So I guess what I hear you saying, then, is that versus historical patterns, we should assume a higher level of core refinancing because of changes in borrowers.

Right, and we've talked about that for quite some time, and that's why if you've followed our business, whereas ten years ago the persistency rate on a normalized book of business would have been in the low 90's, today it's in the low 80's. And as a result, because of the information as well as the change in origination that's resulted in refinancing, plus the no-cost refinances, just a higher level of refinances going forward, but that's been the case probably for the last three years.

Thank you very much.

Our next question or comment comes from Jim Kitsinger. You may begin.

Good morning. Mike, I'm wondering if you could give us some granularity around some of the older bulk books. I mean, they're far enough along now, I would think, to see whether or not they're within kind of the return expectations you guys had on the bulk side.

Jim, we entered the market probably in late '99, and those books of business, '99 and into '00, are still well positive. Are they a little worse than our target 65 projected out to be? I would say yes. We've adjusted through time our assumptions with regard to persistency, some severity assumptions, some incidence assumptions, so those older books there are probably a little worse than anticipated. The newer books a little better than anticipated. In total, we think we're on track to that 65% loss ratio that we've mentioned over time.

Okay. The average premium, you know, is sneaking up into pretty juicy areas. Curt, you talked about a forecast for next quarter, but if the persistency begins to shine through over the next 12 to 18 months, what you've got on the books now is what we're going to see, you know, in premiums earned. Should we be using, you know, this kind of level as we model this out two, three years, assuming there's not another refi boom, and given that a sizable amount of the book is now fixed at pretty low rates?

I'm not quite sure what you're asking, Jim.

Jim, Mike here. I think what you're saying is relative to persistency improving, will we continue to see an improvement in average basis points? Is that what you're saying?

Yeah.

I would say no, because remember, a lot of what's driving the average basis point is the bulk business. That's going to be a shorter persistency rate book. Larry can talk a little bit about that. So, yeah, the flow book will come back and get into -- in theory let's say it gets back into the high 70's, early 80 percentages, if we're lucky, but that's a lower rate compared to the bulk. Larry, why don't you talk a little bit about consistency in the bulk?

Well, the bulk business has held up pretty well in persistency because of the prepayment penalties. We assume it generally to run off quicker than flow and in this interest rate environment, it performs a little better than flow, but as rates stabilize here we think that will turn around. But a lot of it is two and three year arm with a pre-paid penalty of two or three years, and once the two or three years is up, the prepaid penalty goes away and we see a lot of runoff. A lot of it is also arm tide to the LIBOR, so it's also a function of what LIBOR does. A lot of the businesses cash out, and it then becomes a function of what property values do, and does the rate environment and the appreciation of homes allow people to continue to cash out and churn the book or stay on the books.

And, then, relative to the premium rate on bulk, you know, it can move around quite dramatically. As Curt said, the pricing has been strong, the product mix has been good, but that can change. The execution can change from coverage down to 50 or 60 LTV's to an execution down to 80, and we would change our coverage, and then, correspondingly the premium rate could change or the mix could change. So the bulk side, the premium rate quarter to quarter could vary much more than on the flow side where it tends to be Freddie, Fannie, coverages on 90, 95, these 25-30 percent coverages,pretty stable quarter to quarter.

What you're saying is, my ability to model any stability into this relative to the past, you know, three to five years of change, it's going to be pretty hard for me, as an outsider looking in, to have a highly confident forecast on that number looking out a couple of years?

Well, if you get out a couple of years, I would say, for the next five quarters, you could, but not beyond that, Jim.

Okay. And one last question. Can you comment about market share and, you know, given what's gone on with a couple of your competitors backing away, how you see that shaping up here now?

Yeah. Again we were encouraged, as we took a position a year ago that we needed to take, and it's been pretty well document relative to the points we've made by S & P, relative to the validity of it, but be that as it may, it was the right thing for us to do as a market leader. General Electric and United Guarantee have recently followed that. Year end, I believe, is when they will implement their discussions with other customers. So, while I don't know where that will lead, I do know it will lead to some market share improvement with those customers that had dealt with us in reducing our market share. Collectively, anyway. Individually we may still have accounts that we may not increase with right away, but over time I think we will. So I can't quantify what the improvement might be with those customers, Jim, but I can say on a general basis it will improve collectively.

Thanks, guys.

You bet.

Our next question or comment comes from mark Gioroni. You may begin.

Hi, fellows. One question about the bulk business and the sub-prime. You talk about how losses are continuing and I'm trying to figure out there, what's the difference between the maturation portion of these losses continuing and the decline in the economy? If the economy was fine, how much longer would losses continue to increase in this book just simply because of the maturation process of it?

Well, on the bulk, like I said earlier, we started in '99, we wrote a few billion dollars into '00. We probably for the year were close to ten, and starting in '01, 2, 3, we probably were on a track of $20- 25 billion a year. So we really only have maybe three years at, let's say, a steady state level of about 20 to 25 billion. We need probably a couple more years at that kind of level to kind of get to a steady state with regard to delinquencies and whatnot. We're still growing the in-force, and as a result we haven't reached that steady state maturation that you mentioned.

On top of that, the economy is softer, and it's really kind of hard to split the effect of the two, but from a maturation standpoint, I think we need probably another six quarters or so at the levels we're writing to kind of reach that steady state point.

Okay. Earlier you mentioned the 65% loss ratio for this business. For your overall business when I think about things settling out, whenever that happens to be, what do you think of as the sort of normalized loss ratio for this overall business and if possible a normalized expense ratio?

On the flow business I would say 35 to 40% would be the loss ratio that I would work with. Now, over the last ten years that has not been the case, given the economy we've worked in, but also I will tell you, and I've talked about this before, too, that credit standards today are not what they were prior to 1998. A lot more business is being done with lower FICO scores and higher total debt-to-income ratios than was done prior to. So I think that long term has added one claim on per hundred. So I would expect on the flow business to be running in the neighborhood of 35 to 40% loss ratio and expense ratio on our all-in book, flow and bulk, in the neighborhood of 15 or slightly less.

Great. Thank you.

You bet.

Our next question comes from Geoffrey Dunn. You may begin.

Good morning. I wanted to circle back to your loss line. It seems that that's where the most volatility is on all the earnings projections. Last quarter you thought your provision would be 75 to 80, this quarter it comes in at 97. Is there any way that you can give us firmer guidance as to if we see no economic impact, you expect your reserve provision next year to be X, if we start seeing certain things at some point in time it's going to be this? Because this quarter we had a wide range of estimates on the street, and it seems like you're setting the stage for a similar circumstance for next year because of the questions in the economy, and it seems if you run your assumptions through you could be anywhere from south of $4 to north of $5 next year.

Yeah, I think the best think you can do is to assume that the economy didn't get -- doesn't get much better, I guess, and to run it at an assumption that notices continue to increase at a certain rate every quarter, and if that's the case, you continue to see not only a sequential increase in claims quarter to quarter, but an increase, if you will, in incurreds, and probably the same level of increase in the reserves. So that would be a baseline that you could use. You know, the second half of the year I think is going to be a key relative to what happens to the year forecast relative to what happens to notice inventory. You're really looking at a forecast that says where's the economy going to be and what impact will it have on this book of business in the second half of the year. The first half, I think we've talked about, this fourth quarter and going into the first half of '04 we see an assumption of continuing to increase notices every quarter as well as an increase in paids, and therefore taking the reserve up on a commensurate basis that we have. So the only assumption I can tell you to make is assume that it doesn't get better and have a base forecast.

Okay. So based on all the different forms of guidance you've given us on the call, it sounds like earnings decline in '04 against '03 is appropriate.

You could assume that, depending on what your assumption was in the second half of the year.

Thank you.

Our next question or comment comes from Alex Orloff.

Good morning. Couple of questions from the top line. The first question. You mentioned the persistency rate might go up to like 70 to 80% on the non-bulk business. What is the timing on that? Is that - you'd expect to happen in Q4 or over the next several quarters?

No, no, that takes a year to get there, a year-plus to get there.

Okay. That's what I thought. You commented briefly on the growth in the new premiums written, in the new business written. How do you expect-- the growth in the last quarter was pretty substantial. What do you see the growth going over the next several quarters? I'm talking about in new primary insurance rate.

Well, are you talking about NIW or premiums?

NIW.

NIW would decline. Flow business in general is declining as we talked about with respect to the lower refi activity. The purchase money is increasing, but net-net, NIW should trend down year to year.

Last question, once again on the tax rate, you mentioned that the tax rate is likely to stay roughly at the last quarter's level. Again, the time horizon on that, is it kind of a permanent decline, or if we look at to 2005 is it still going to be at those levels?

Depends on the mix but certainly in the next two to three-quarters it will stay about this level.

Great. Thank you.

Our next question or comment comes from La Von Von Redding.

Good morning. A couple of quick questions. As you lay out the scenario, it seems a fairly negative one in terms of your expectation on claims incurred and reserves. Is this something that you as being more MGIC specific or because of your focus in the Midwest or do you think this is kind of an industry-wide issue?

Well, the credit standards on the flow business are an industry issue. MGIC has more business in the Midwest than all others, and as a result of that we should participate more in that sector than others. So I would think because of that, our dominance in the Midwest, that, we might participate slightly more than others in our industry. But the lack of employment and the credit standards that we've been dealing with are still issues nationally.

So while, as you mentioned, it's negative, that may be too strong a word, but in the short term, as we look at things, which is how we look at things, we don't see improvement, and that's what we're modeling. To the extent things improve, things will be better than we're talking about here.

Okay. The other comment, I guess, comes from GE and United Guarantee kind of talking about behaving more rationally, in my mind. I think you commented that that's something going to be implemented toward the end of the year, so you're actually not seeing any of that?

None of that has happened yet. Their discussions with lenders takes place-- well, those contracts run out at year end, so they will be having discussions with customers at that point in time.

And while you've made public your thoughts on the subject, what about, I guess, if others were to do it, or other public companies may not be as willing to be as public about their comments on it, when would you be able to see them behaving more rationally in the market if we're talking year-end changes at the other two guys?

Those share discussions with the United Guarantee and General Electric customers, those allocation of business would start taking place probably a couple months later, because what's in the pipeline will continue to be booked. So, however customers deal with those changes that would be reflected, you know, probably March or so, and impact the rest of us, if that's your question. I'm not sure if that's your question.

Yeah, that's pretty much it. I just want to get a sense for when you might see people behaving somewhat differently.

Right, and again The S & P is pretty much documenting what we've been talking about for the last year in their recent report relative to the returns on that business. Again, from my standpoint it's a position ultimately that needs to be taken by all within the industry, given the slightly higher loss rate that we see going forward on the business up into the 35 to 40% on the flow business.

One final question. I was a little confused by a comment that you made -- that was made earlier. I guess one of the questioners asked the question about a steady state kind of maturation model, and you commented on needing six quarters, at least, in your mind, to kind of get there. I think the assumptions were that the economy was kind of normal. Could you just kind of explain how you kind of came to that six-quarter kind of concept?

I think at that point in time the percentage of new business that we're booking will equalize the percentage that we have in our in force book. To date, the amount of new business that we're writing in the bulk area is slightly higher than what it is relative to the percentage of bulk business on our in-force booked. I think it's like 27% versus 24%. And so in another four, five, six quarters, I think we'll be at a point where those two are equalized, and we'll be at a steady state relative to inforce to new business.

Gotcha. Thank you.

Our next question comes from Steven Eisman. You may begin.

I have two questions. One, obviously as the refi boom ends, your underwriting is going to be --. Can you hear me now? Two questions. You could be of help here. As refi's end, ,your underwriting trends are going to be coming down but I would assume at the same time that your contract underwriting revenue will also come down.

Yes.

My question is, is that going to be a wash or a net positive? That's the first question. And the second question is, given the data that you put out, I can calculate the number of -- not just the -- obviously the dollar amount of paid claims, but also calculate the number of paid claims, and by my calculation, the sequential increase in the number of paid claims for the quarter was around 1200, which is roughly two to three times [inaudible] maybe the company's --.

Well, I think I -- there's a couple of questions there. On the first part, with respect to the fall-off of revenue and underwriting expenses, they probably will go hand in hand, okay? That should be a wash unless we find our economic events that we can work on and be more efficient. And I would say, as I've said, it would be a wash, but what happens when we're in huge contract underwriting volume market, Steve, is that our sales and underwriting organizations get so wrapped up in trying to arrange that contract underwriting. You know, we have like 800 people on a contract basis, situated throughout various lenders offices throughout the country. It takes up a great daily of sales and management time that could be more wisely dedicated to getting new customers and better service to customers. So, while on a cost basis it's a wash, relative to procuring new business it's detrimental to us that we don't have that time spent chasing down new customers and getting more business from existing customers.

So just the fall-off being, let's say two, three-quarters now, where contract underwriting is down, bottom line impact is a positive or a wash?

Probably pretty close to a wash, I think, at this time because we're early into it. Remember, this just started this quarter. Now, the second part of your question, I think, you faded out again a little bit --.

I'll repeat it.

But I think you talked about the fact that somehow you've calculated maybe some theoretical rate of number of claims or something, correct?

I mean, you give out the dollar amount of paid claims.

Right.

Then you give out the -- which was $124 million.

Right.

You also gave the average severity.

For different types, yes.

For different types. So, if I just take the dollar amount of paid claims and divide it by the average severity, I come up with some number that roughly approximates the actual dollar amount of paid claims.

Yeah.

So, for example, that would, with a simple calculation, be 5,511, by my math in the third quarter, versus 4,273 in the second quarter, for a sequential increase of 1,238, which,just given the number that you give out, is a much bigger sequencial increase than the company has ever had. In fact, as I understand, is that significant or not?

The number is not that great. I think you've got some other factors involved there, because it's apples and oranges. But there has been an average increase on the number of claims we processed on a daily basis, but it hasn't been to that level.

I could follow up off line.

Yeah.

Thank you.

You bet.

Thank you.

Our next question comes from Bruce Harding. Sir, you may begin.

Curt, I thought I was going to get in on time, I apologize, but I pretty much missed all your prepared remarks. Just heard you say something about the tax legislation, if you could, if you don't mind repeating that. And then, can you just tick off any other highlights on the political front? I didn't - I think I had a pretty good understanding of what FANNIE was trying to do a week or so ago with regard to its policy paper regarding the federal home loan banks, but if you can make any comment on that? Then, any implications beyond the obvious on GE dropping to a AA, and then, you know -- well, I will stop there.

Okay. Relative to the AA, let's go backwards, anyway, the only implication there, I think, it was a wise business decision on their part relative to utilizing AA capital in a business that's AA necessary. So I think that was a wise use of their capital. The AAA had really no benefit to them, given how the business is done today.

On the mortgage insurance tax deductability legislation, we have strong bipartisan support in both the Senate and the House. Unfortunately, everyone is detracted, given what's going on in the world, and in order to get this legislation done, you need a tax bill to attach it to, and I don't know if there will be one this year. And as a result of that, I do think, however, next year there will be various vehicles that this could be attached to. So I'm highly confident, and I am optimistic by nature, that next year, because of the strong bipartisan support -- again, we have no one that's a detractor here because it is going to help more people get into housing -- that it will get passed.

So I guess the question to me is not will it or won't it, it's just when, and that will be very positive to our business, as I say, as we deal with 80-10-10 structures and 80-15's and 80-20's, along with, as I stated, you may have missed, the changes in the Basil 2 and the credit given to mortgage insurance, also very positive, as well as, hopefully, as the economy improves, the banks will start making commercial loans again, and delve into the world of the second mortagages. Those will all be very positive in getting that 25% of the market that we've lost to those structured transactions back to our industry. And also, I think it's hurt us on an adverse selection basis, too. I think a lot of good credits have utilized that scenario because they weren't offered elsewhere.

Relative to the GS 2 legislation, that's -- what Fannie Mae was floating really is what -- whatever they're doing, they're doing. I can't comment on that. We had questions relative to spread accounts and federal home loan banks, and the reality is that the federal home loan banks use primary mortgage insurance, and the discussions were really below that layer of primary mortgage insurance could utilize spread accounts. So that discussion was one that, in your world, things bubble up and people overreact, and the reality of it, as we understood it, anyway, was one that would live within a normalized primary mortgage insurance market but have the ability to do something underneath that. So if anything, it would help promote more mortgage insurance. Did I hit your questions?

Yeah, Curt. And then the -- let's say everybody, you know, followed your lead, as GE and AIG have, and dropped the 40% seed captives, is it likely your share would go back to previous levels and, you know, any implications in terms of, you know, when that might happen?

Well, when it would happen, I have no idea, relative to the other companies. Again, we can only speak for us, and we've got the three major companies in there now. Relative to market share, it would be my expectation, and the people would certainly, relative to our sales force and underwriting organizations, it would be expected that we would, at a minimum, be at those levels of old going forward. What we have done through this process, Bruce, is really worked hard relative to the, if you will, the ten customers that dealt with us on a less basis than they had prior, and really tried to maximize our business elsewhere with many other customers. So we've done a good job of expanding our share elsewhere to keep it at that 22%. And so, you know, if everyone came back to a normalized level relative to the mortgage insurers, I would be optimistic we might do a little better given the good job our organization has done in moving share with other customers.

And, Curt, on the bulk, to Mark's question before about three years of seasoning on bulk and probably need another couple of years before you get to that steady state, have you changed your bulk insuring practices, or are you cherry picking from these pools for different types of loans? Or are the policies that were in place back when you got into the bulk business, in terms of your decisions on which loans to take and which to exclude from these packages, still the same, in terms of the mix? Thanks, that's all I have.

We've changed the pricing a great deal, as Larry has talked about, and looking at, in particular, at the combination of higher FICO's with lower debt to income debt ratios. We've changed that pricing significantly and bounced a lot of those loans out. We learned that from the '98, '99, and early 2000 books, and it was the second half of 2000 that we really jumped pricing and have moved it up since. Cherry picking is too strong of a word, but I will tell you we have more opportunity than what we're writing, so we are looking at transactions and doing the ones that we feel are the most profitable to MGIC. Anything else, Bruce?

Thanks a lot, Curt.

Okay, thank you.

Our next question comes from Chris Buonafede. You may begin, sir.

Good morning. Can you guys tell me, obviously, underwriting expenses have come down due to contract underwriting. How much in dollars do you think that can come down in terms of the contract underwriting piece as we go forward over the next year or so? I'm trying to get at how much leverage is there in terms of dollars.

Well, I think we're still working that issue, because asCcurt says, we still have a market share strategy, and we're looking at working with other account, et cetera. I think certainly the next quarter will take it down a couple more, then what we do next year, a lot is going to depend on what the competitive landscape is and what we need to do to support customers, et cetera. So at this time, it's slow as we go.

I guess maybe, can you guys tell us what the dollars of contract underwriting expenses are embedded in that $77 million or so?

No, we don't break that out.

Okay. Fair enough. Thanks.

You bet.

Our next question or comment comes from M. W. Montgomery. You may begin.

Thanks very much for taking my question. I understand that you booked substantially all of your underwriting profit in year one, and that later on you say, all things equal, a decline in new insurance written would expect to the adversely impact the overall result. So it seems like persistency, void of new mortgage origination growth, has a negative aspect to the result. So my question is, for modeling purposes, if we were to have a situation where we didn't return to all-time mortgage origination growth, what would the negative aspect of persistency be on the results, please?

Again, I'm not quite sure I understood what you're saying, but I know the risk factor you're referring to. And unfortunately -- or fortunately, all things are not equal, when we go through those kind of climates, because we come off a book of business that's been higher, and go to one lower in volume. The reality is the purchase money transactions dominate those years of origination, so we participate in a much higher percentage of the loans that are made that year, so they don't fall off like the general market does. That's point one.

Point two is we're much more productive relative to contract underwriting. Again, in a refinance transaction, about 1 to 2 per 100 go to a low down payment mortgage, or, I mean, out of 10, where as in a purchase transaction, 4 out of 10 do. So the work we're doing in that area again results in mortgage insurance much more dramatically than they do in refinances. So on the expense side we improve dramatically.

Finally, on the persistency, which we've talked about, again you're in a market where you're not refinancing, so as a result you're generally -- and I would say most generally, going to be increasing the book of business. Your persistency rate is improving, so your in-force back of business, even though you may be writing less, you're canceling far less, and as a result of that you're -- you'll grow that, and that generates 80% of our revenues. So while that risk factor has already caused some issues with the book today, that all things equal is not what happens in our business, and that's the part that I guess people can understand.

Okay. Thank you.

Our next question is a follow-up from A. J. Rowell. You may begin.

On a monthly basis have you seen Tennessee rate stabilize or start to improve in the third quarter or what's the trend on that, on a monthly basis?

Well, this last month, yes, you know, so that's what we talked about, going into the next quarter, as Curt mentioned earlier, we anticipate seeing a slight up tick. Now, remember, we measure it year to year, so there will be a slight uptick in December vis-a-vis a year ago and vis-a-vis the -- not a year ago, but vis-a-vis the third quarter.

A. J., remember that the processing lag between the reporting activity and the lenders coming up in, sought should have more in month to month application data than it is in the cancellation rates in the third quarter.

And on the average premium rate, now, correct me if I'm wrong, as refinancing is declining and purchase transactions dominate, I would think the drag -- the reduction of the drag as folks, you know, less folks refinance, that the average premium rate should stabilize, maybe even improve, because you're now adding higher premium business into your book.

That's correct. What happens in a purchase money market, again, back to all things not being equal, is you're going to do more 95's,9 7's, and also more ARMs, and each one of those is a higher premium charged, and so you're exactly correct.

Thank you.

But that's the flow business.

Right. Thank you.

Our next question comes from David Hochstim.

Hi. Sorry if I missed it, but did you say what the average claims in the Midwest were relative to the book over all, or just the delinquency rate?

Just the delinquency rate. Just that the rates were higher, David. The delinquency rates as well as claim rates.

And then with respect to -- we assume it was less than investment gain, if you didn't disclose it? Not significant. Following up on what Curt said about the federal home loan banks, is there some opportunity forthe industry to generate incremental business if the federal home loan banks are regulated or subject to the same standards as Fannie Mae and Freddie Mac?

Well, again, I don't know if it's incremental business. It has opened a new avenue to serve relative to the business itself. I don't know if that would be incremental in that we would be doing it with Fannie and Freedie. We have been doing some full coverage beneath their writings, so that has been incremental to us to date, in addition to the primary insurance we're doing, but we would have been doing that primary all along. So the incremental opportunities, you know, I'm not quite sure how much that would be. I don't think it would be significant, David.

Okay. Thanks.

Our next question comes from Blaine Marter. You may begin.

Can I infer from your comments today that you like the level of bulk business where it is as a percent of in-force and as a percent of new premium written, so kind of a mid-20ish?

Yes. We actually are managing at that level. As I mention we'd we've had more opportunities to write it, more transaction. We may have quarters where we go slightly higher to that, approaching 30%, where we feel there are great pricing opportunities relative to the returns, but in general in the 25% of our business.

All right. I guess I need to be convinced that this is good business. I mean, how -- can you just give me a sense of the returns? How are you judging this business? My sense is that would you actually be rewarded in the market if you actually slow your premium growth a bit, maybe buy in some stock here.

Well, again, on a return basis, which you don't have, you don't see the premiums broken out relative to the incurreds and the paids, it is very profitable for MGIC. The early writings, the '98, '99, and early 2000 were basically break-even for the company, and since then we're having loss ratios and expect loss ratios to outperform that 65%. So the reality is, on a profit and loss basis, it is something that is very profitable for MGIC.

Remember, one of the keys to this strategy was that, and not part of captives, so you don't have a 30, 40% seed given right off the top.

Right. Thanks.

Our next question comes from Paul Miller. You may begin.

My question has been answered. Thank you very much.

Follow-up from Geoffrey Dunn.

Jeff actually had to take another call but his question is, your expectation for the long-term expense rate show on the overall book is 15%. Coming out of the refi wave what do you think your chances of achieving that by year-end '04 are?

I think we have a good shot at that. It would be an expectation. Thank you.

Our next question comes from Robert Hawkin. You may begin.

For some reason I wasn't getting in on the queue, but I'm glad to be here. I'v got a quick question and a then follow-up. The 45% persistency, is that the flow and the bulk?

Yes.

So as the overall persistency rate improves in '04, the benefit is really more to the flow side than it is to bulk side?

Yes, because right now the bulk is higher than the flow.

And that means that, you know, and everyone has talked about all things being equal, the flow should be then, you know, should, in fact, become a larger percentage of the business, all things being equal. And as you then normalize out with persistency rates, as you've indicated, maybe in the low 70's, your mix of business, you know, your IRR's of the business between the flow and the bulk, how does that suit you? How do you feel about that, particularly with the flow business now capped at 25%?

The bulk business at 25%.

Just in terms-

Of the captive?

The captive capped at a 25% seeding commission. In other words, where would you rather be? Clearly, when the persistency rate increases in '04 your mix is going to shift, and how do you feel about the IRR's of that business in terms of, you know, that is going to shift more towards the flow business, I take it.

Yeah, I mean, where would you rather be? You'd like -- your first preference is to write flow business, no captive. Your second preference is to write bulk business. Your third preference is to write flow business with the 5-5-25 cap activities, relative to returns. So that's what-- what we'd really like is to be back to the days of old where there were no captives and just writing flow business, but that's not reality.

In a way, what will happen is that in '04 the optics may look a little bit better but you're saying, you know, the I R's, or the mix, is going to be, you know, perhaps a little less attractive.

I wouldn't say that, Bob. I'm not sure what leads to that conclusion.

Well, if the, you know, the business that you're writing today, and then through '0 H -- through '04 is going to persist longer --.

Which is very good.

-- it would be -- would become a larger mix of your business.

Again, I see that as positive. Maybe it impacts our returns a tenth of a percent or something, but it's going to generate a lot more revenue and income for the company.

Keep in mind, we talked about maintaining the mix of NIW and in-force at the 25%, so it wouldn't revert back to 15% or 10% of the in force. We're going to try to maintain at that 25% level.

And I'm going to follow up off line on that as well. Okay. And one other follow-up to the federal home loan banks question. I take that it what you're, you know, what the political issue is, is that if the status quo is retained, you know, you will continue to have, you know, a nice niche business, which is this G E pool insurance product that you're offering the federal home loan banks, and so that's really the issue, plus the primary business that you're getting from them on a flow business.

Yeah, but that primary would cap regardless if it's Fannie-Freddie purchased or federal home loan bank. The initial opportunity has been relative to that niche, as you mentioned. And, frankly, I'm a person for competition, so I think their part of the marketplace is something good for all.

All right. And one last follow-up to Bruce's question. You know, your comment about how any change in the tax deductability feature, you know, requires a tax bill and so forth. Why haven't the GSE's, you know, been more forceful in promoting, you know, this product? It would seem to me that they're losing out in a number of respects. I mean, they're in effect writing 80% and getting guarantee fees on loans that are really riskier , and then they're losing, you know, a significant amount of principal balance on which to get guarantee fees. Why haven't they been more aggressive with you in terms of promoting legislation in this regard?

You would have to ask them on that question, Bob, but the realities are, from my standpoint, the structured transactions are not good for the marketplace at all. I think the risk has been underpriced, relative to the banks that hold those notes. I think Fannie Mae and Freddie Mac, as you referenced, are buying 80's that are really 0's, 95, and 100, and they don't like it, but it's been a customer issue. So why they haven't on the legislation, that's a question for them. But I don't think it's -- the whole process has not been positive for the marketplace.

Where is your political support on this issue? The banks are huge, you know, are huge beneficiaries in terms of their --.

I don't know if they're huge. It's just been an outlet that, really, the only outlet they had given the lack of commercial business, so this has been a niche that they've been able to participate in. I think, longer term, this isn't something they want to dominate and I think that will come through in losses. So there is no -- you know, there is no opposition to the legislation, Bob. That was my point earlier. It's just the quirkiness of our political system that you can have both sides of the aisle agree that this is good, and numerous coalitions and others support it, and yet ,because of other more pressing matters, legislation doesn't happen. So that's where I say I'm confident it will happen, it's just a matter of when, not if.

All right. Thank you.

Our next question comes from Jonathan Gray. You may begin.

Yes. I'm a little baffled by the fact that the bulk loss experience, presumably, reflects home prices around the country. It is all secured, collaterallized lending, and yet the loss experience on that book I would assume has been surprisingly negative. It's been more negative than the company had anticipated a year and a half or a year ago. Why is that the case? With home prices so strong, that the loss experience seems to be performing badly relative to, let me say the company's expectations. Since you've been raising your guidance with regard to losses and provisioning, by definition I would assume have been surprised.

I think, Jonathan, first of all, it's good to have you back on the call.

Thank you.

I think it's more across the board economy, both books. As Larry talked about before we're hitting our target on the bulk business relative to loss ratios and expected loss ratios, okay.? Overall claims are up more but it's across both books, and it has to do a lot with delinquencies from all parts of the country, really, and no significant deviation, if you will, between those books.

Well, my question, then, you know, really remains. In other words, how is it that home prices are fairly strong everywhere? There aren't any particularly weak markets. In fact, the markets are surprisingly strong, virtually everywhere in the country. How is it that the loss experience seems to -- it's nice that people -- unfortunate that people are unemployed, the employment rate has moved up, clearly. But again, as long as there's some equity in homes and home prices are rising, it's a little difficult to understand how it is that the loss experience is performing worse than expected across both books. I don't understand it, but I guess there's no clear answer.

Jonathan, what we've been saying for almost eight quarters now is, we've had a steady increase in delinquencies across the market. Maybe there's a couple of states, California is an exception, they've been all right, but particularly in the Midwest and the larger states, Florida, North Carolina, Texas, and all the midwestern states, we've had an increase in notices sequentially every single quarter for eight quarters, and it is a function of the market we serve, the first-time homebuyers, and the economy and their employment picture. And real estate markets haven't been bad. Our mitigation is down slightly, but it's still a factor, but overall what we've seen is a continuation of the deterioration, if you will, of notices, some deterioration in claim rates, and it is a function of the market we serve.

And don't be fooled by averages or median house price movements. Take any market, it might have 3, 4, 5%, on average, price appreciation but within that market there are areas doing better, areas doing worse, even declining. And when you have an individual who loses a job, can't maintain the home, and it's in a part of the market that may not be keeping up, you result in a loss.

Jonathan, the real-estate values are up, no question about it. The rate of being up has slowed rather dramatically from a couple years ago. And I think it's telling from our books of business, if you looked at our 2000 book of business, we had the opportunity to mitigate 38% of the claims that we had through rather than paying the percentage, either buying and reselling that property and losing less than the percentage, or doing a pre-sale in conjunction with a lender that was less than what the percentage option was. Today about 26% are in that category, so it's fallen from 38 to 26%. Still indicates values are up, but not nearly what they were growing relative to the past. And as a result of that, on first-time homebuyers, if you're not having, where they've got three to five percent, possibly 10%, into the transaction, and then you cut across the realtor and other costs involved with reselling, you are going to get claims, because there's minimal equity unless there's been dramatic increase in the real-estate values.

So, longer term, what's going on in the real-estate market is very positive for us. When they go up very quickly that leads to claims a couple years down the line. The slowdown is very helpful, but it has eliminated opportunities for people with minimal equity to sell their house prior to foreclosure, and it's also cut back on our opportunities to mitigate loss by 38% three years ago to 26% today.

Thank you.

Our next question or comment comes from Ed Grotians. You may begin.

Hi. I know you got out of the D P captives, but I have this question. Have you been sharing losses with the captives?

Other than the quota share transaction that we have in place where they share first dollar losses, we have not had any of the excess loss treaties hit into their layer.

And I guess I'm surprised at that, given the trends that you're seeing, what is, you know, I guess what is it, the fifth claim goes to them? That's our whole point.

We're not surprised and that's why we did what we did.

But are the 25% less sessions, are you sharing in those?

No.

Those are higher trigger points?

None of the treaties that's, whether they be at 3, 4, 5, or a 100, have the risk kicked in.

And the quotas?

Quota share we've had in place for sometime, also have some customers utilizing those today but there they get first dollar losses as a portion of the premiums that they receive.

And just to follow up, you talked about the mitigation going from 38 to 26%, is that all just the slowdown in the appreciation of home values that is influencing that or are there other factors there as well?

No, that would reflect -- I mean, ultimately the value is the ultimate scorecard, so predominantly reflects the slowdown in the values within that market.

All right. Thank you very much.

Operator?

Yes, sir.

if there are no more questions, we thank everyone for their interest in MGIC as always. Thank you.

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