MGIC Investment Corp (MTG) 2005 Q2 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the MGIC Investment 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. If anyone should require assistance during the conference, please press star then zero on your touch-tone telephone. As a reminder, this conference call is being recorded.

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

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

Our earnings release of this morning includes additional information about the Company's quarterly results that we will refer to during the call. It includes certain non-GAAP financial measures. The release and this additional information may be accessed on MGIC's website, located at mgic.com.

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 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 turn the call over to Curt.

Thanks, Mike. Good morning.

Second quarter net income totaled 174 million compared with 155 million a year ago. Diluted earnings per share was $1.87 versus $1.56 last year. Our new insurance written rebounded in the quarter with writings of 16.6 billion, which was up 46% from last quarter and 3% from a year ago. The new insurance written was comprised of 10.4 billion of flow business and 6.2 billion of bulk business. Persistency ended the quarter at 61%, up slightly from 60% last quarter. The breakdown on that relative to the flow and bulk was flow persistency was 64%, and bulk was 52%. Insurance in force was basically flat with last quarter, at 172 billion, and down 5% year-over-year. Our average premium earned also was relatively flat at 72.5 basis points, compared to 72.4 basis points last quarter and also 72.4 a year ago.

In the quarter, we again experienced improvement in our delinquency inventory, with delinquencies totaling 76,000, down from 78,000 last quarter and 81,000 a year ago. Paid claims were 158 million, compared to 149 million last quarter, and 140 million a year ago. However, as a result of the lower delinquency inventory and stable claim rates, losses incurred totaled 137 million, 11% lower than the 154 million reported last year, but up from the 98 million we reported in the first quarter. Underwriting expenses in the quarter were 69 million, flat with last quarter, and down 6.4% from a year ago. Our joint venture results continue to be excellent with a strong quarter -- another strong quarter from Sherman and C-Bass. Finally in the quarter, we repurchased 3.4 million shares, and have now repurchased 4.5 million shares for the year with approximately 5 million shares remaining under our recent board authorization.

Regarding the second half of the year, in many cases it should be more of the same. New insurance written we think will approximate the first half of the year, as should average premiums earned. Persistency should improve very modestly; and as a result, insurance in force we think will be down in the neighborhood of 3 to 5% year-over-year. Our delinquency inventory, which has improved significantly year-over-year, should remain level with our current position, and I think paids will also approximate the amount paid in the first half of the year. Expenses also will -- or should remain relatively flat in line with our volume expectations. And finally, I expect our joint venture results to be down from the first half, given the exceptional results both companies posted in the first half of the year.

With that, let's take questions.

[OPERATOR INSTRUCTIONS]. Paul Miller, FBR.

Thank you very much. Hey, Curt, you guys purchased 3.4 million shares this quarter, which is, I think, the most you've done in the last couple of years. And you said you had about 5 million shares remaining. I know you had -- you divvied it up about $575 million in May, I believe, when the announcement hit. Can we expect this type of pace through the next couple of quarters, or was this just an exceptional quarter because your stock was trading so cheaply?

Mike Lauer. Let me just clarify a couple numbers. We -- you mentioned a dividend of 500 and something. The dividend was 375.

I'm sorry about that.

Yes, and with that dividend, we actually dividend up 325 million in June, and we're going to dividend up another 50 in July, effectively. That's the flow of the funds. In addition, we also have our normal $44 million quarterly dividend that would come in -- in July and subsequently, I think in October. So we have two more $44 million dividends plus the remaining 50 I just talked about. With respect to the purchase volume -- repurchase volume in the second quarter, we wouldn't continue at that level, no. So everything else going forward, again, will be subject to the amount of dividends that we upstream, availability of operating cash, and anything else we're looking at with respect to use of funds.

I mean, I know you have a 5 million share remaining. Can you give us -- I don't want to put you guys in the corner but any indication of when you expect to complete that? Over the next 12 months, or the next two years, or the next --?

Well, that's the corner. I guess -- That is the corner.

You're not putting us in that corner.

But I think specifically as I mentioned, you know, we just got that authorization for 5 million additional shares and subject to dividend availability and operating performance and other uses of funds, we'll continue to buyback. But I -- we're not going to comment on what level we'll be at.

Okay. Thank you very much.

Thanks, Paul.

Edwin Groshans, Fox-Pitt Kelton.

Good morning.

Good morning.

I was really just -- I just wanted to focus in sort of really on the -- the bulk new insurance written. It's up nicely from the first quarter and I guess back to levels what we see at about a year ago or so. And with your comments about staying -- new insurance in total staying at the levels of the first half, does that mean the you're seeing better spreads out there, or more -- I know you've said in the past you've seen a lot of business come through. You just weren't winning in the bulk space. Is that changing?

Larry, you want to --

Yes. On the bulk business in the second quarter, we had a customer who had not utilized us the last few quarters come back. So a good chunk of the increase was due to them. With regard to going forward volumes, once again, the spreads are still pretty tight, although, we are seeing a bit towards the lower tranches, a bit of widening occurring. So that may help us, but if -- if -- aside from this customer I spoke of, we think volumes would be more in line somewhere between first and second quarter volumes for the remainder of the year, unless spreads widen here, which showing some signs but not a lot of movement. Because the -- some of these customers coming to us, as we said in the past, can be pretty lumpy, choppy. So.

Yes. It really was a reflection, as Larry mentioned, that we had a strong customer who had returned to us, and we had some nice opportunities with this customer in the quarter. And our expectation is to do like 3.5 to 4 billion per quarter in the last two quarters.

And then on the flow side, it moved up nicely also. Are you seeing more come back to you, I guess -- is that a reflection of single file or is it just a movement back from the flow channels?

I think seasonally, we usually do a little better now this time of year, also, relative to volume. Single file continues to be successful. We continue to hear from large aggregators that are planning on adding single file as a product, which is promising. I think it's about 5.5% of our volume. We crossed the 1 billion threshold sometime within the quarter. So that's making an impact. I think it's also shown by our FICO scores. You saw a little uptick in FICO scores that we're insuring, which I think is a reflection of single file because it impacts that credit range, the higher credit scores. So all in all, I think penetration is -- for our industry is improving slightly. And that's a reflection of that, also.

And this is Mike Zimmerman. I'd also -- the origination market in general is up pretty significantly from original year forecast. I think the latest forecasts are around 2.6 trillion versus last year's level about 2.7. Beginning of the year, I think everyone was looking for about 2.1 or 2.2. So some of that increase was also just a function of more origination activity.

Excellent. Thank you very much.

Thank you.

David Hochstim, Bear Stearns.

Thanks. I wonder if you could -- maybe Larry could talk about some regional -- if there are any regional trends in terms of credit deterioration or improvement in the Midwest. Is that looking any better? And could you give us some sense of what kind of exposure you have in California and Florida and, say, Las Vegas if those are the markets that are most susceptible to some downturn, how little exposure do you have in those places?

Yes. Well, with regard to some of the geographies, with what's going on in the auto sector, Ohio and Michigan are relatively soft in terms of the employment picture here. California, it's still humming along. We're monitoring the housing statistics closely. We have not seen much in the way of deterioration. It's still chugging along, but we're keeping our eye there. With regard to the dispersion of risk, California between bulk and flow, I think we're about 8% overall; and Florida's probably right in that area. So we're well distributed. Bulk, certainly we've got more in the way of California, but that's offset by the flow of California. So all in all, we're about 8% both California/Florida.

And then -- and on the increase in the bulk in the second quarter, some of that's due to a lender having less ability to do senior substructures and needing credit enhancement [ph]. Why wouldn't that lead to higher bulk volumes for you later in the year? Maybe that's kind of a more common problem for issuers.

Well, a big part of our increase was a lender that came back to us. So relative -- relative to that continuing within the final two quarters, I'm not sure if that'll happen, David. But if the spreads aren't there, regardless of who it is, the business won't be there.

Yes.

And I characterize our bulk customers. We've got some customers that are pretty repeating, month after month, quarter after quarter. They're routinely there. And then we've got others that come in and utilize our transaction, I think, in part, just to keep various options alive and the channels open, not knowing what execution is going to be most attractive.

Okay. And then I guess finally, just in terms of those of us outside trying to guestimate what Sherman and C-Bass might generate in the way of earnings, can you give us some sense again of what -- what's kind of recurring and sustainable or semipermanent? If we try and take out more than the next one or two quarters, what -- is there any way we can guess what earnings could look like in 2006?

Yes. I don't know, David, because I think maybe what we can do is -- is talk about a base case, if you will, that -- base case would be that their earnings would be up, I guess I would say conservatively 5 to 10% would be a good base case for '06 relative to -- at this point in time. As we get closer to year-end, we can look at whether or not there's anything materially changing with respect to that forecast. But at this point in time, on a year-to-year basis, I think flat to up 5 to 10%.

They're a growing part of our business as they have for a long time now. Annually, they keep outperforming. They've got -- both companies, great management teams.

Right.

I think we need to get increasing guidance to you relative to that. We're talking about that yesterday.

Yes.

Because they are a growing part of what we do and how we do things.

I mean, they're up to 25% of earnings now.

Yes.

And -- and every quarter Mike tells us earnings will be -- well, it's the first time he said they'd actually be up in a year. That's pretty impressive.

Yes, I -- [laughter]

Usually they're down every quarter.

I had to laugh myself when he said it. Because I was wondering what he was going to say.

Okay. Thanks a lot.

Thanks, David.

Bye.

David Chamberlain, [inaudible].

Hi, guys.

Hi, David.

Quick question. Just could you talk about the bulk flows and -- bumpiness and where you are in terms of the seasoning curve [ph] you think, with regards to say the '01s, and '02s, and '03s that you wrote and how that impacts you think delinquency going forward in the gold channel [ph]?

Larry. '01, '02, '03?

Did you hear that question?

Yes. If I understood the question, where are some of the books with regard to the claim delinquency curves?

Yes.

Well, if you remember back, we wrote some pretty good sized books back in '02 and '03, 20 some-odd billion. So those are --

On the bulk side.

On the bulk side, yes. A couple are three years old. And they're pretty much on the delinquency side probably top of curve. And so in '04 and '05, we're going to write less, and those are moving up to curve. So we've got a good piece of the bulk in force at maximum delinquency claim part of the curve. So all other things being equal, just simply having smaller books on the early -- on the upward curves should result to flat to less delinquency activity, once again, subject to economic developments and credit quality and all that. Just from a size of book, seasoning of book, we've got the big books at peak of curve.

And just -- second question just on the flow. And I know in prior quarters, you've been talking about what part of the business you thought was being disseminated by the 80-10-10. Can you give us just a sense maybe if that's changed at all? Have you seen any pullback form the home equity product or any -- I mean, it sounds like you've seen some increased penetration. But just kind of give us a sense there of what you're seeing.

Yes. We haven't seen a pullback on home equity product. We've seen some increased penetration I think because of our single file product, but relative to what it's competing against, we haven't seen a pullback.

Thanks, guys.

You bet.

Geoff Dunn, KBW.

Hi. My questions have been answered. Thanks.

Thanks, Geoff.

Mike Grasher, Piper Jaffray.

Good morning and congratulations on a nice quarter. I wanted to come back, Curt, to your comments about the paids should approximate the amount in the first half of the year. Do you -- are you still guiding us to paid equalling incurreds overall?

Not overall, for the second -- or, for the third and fourth quarter. If you recall on the last quarterly call, I said that the -- we anticipated incurreds probably less than paids in the second quarter, but going into the third and fourth, we don't anticipate any material change in delinquencies. All that being equal, we would think that incurreds then would probably approximately equal paids, maybe slightly more, slightly less. So the guidance on paids is that paids will stay at about this level on a quarterly basis the next two quarters, and that'll still be within the range that we talked about early in the year, between 600 and 630, I think. And on an incurred basis, in the third and fourth quarter, I think incurreds will approximate paids. So they'll be higher than they are in the second quarter.

Okay. Thanks.

That's all subject to delinquency levels. I'm assuming there that delinquency levels stay flat. If, in fact, they're lower, that could be positive.

Understood. Thanks very much.

Thank you.

Rob Ryan, Merrill Lynch.

Good morning.

Morning.

Could you detail a little bit for us what you're seeing in sort of the components of what go into what the reserve levels should be and therefore what the loss provision should be? Namely, not only the default inventory size, but your expectations for the claim rate and average claim size?

Well, I guess number one, as I mentioned before, the driver on the incurreds would be first the level of paids. And as I indicated, they'd probably be at about this level. So you replenish, if you will, the reserves for the paids. We haven't seen much change in the severity, as you noticed quarter-to-quarter. And I would also point out that unfortunately, we haven't seen any significant change in claim rates. Especially in the -- I would call the Midwest or the manufacturing states, which would be the higher delinquency states. So albeit the fact that the delinquency levels have slowed down, we've not seen any significant improvement in claim rates.

So the driver, if you will, will be initially the level of delinquencies; they've come down. That's positive. As far as claim rates, I would say the larger area of delinquencies being in the Midwest and manufacturing states, they've not come down. They've remain flat. That's positive, if you will, from year-to-year development, but not that significant with respect to improvement. So I guess what we would be looking for for improvement in incurreds would be first the level of delinquencies and where those markets are coming from. We've not seen any significant change in claim rates. They look to be on track and as far as -- or not claim rates, but severity. They may be up modestly just because of -- of average loan value. And an improving trends with claim rates would be beneficial, but we've not seen that in the major states.

Great. Thank you.

Howard Shapiro, Federated Investors.

Hi. Good morning.

Morning.

Question about your exposure to some of the more exotic mortgage products. I noticed that your ARM percentage has stayed pretty flat at about 13%. ARM origination market share is a multiple higher -- several multiples higher than that. Can you tell us two things? What is your exposure to some of these more exotic products like interest only ARMs or the MTA ARMs? And is there any circumstances where you would consider increasing your exposure to those products and then maybe helping your insurance in force growth?

Let me take a shot at the -- the ARMs. Our full business largely ends up with Freddie/Fannie. And they're a great place for fixed rate, but the banks and whatnot are pretty -- pretty good, competitive on ARMs. So through time, our flow business has been roughly 90/10. 90% fixed, 10 or-so-percent ARM. And -- and that's what it is now, and I guess historically, we kind of moved into maybe the 15% range if rates got much, much higher making fixed even less attractive. But fixed rate's pretty attractive and we're at 90/10 there on the flow side. Most of that product is the 2-6 treasury ARMs fully amortizing, no potential NegAM. So it really -- in the way of some of these new option ARMs, as far as the in force -- total in force bulk and flow, we're probably a percent of the portfolio bulk and full combined in the potential NegAM, which would include the option ARM. So we haven't seen much in the option ARM. One. Flow side, we're not -- it's mostly fixed. And then on the bulk side, there's a lot of ARM activity, but it tends to be the two-year, three-year ARMs which are fully amortizing. They've got an adjustment of 2 or 3% after the first two or three years. And then a lot of it is tied to six-month LIBOR adjusting a percent or so every six months with lifetime caps. But once again, the payment and the interest rates move together so there's no potential NegAM.

Interest only, there's been a lot about that. It's a recent phenomenon, meaning we're seeing more on recent writings on bulk, probably 20 or 22% of the bulk is interest only. But the interest only, the vast majority of that 20, 22% has a IO period of at least five years. And on the flow side, recent writings may be 5%, and once again, most of that is five-year. So the point being, I think IOs is a good tool to improve affordability, if you put that IO period out there five years or more. Why? Because in five years or more, a lot of people will have moved on. Our half life on bulk, it's probably a couple of years. Three years or so on flow. And so a lot of people, over half, will be gone by the time the IO period kicks in. And if they're still there, five-plus years of appreciation has a better chance to have caught up and give some latitude out, if you will, if somebody can't cope with the payment shock at that point. We're concerned about IOs in that three-year period.

The sector we serve, Howard, is just not one that participates as much in the exotic instrument stages as others. So your question relative to market share, would we get more industry penetration by doing more in that sector, I think it really isn't available to us even to consider that that's been the predominant piggyback market as far as I'm concerned. So that hasn't, to date, been even available to us regardless of our underwriting guidelines. And I think for many of the borrowers we serve on affordable housing levels, it's not an instrument being offered anyway.

Would it be fair to say then that your ARM exposure probably overstates your actual credit exposure because you're not in some of these riskier products, and if you are, your exposure is after five years, really?

Yes, that's a fair statement.

Okay. Thanks.

Brad Ball, Prudential.

Thanks. The question is on the average premium rate, which is flattish, and I think your guidance is for it to be flattish going forward. Despite the fact that you had a pick up in bulk NIW in the quarter and it sounds like bulk NIW may be running at a higher level now that you've got this new customer back. Should we expect to see the earned premium rates pick up later like into '06, or is the flattish average earned premium rate reflective of the -- the still pretty competitive pricing for bulk?

It's reflective of the competitive pricing. And a customer returned to us. I wouldn't say that that's going to happen each quarter, but there are executions available that spiked the volume for this quarter that may not be available in the third and fourth quarter.

We should think of the customer returning to you as a one-off transaction as opposed to --?

Yes, and that's something -- exactly. There's two -- two camps of bulk customers, as I said earlier. Those that come to us routinely and those that come to us on a spot basis. The one in the second quarter causing a pretty good jump in the volume is in the spot basis.

Okay. And just again, on the average earned premium rate, we think of bulk as generally driving higher earned premiums. But your comment earlier about the pricing environment is keeping that from moving higher as you increase bulk as a proportion of total business?

Right now, I would say that's correct.

Okay. Thanks. And then one final question. On -- you mentioned the single file business and this increased success you're showing there. Are you still seeing this sort of -- the same level of competitiveness coming from the banks in the piggyback loan area? What portion of mortgages do you think are going to the piggybacks with the loss of market share today?

Yeah. I'd say it's still in the neighborhood of 40 to 50% of the market is lost to piggybacks, particularly in the higher-cost areas, but -- so, yes. It's -- we haven't seen any pullback within that sector, Now, they've been giving guidance, I know, and maybe there will be a pullback at some point in time, but we haven't experienced it yet to date.

Okay. Great. Thanks very much.

You bet.

Bruce Harting, Lehman Brothers.

Mike, how much excess capital would you -- would you say you have, and how do you define that? Is it on a risk basis, or is it more of a balance sheet calculation? I'm just curious.

Well, it's theoretical because -- and I won't get into it. But if you look at the risk to capital ratios, we're running at 7 or 8 to 1, and you think about where you have to be to be AA. It's somewhere north of 15 to 18 or something, or maybe as high as 20 to 1. So -- so -- but all of that is changed and subject to the rating agencies and the newer capital models and their risk rating the business, et cetera, so I -- I mean, I wouldn't get into a discussion talking about how much excess capital we have other than to say that we're well capitalized and to the extent that we can upstream dividends for -- periodically, we'll continue to do that. So I can't hang a number on it, because that could change overnight.

And with regard to delinquency trends, are you -- I heard you earlier say delinquency's flat through the end of the year. Is that as far out as your guidance is going right now? And is that just for public disclosure that you only go out that far? Do you -- and I assume you have -- you have your own sense of where you are going to be in a year from now?

Well, we have a good sense where we think the books will be. I don't have a sense where the economy's going to be, though. That's the issue. I mean, as I pointed out and Larry talked about, too, we still have a significant portion of delinquencies in the Midwest, in the manufacturing states and we're still concerned about job growth in some of those states. And the fact is that delinquencies have remained flat. They haven't gotten significantly better. So that still continues to be on our check list, if you will, and to the extent the economy can improve in those areas, that would be positive. But -- so I think as far as we'll go, I think, looking out beyond the next six months is to say that based on what we know today, we're comfortable with the forecast being flat. If the economy continues to improve and if we can see some improvement in some of those Midwest states, we'd see that by year end, and that would be positive. But we haven't seen it yet.

And then -- thanks. And then the last question is, I'm not -- I don't completely understand where you and the other MIs fit into the -- this growing Wall Street securitization of mortgage trend. You talked earlier to the question about -- to Howard's question about the ARMs, and your ARMs, your percentage of risk in force. And insurance in force has stayed pretty flat, but ARMs went up considerably last year as a percentage of total U.S. originations. So does that -- should we assume you're avoiding most of the -- you did say you're avoiding a lot of the "exotic product," but then exactly how do you fit in and without naming the names of your specific clients on bulk, is it safe to say that it's the major Wall Street houses who are securitizing? And then -- and then where do you pick and choose your spots to play in bulk? Thanks. If not ARMs.

You know, I -- some of these exotic products you mentioned, the option ARMs and the interest-only, you see it in the paper a lot, but there's still a lot of the more traditional ARMs and fixed rate and all that being done. And that's the vast majority of the product. Some of these other things are just coming into market, but it gets a lot of press. And then on any bulk transaction, there's a good chunk of the transaction where we don't provide the MI. The MGIC is part of the credit enhancement on 80% of the loans, but there's maybe 5%, 10%, 20% that goes without, maybe because it's a very low LTV or maybe it's some of this exotic product. So we can -- we're going to be a big participant in the capital markets because the vast majority of the products are still the traditional type ARMs, not these exotic products. And maybe it grows, but right now it's -- it's not a big part of it, and if there is some of that, and we don't care for it, we can still participate in the transaction without necessarily doing some of that product.

Okay. Thanks.

[OPERATOR INSTRUCTIONS]. Paul Miller, FBR.

Thank you very much. I'm just wondering since you guys are releasing reserves, and I believe you paid losses -- if I'm incorrect, correct me -- it's about 150 -- it's about 150 million. When can we expect to see -- since you're releasing reserves, can we expect to see paid losses start to decline in the next three to four quarters after we go -- I mean, are we in a peak state? And then in '06 start to see these paid losses decline as we see your delinquencies decline?

For the quarter, Paul, second quarter paids were 158, and they were 149 in the first quarter. I think that right now we're forecasting the paids in the third and fourth quarter will be approximate the level of -- in the second quarter. Going into '06, again, we could -- we could start to flatten out. A lot will depend, of course, on what happens with the economy the next six to nine months. We'll have a pretty good feel on that as we get to year-end and look at delinquency levels. But our forecast right now would say that paid's pretty much flat to slightly up in '06, but subject to economic changes. As Larry talked about on the bulk side, the larger books have peaked and so that would trend to think that all things being equal if the economy continues at the current rate that we ought to be at a flatness, if you will, on paids. How fast they come down will be subject to development of delinquencies as well as economic improvement. I think we have to keep looking at it every quarter and report up.

And then, the other question is I know there's been a lot of discussion about exotic ARM products being out there. And the big exotic ARM product, I guess everybody's referring to is these pay option ARMs that is starting to -- a secondary market is starting to be developed very quickly, which really didn't exist before. We're already seen Fitch and S & P weigh in by increasing the equity levels of some of these -- on some of these securities. Are you guys big players in these pay option ARM securitizations? And what's your overall opinion of these pay option ARMS in general?

Once again, on the pay option ARMs in a portfolio bulk and flow together, we're probably under 1% of any and all kind of potential NegAMs, which this pay option ARM would be subset. So it's not a big deal to us as of yet. Like I said a couple of minutes ago, some of this product may be in some of the bulk transactions we see, but we may cover 80% of the loans and leave some of these products behind. So they're still being securitized in part of the transaction. It's just not part of the portfolio that we ensure.

And can you just give us any -- how do you guys view the pay option ARMs? I mean, do you think there's enough data out there to make value/risk award adjustments on it? Do you think the product's too new in the secondary markets?

Well, I guess ARMs in general over the last 15 years if you look at the interest rates, they've been pretty flat to down. So I'm not sure ARMs have really been tested in an upward rate environment. So I would say there's a lot to learn yet. Now, having said that, I think the option ARMs do provide some interesting possibilities, if you're in a bind, reducing or keeping the payment low level and racking up a little NegAM to get you through may not be a bad deal. I mean, if you can't make the payment, that's going to cause the delinquency. So that's the first step is the delinquency. So by having some of your options with regard to your payment, maybe it voids the delinquency at the expense of NegAM. So there's a potential positive tradeoff there. I think the thing that I see or worry about most off on some of these option ARMs is what do they start them out at? What do they qualify them at? And that's -- that's starting out at 1 or 2% and that's what they are qualifying them at. That then increases substantially the payment shock and inability of the borrower to weather any payment shock if they're qualifying them at a very low start rate. It's akin to the old teaser rates.

If the borrower, Paul, uses the option ARM to help them out of trouble, it's a good thing. The borrower that maintains the low payment just so they can keep it there and spend the money when they're not in trouble, that's a bad ultimate decision. So it all depends on how the borrower uses it, but clearly, if it's a case of a borrower that has that choice or going delinquent, it's a good thing.

When you guys -- I mean, have you guys been involved in underwriting pay option ARMs at this point? And if you do, do you qualify to fully indexed rate or do you qualify -- I mean, how do you view -- how do you qualify the borrower? If you do -- if you do get involved in the pay option ARMs underwriting at this point?

Well, usually a lender will submit a program to us and in the program description they'll indicate the start rate and qualification rates and so we can, depending upon FICO, LTVs and some of the other components of the program, say yea or nay to the program. So we, let's say in today's age here, we don't necessarily maybe underwrite each and every loan as year gone by, but we tend to review the programs and weigh all that in, in our approval or disapproval of the program.

It's part of the underwriting decision, along with the FICO score and debt-to-income ratios. It's not as easy as just saying we're going to do it at the -- at whatever the start rate may be or the fully indexed rate any longer. That was ten years ago. Today you consider so many different things because of the automated underwriting models that it's just not as simple an answer, Paul.

Thank you very much, gentlemen.

You bet.

Geoff Dunn, KBW.

Thanks. Not having much luck with questions. Mine was already asked. Thanks.

Okay, Geoff. Ask it again. See if we say it the same way.

A new pronunciation of that name, Geoff.

[OPERATOR INSTRUCTIONS]. Will Keaton, Galleon Group.

Hi. I was wondering, you said the pay option ARMs are not a big part yet. What -- how comfortable are you with the pay option ARM in terms of percent of NIW going forward? Like, to what level would you be comfortable with it going to? And then secondly on capital, you mentioned that perhaps you might be slowing buybacks. I was just wondering if you could give us indication of your preference for use of excess capital, whether it's buybacks, dividends or acquisitions?

Let me -- this is Mike Lauer. Let me ask the second question fist. And then Larry can talk about pay option ARMs. What I said was that -- that -- the question specifically earlier was will we maintain repurchase at this level. And what I said was it would be subject to our continued ability to upstream dividends. And in the second half of this year, we've got normal dividends of 44 a quarter. Two of those, and another 50 from this last dividend. That's the availability. And going into next year, the repurchase program, again, would be subject to our ability to upstream dividends to the writing company beyond the normal dividend of about -- let's say the normal dividend's about 200 million. Now, recall that out of that dividend, we need to pay common stock dividends and interest, et cetera. So repurchase activity will be subject to continued upstreaming of operating dividends from the writing company.

Regarding the option ARMs, at this point, I don't think we need to put a cap on it. One, it's not a big deal. Two, it depends on the program. The FICOs, the LTV's what you qualify them at. So you've got to weigh all of that. We do like to have dispersion -- geographic dispersion, product dispersion, customer dispersion, so -- but at this point, it's small and there's a lot of other factors to consider, and I guess we'll just take a wait and see.

Okay. Thank you. And then just in terms of capital returns. Could you rank for us your preference in terms of dividends, repurchases or acquisitions?

Well, I think now as evidenced by what we've done, share repurchase is very important to us. Raising the cash dividend is very important to us, as we've done the past year. And we continue to look at some acquisition opportunities. So -- but as demonstrated right now, share repurchase is the most prevalent use.

Okay. Thank you.

Andy Wagstaff [ph], Touchstone Investments.

Hi, guys, thanks. With regard to looking at your data, have you guys been able to discern any trends with loans that are written with non-income verification versus kind of full dock [ph] loans to see how the performance of those loans is relative to the rest?

Yes, we -- we have. Those whole days of generalization bases have been about four to five time higher claim rates relative to comparable credit scores with documented loans.

So four to five times higher.

Yes.

And as we kind of move into this next generation of products with the pay option ARMs and whatnot, I mean, would it be concerning to you if some of those loans are written with non-income verification, or is that really not -- not a big issue?

That's a big issue.

It is?

Yeah. I mean, a lot of changes are happening, I think in the all-day world relative to trying to get that information on income and debts soon after the fact, and different originators all have different experience within that world. And so our -- our experience to date, I wouldn't expect to be the experience going forward. I think it'll get better. But clearly, in looking at our loans we've insured to date, that's been the experience. And so if you add on another high-risk factor to that, that is not a combination that you'd like to do a lot of business in.

Right.

I think the market is really trying to grapple with affordability. Home prices have moved a long way --

Absolutely.

-- Relative to income. And the old rule of thumb is the money's got to come from somewhere. In the past, the money's come from interest rates coming down a couple percent over the years, but now we're flat to up on interest rates, so that can't be a help anymore. So they're -- the market is searching for ways to put people in homes and deal with that affordability. Interest only and I -- once again, I think that's a good tool if it's structured right. You're getting into these option ARMs and stated income, and they can be good tools. But once again, you've got to put them together correctly and -- and make sure's suited for the borrower. If you start mixing and matching these with the wrong borrower, they can be pretty detrimental.

Larry's right on. These are all niche programs, I think, that can't be generalized to the whole market. And we're on the Alt A stated income and the NINAs within that world. That clearly has a place in the marketplace. It clearly makes sense for a large number of borrowers. But makes no sense for other borrowers. And so to the extent you can utilize these programs in the right area, we applaud that.

Yes. And historically, when they were using these kind of products, were they doing -- were they writing these loans with non-income verification or is this kind of a new phenomenon?

Well, the non-income verification has been around a long time. It's been predominantly done for people that were self-employed and just didn't want to go through the process of showing that tax returns. It wasn't generally done for wage earners that could show you their pay stub. And so it was a program that's been around a long time and it makes a great deal of sense in that sector. It makes a great deal of sense from the aspect of a borrower where you can look at their job and say, this is a reasonable income, you don't have to show a pay stub relative to what profession they may be in. So they are programs that have been around for quite sometime, it's just that they've accelerated into the more generalized market where it may not always make sense.

Have you seen any heightened levels of fraud in those loans in some of the -- in some of the high end markets, California and Florida and whatnot that are -- that are different from what you might see in other states?

I don't think we've seen a state differentiation.

Okay.

Relative to fraud. No.

Got it. Okay. Hey, thanks for your time, guys.

Yes.

Operator?

I'm showing no further questions at this time. You may proceed with closing statements.

Okay. Well, thank you. As always, thank you for your interest in our Company. Have a good day.

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