Welcome to the Arch Capital Group Ltd. Third Earning -- Third Quarter Earnings Call. My name is Richard, and I will be your operator for today's call. [Operator Instructions] Please note that this conference is being recorded. Before the company gets started with its update, management wants first to remind everyone that certain statements in today's press release and discussed on this call may constitute forward-looking statements under the federal securities laws. These statements are based upon the management's current assessment and assumptions and are subject to a manner -- a number of risks and uncertainties. Consequently, actual results may differ materially from those expressed or implied. For more information on the risks and other factors that may affect future performance, investors should review periodic reports that are filed by the company with the SEC from time to time. Additionally, certain statements contained in the call that are not based on historical fact are forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. The company intends the forward-looking statements in this call to be subject to the Safe Harbor created thereby. Management also will make reference to some non-GAAP measures of financial performance. The reconciliation to GAAP and definition of operating income can be found in the company's current report on Form 8-K furnished at the SEC yesterday, which contains the company's earnings press release and is available on the company's website. I will now turn the call over to Mr. Dinos Iordanou and Mr. Mark Lyons. Gentlemen, you may begin.

Thank you, Richard. Good morning, everyone, and thank you for joining us today. I'm sure you had a tough time getting in to where you are if you're in the northeast, but it's the same for all of us. We had a good third quarter, mainly aided by light cat activity and a low level of large losses. Unfortunately, Sandy has reminded us that storm activity can and will happen at any time during the hurricane season. On an operating basis, we earned $0.87 per share, which, on an annualized basis, represents a 10% return on equity. Our reported underwriting performance was very good, with a combined ratio of 90.2%, while our investment performance for the quarter, including the effects of foreign exchange, was excellent, with a total return of 245 basis points for the quarter. Cash flow for the quarter was strong at $335 million, an increase of $25 million over the year-ago quarter and $83 million over the second quarter of this year. Our book value per common share increased to $36.79, a 6.8% increase from June 30, 2012, and a 19% increase from a year ago. We also had very strong revenue increases for the quarter, which I will comment on in a few minutes. The market environment continues on a path of improvement across the board. Premium rate changes for our book of business ranged from essentially flat in some sectors to as high as 17% with long-tail specialty lines receiving the highest increases. As we mentioned in our last quarter call, 2 other trends of note are emerging in the U.S. Insurance Group. Our E&S submission activity continues to increase year-over-year, a continuing sign that standard line underwriters are returning more business to the E&S market, where in our view it belonged in the first…

Thank you, Dinos, and good morning, all. On a consolidated basis, the ratio of net premium to gross premium in the quarter was 80.6%, essentially unchanged from 80.4% a year ago. In the reinsurance segment, the net-to-gross ratio was 97% in the 2012 second -- third quarter compared to 96% a year ago. And the insurance segment maintained an approximate 74% net-to-gross ratio. This 74% net-to-gross level in the Insurance Group is a result of our planned strategy to over the last 5 years grow the lesser volatile, smaller account businesses and significantly reduce our exposure in the higher severity businesses. The consolidated combined ratio for this quarter was 90.2%, with 3.7 points of current accident year cat-related events, net of reinsurance or reinstatement premiums, compared to the 2011 third quarter combined ratio of 94.4%, which reflected 8.7 points of cat-related events, also net of reinsurance or reinstatement premium. The 2012 third quarter reflected 2.3 loss ratio points from Hurricane Isaac, with the balance emanating from the other 2012 cat events, such as U.S. tornadoes and hail storms that occurred during the first half of this year. Additionally, crop business for the 2012 year, which Arch does not classify as cat, was booked to a 300% loss ratio against year-to-date premium, and this represents approximately $0.045 per share. The 2012 third quarter combined ratio also reflected 7.1 points of prior-year net favorable development, which also is net of reinsurance or reinstatement premiums and acquisition-related expense, compared to 8.5 points of prior-period favorable development in the 2011 third quarter. Approximately 75% of the net favorable development in the 2012 third quarter was from the reinsurance segment, with approximately 60% of that due to favorable development on short-tailed lines from more recent underwriting years, and 40% due to favorable development on medium-…

[Operator Instructions] Our first question comes from Amit Kumar from Macquarie.

The first question relates to Hurricane Sandy and I know it's too early. If you were to go back and look at Hurricane Irene and then look at Hurricane Sandy, I would be curious to know what were the sources of losses for Irene and now when you look at Sandy, how would they differ in your view?

Well, there is a lot of similarities, but also there is a lot of differences. Too early, so take my comments based on what we know today and things will change. Both events' similarities -- both events, at least according to the New Jersey, New York and Connecticut insurance departments, they're going to be classified as tropical storms. So in essence, it will have an effect not dissimilar to Irene. The hurricane deductibles might not apply so the normal deductibles will apply. That, we model a bit and it gives you basically for commercial -- for a book of business that is half commercial, half personal lines, about a 30% increase in the ground up loss; for a personal lines on the book, it's probably a 50% increase. So that's the similarity on the factor. Where Irene did not have significant storm surge, this event, Sandy, had significant storm surge. And even though commercial policies, probably we have less disputes as to was it wind-driven rain and wind damage versus flood, because most commercial policies deal with the coverage of flood either by explicitly excluding it or including it with a sub limit. The homeowners policies exclude it, but then you have some dispute that are going to emerge from that. So it's -- that it will be different than Irene on the basis that up and down the coast, even downtown Manhattan, the South Shore of Long Island is a significant amount of damage that is from storm surge, which should not be covered by policies unless they bought flood insurance. Only time will tell us as to how that will get interpreted, and what losses might be put on the backs of the insurance enterprises. So that's what we know today. It's too early. This is a complicated storm with complicated issues. And probably we won't know for quite a bit of time. The one area that it's going to be totally different than Irene -- Irene did not have significant business interruption losses coming out of it. This storm will cause a lot of business interruption. The whole grid system is being devastated. Almost every small business owner is out of business. Small commercial policies don't have big time deductibles on the business interruption. They have an amount -- those amounts are going to be breached, and the business interruption can be significant, so -- but way too early for any one of us. I'm not smart enough to know as to how big this is. The only modeling we have done is what I mentioned to you that with or without the deductible, I think the loss balloons by 30% on commercial policies and about 50% on personal lines.

Got it. And do you think the market is underestimating the demand surge component of the loss?

I don't know if they are -- nobody has come up with any estimates. Even the modeling agency, the modeling -- I mean, the latest we had, it was actually accu cut [ph] -- and it's a wide margin, $10 billion to $20 billion. I mean, $10 billion to $20 billion is 2 acts. I mean, that's not a pretty accurate prediction. I would -- if you have to press me, and this is based on what I've seen, no specific information that we have, I will go to the higher end rather than the lower end.

Our next question comes from Mike Zaremski from Crédit Suisse. Michael Zaremski - Crédit Suisse AG, Research Division: So, Dinos, you expect to buy back stock in 4Q from the prepared remarks, but it's also way too soon to assess the Sandy losses. So correct me if I'm wrong, I believe Arch's goal has been to never have more than 1.5% market share of a loss. So if you're planning on buying back stock, is there a readthrough there on how you're thinking about the Sandy loss?

All I can tell you is that independent if Sandy is $10 million, $20 billion or $30 billion or $40 billion, we have excess capital, so that would won't affect the decisions. Michael Zaremski - Crédit Suisse AG, Research Division: Okay. Got it. And we can take this off line, but I was -- why would -- it being classified as a storm and not a hurricane, why would that cause commercial and personal lines losses to be up 30% and 50%, respectively? I thought the delta between the deductibles was only like a few thousand dollars?

Not on the high-end homeowners policies. Usually, they might have a 1% deductible in the northeast for a hurricane. On a $2 million, $1 million home, even 1% it might be $10,000 or $20,000, but far significant that the $1,500 or $500 or $1,000 deductible that they want to be without this. So it's significant. Now, if you walked around in your neighborhood, you will see that this storm has caused a lot of damage, mainly to trees and shrubs and some outer buildings or maybe a tree fall on the corner of the roof, and it can be repaired for $10,000, $15,000. If it was a hurricane loss, that might be totally within the deductible. By being classified a storm, that might be a $10,000 to $12,000 to $15,000 recovery from the insurance company. When you aggregate that over factors and some risks, it becomes significant. That's why when we model it and we run this -- we run a data set of personal lines exposures, with or without that parameter, and we saw that it was 50% more.

And Mike, if I may just add one thing to Dinos' comment. Is that to an assumed cat book, once the seeding company hits its attachment point, the losses are then disproportionately shared. So every buck that is contributed through all these aggregations Dinos was talking about gets shipped upstairs to the cat carrier. Michael Zaremski - Crédit Suisse AG, Research Division: Okay. That's very helpful. So lastly, totally unrelated. I believe Arch has a "ventures group", which recently added someone from a competitor, which is a leader in the joint ventures business. Can you talk about your plans there?

I don't know if we have a ventures group. I mean, that's news to me. But -- so I don't -- I mean, we have individuals who are looking and they have the assignment to look for both on the investment side, alternative investments and on the business side to look at opportunities for us to do either side cars or alternative structures. But if that's what you're referring to, yes, we have both, but it's no different as to how we approach a business over the last 5 to 10 years. And now we're just kind of having specific assignments to individuals. This way, we get better control as to how we do it.

Our next question comes from Jay Gelb from Barclays.

Do you think the impact of Sandy might influence the discussions of the January 1 renewals season? Whereas previously, we were hearing property cat rates could be flattish as of January 1, do you think this could influence both the supply and demand functions?

I'm going to go against my mother's wishes not to make guesses. But I think depending on the size, and I'm more of the pessimist here, as I mentioned, that I think this is a bigger loss to the Insurance Group than some people estimate. It should have an effect on how people think about the cat risk and how they price it. Yes, there is new fresh capacity coming in from hedge funds and/or private equity funds that they view the cat area as a non-correlated area for them to put some money to work. And of course, supply of capital always puts pressure on rates. But I think this might waken up some people and say, "Hey, maybe what we're charging is -- needs to continue to be charged because we do have these kind of catastrophic events." And if I have to guess today, I would tell you that if there was significant pressure for rates to go down, probably that pressure has been relieved a bit, and there might be a possibility, especially on individual programs that they're going to get hit, that you might get even some rate increases.

Okay. Can you discuss Arch's inbound reinsurance program in terms of protection against Sandy loss?

Yes, Mark, why don't you -- yes, for the Arch Insurance Group, as to what our attachment point is...

Yes, we have a -- we attach at $100 million and we have coverage up to basically $350 million. And we always view our purchases as balance sheet protection, not income statement protection. So the higher up it goes in the program, is the higher the balance seeded away from it to virtually all seeded at the top. So we feel we're pretty well protected. However, we have to pay attention to the enuring [ph] covers that are there, and there's material cat absorption within the quoted shares that we have protecting that book.

The first number he gave you is just purely the cat book, but there is a lot of cat protection within our specific treaties protecting our different books of business on the property. We take most of our cat risk on the reinsurance side. We -- as a matter of fact, I think I mentioned this before in other calls, when we allocated PMLs, we allocate approximately $5 of PML to reinsurance versus $1 to insurance. On the insurance side, so we try to buy as much protection as we can. So we don't allow the insurance group to cycle over the 1-in-250 event. So we buy based on our modeling enough coverage to protect us for the 1-in-250, and we assume the deductible at the bottom.

And clarifying comments, Dinos was making the contrast before between commercial and personal life exposures. The insurance group doesn't have any personal line exposures, firstly; and secondly, just as a clarification, Dinos' comments before -- he'd mentioned that he thought things might be underestimated from the Insurance Group as opposed to the intent of the insurance industry.

Of course. Okay. And then finally, in the reinsurance segment, we still see large growth coming out of the new programs. How many more quarters should that persist before it normalizes on the growth rate?

Well, it depends what happens on renewals, right? I -- these are lumpy deals. If we continue to have the ability to continue to renew them, they will still be there. But I can't predict the future, that's why we're not in the business of guidance and we're not projecting revenue. We're not a company who runs on revenue. The only thing I hold people responsible is the deals they write, that they have some positive return. So we're bottom line oriented, not top line oriented. We will continue to be as such. When the growth comes, it should come with a profit, and if it doesn't, so be it.

Of course. I was just looking to isolate when those programs came on so that we can think about the rate of change in premium growth year-over-year. Did most of these new programs come on in 3Q...

We -- it came -- the mortgage insurance started in the -- exactly a year ago. It was the last quarter of a year ago and it continues. And of course, it goes into the future. And the motor, it was the beginning of the year.

First quarter of '12.

Yes.

Our next question comes from Vinay Misquith from Evercore Partners.

The first question is on PMLs for the Northeast. You've given us the 1-in-250, which is great, but just hoping that you could give us maybe the 1-in-100 or 1-in-50 PML for the Northeast, too, please?

Yes, I think we have the 1-in-100. Give me a minute or 2 and I'll dig it out.

Sure. Just a follow-up on that. How do you see this loss sort of in terms of PML? Is it more like a 1-in-50, 1-in-30? How does it look to you?

Well, from which perspective? From size of loans or frequency of event?

From the PML perspective, when you do your forecasting for what the PML should be...

Believe me, [indiscernible] is not a 1-in-100, it's probably a 1-in-30 or 40 or thereabouts. Of course, some people, they say 1938 was the last time we had something of that sort, but things are different. We had Irene, and Irene was a smaller storm, but it was only a year ago. So -- now, you said -- the 1-in-100 PML for us for the northeast is $525 million compared to the $857 million that it was for the 1-in-250.

Sure. And do you happen to have the 1-in-50 by any chance or is that asking for too much?

I don't have it in front of me. We can run it for you, but usually, we don't spend a lot of time worrying about the 1-in-50 or 1-in-100. We spend a lot of time worrying about a 1-in-250 because we want to make sure, and as I said in many other calls, independent of what event we have, the 1-in-250 event, we want to survive it and emerge on the other side with unimpaired capital and ability usually to grow the company because the opportunities, once you have a 1-in-250 event, they're going to be enormous for those that are still standing. And I don't want to be just standing, I want to be able to be running at the same time.

Right, right, fair enough. And so I'm just thinking about the 1-in-100 versus the 1-in-50, and I'm sorry for just belaboring the same point. Just -- would there be a significant sort of stepdown from a 1-in-50 to 1-in-100 because a lot of the losses really are on the higher layers of programs?

You got to understand where we play on the curve. We don't like to be on the 2 ends of the curve. We don't like to be exchanging dollars and be way down low, so small storms are not going to give us a lot of losses. And we don't want to be way up above on the top, like the $40 billion and above, because we don't believe you get paid enough even for the cat risk to take those kind of bets. So usually for us, it's midsize storms. So if you're talking about a 1-in-10, 1-in-20, which is the bottom end, our numbers are going to come way down, way down.

And one other clarification, Vinay. To augment what Dinos said about where we play, it's the extent to which we play. So as you go from 50 to 100 to 250, whatever on the curve, its censoring effect of policy limits and cat limits put out come into play so it can't grow beyond that amount.

For example, if I gave the 1-in-1,000, which we do, it's not going to be a significantly higher number than the 1-in-250. Because as I'm telling you, I take most of my cat risk on the reinsurance side, which I'm controlling aggregation of limits. So in essence, at some point in time, the world can go to oblivion, and it is what it is. And we want to survive those kind of events because one thing we will never do is bet the company, independent of how big the event is.

Fair enough. And on the reinsurance side, are your exposures half and half between personal and commercial or is it more weighted ?

I don't know if it's half and half. I -- we don't view it as such, so I don't have a number. But in knowing the accounts that we're on, it's hard to tell because a lot of the companies that we participate on, they have both commercial books and personal lines books. And without having the files in front of me, I don't know if their commercial book is bigger than their personal line. But I would say, probably our book is probably half and half from an exposure point of view. Some clients, they will have a bigger commercial portfolio than personal line, and some, they're going to be vice versa, but we have many clients.

All right, fair enough. And you guys are also -- I mean, they're reasonably big fac writers, so do you have some extra reinsurance protecting that book or...

Our fac book is not a cat book. We take some risk on the fac book mostly on flood limits, specific flood limits here and there. But it's not a book that is a traditional cat-exposed book. It's more of a fire and other perils-exposed book, and that's what our facultative teams do.

And in terms of analyzing it, it's really a sister process to the Insurance Group. You've got deal-by-deal information and much more granular in your ability to analyze it.

Our next question comes from Greg Locraft from Morgan Stanley.

I have 2 sort of follow-ups from items. First on the deductibles issue, it's obviously a big reinsurance -- it's going to hit the reinsurers hardest as a segment. Who determines what the storm actually was when it hit shore? Is it when the eye wall passed over, if there was an eye wall? Is it wind speeds, is it -- I mean, what defines it and whose the arbiter there?

Well, it's wind speeds and you have Noah who is going to give different measures. And unfortunately, you have also the regulatory process at the end of the -- listen, if you don't like what the regulator says, you can always end up litigating it, but I'm not going to make that determination. I'm not in the personal lines world, et cetera. Listening to some of the calls, I think already a major personal lines company says they're going to abide by what the insurance department said. Once they do that, you got this follow-the-fortune clauses in the reinsurance contracts so the reinsurers are going to behave in the same manner because they have no way out of it.

Our next question comes from Matthew Heimermann from JPMorgan. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: I guess, one question I had was -- and this just gets to exposure. When you talk about the reinsurance side, how does the facultative book that you have kind of change the exposure in this type of event relative to kind of a more traditional wind event that has a higher personal lines component to it?

I'm not totally understanding your question. How does -- what the facultative book changes which parameter? Hello?

I'm sorry, his line has dropped. [Operator Instructions] Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: I don't know what happened. But sorry for asking the question poorly. You have your direct and facultative business within the reinsurance business. I would assume that, that's virtually all -- that's all commercial business. And so I'd just be curious when we think about your performance in past events, if that facultative book potentially means you get -- you maybe have a different performance, maybe a little higher scare of losses than you would in kind of a traditional wind event, which is where there's obviously a bigger personal lines component as a percent of the total. And then just be curious if that's true, are there other offsets when you think about your portfolio in total?

It's -- on the fac book, as I said, is on the reinsurance side. On the reinsurance side, we do both treaty and fac. Our fac book is not a cat book. So in essence, a lot of what they write is excess limits on fire policies, not including the cat component. So they don't -- they're not searching to write a lot of exposures that give us a cat loss, flat or wind kind of losses. They're more of a fire book, and I think of it as an excess fire book. Occasionally, we do write some accounts and they're all commercial, that they will have a cat component to the coverage. But it's very, very small. Just to give you an example as to how small it can be, when we allocate PMLs to the different units, right, the nationwide PML allocation for our fac book is about $75 million. So it's not a cat-driven book of business. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: That's helpful color. I guess, the other -- just a numbers question for Mark, was the -- if I'm doing the math right, then that would imply there's about a $6.5 million pretax impact from crops if we want to think about the net dollar impact.

Matt, you hit it exactly. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: Okay, all right. Good to know all the risk from time to time.

Listen, we booked it at maybe a higher combined ratio than most. But to us, it's such a small piece. I'd rather get it right than get it wrong. So nobody is going to stick their hand in my pocket and take the money out, unless I have to pay it. So if I have it out that is good; if I don't, I can always return back to shareholders.

Just to clarify, we say the $6.5 million, that's ultimate loss not the amount of the loss in excess of the premium. I mean, that would be $4 million plus. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: Yes. That makes sense. I was just trying -- I was just trying insulating the per-share comment you made. Then I guess the last one is just in terms of business. I think you have a slightly, maybe not surprisingly either, but conservative view of kind of where returns in the market are and might be going because of rates. But if you do get more optimistic, and if rates look -- move up and you get more optimistic, what types of business mix shifts could we expect to see kind of on a product basis within your insurance and reinsurance portfolio?

Well, let me give you a couple of comments on that and then I'll turn it over to Mark because before becoming CFO, he ran the Insurance Group, too. So he was seeing the information on a granular basis. One area we have shied away from consistently now, both on the insurance and reinsurance side, is what I would call excess, like umbrella business, specialty casualty business. This is an excess workers comp which has shrunk significantly, workers comp on a guaranteed cost basis, et cetera. Now if you interpret my comments, my prepared comments, we still don't like those segments even though those are the segments that they're getting the most significant rate increases. Because our view from an absolute point of view and where new money rates are on the investment side, doesn't give us the economics that we expect from that business. But if we get another year of rate on rate, I believe that those are going to be getting above the thresholds. And as we've done in the past, don't forget, we had a lot of casualty business in '03, '04, '05, '06, we're not afraid to step on the accelerator at that time, but we have to make that determination that on an absolute basis, based on the current investment environment for these long-tail lines of business, we're getting the right economic return. Mark, do you want to?

Yes, Dinos' right on point. I'd think the key thing to reemphasize is the way we look at things. And we look at things first and foremost on an absolute basis, not a relative basis. So the fact that rates in some of the lines that Dinos' mentioned, say excess casualty, are getting double-digit increases and they've gotten increases in excess to loss trend probably for the last 3 quarters. But given our view, on a relative basis, that sounds great. On an absolute basis, we still think it has a long way to go. When you move to the other example of excess comp, because the duration of that is so much longer, it's about 16 years compared to the excess cash being about 7 years, the leveraged effect of interest rate movements is gargantuan on that line. And interest rates will be the primary driver of turning that to a green light line of business. So the answer is a line-of-business by line-of-business, sometimes class-of-business decision process where we open the -- I don't believe I'll say open the floodgate given Sandy. But if those decisions will be made at that level, but driven by our view of the absolute returns.

Our next question online comes from Ian Gutterman from Adage Capital.

First, just one last thing on the fac book, and then I just have a question on PMLs in general. Like you said, you obviously don't think of that fac book as a cat book, but always the consequences of sort of these unexpected cats is we get non-modeled cat losses, right? Things that we didn't think were cat ended up being cat. Is there anything you see, though, about Wall Street-type risks or things that are maybe Central New Jersey that you used to not think of as being in a cat zone, or Pennsylvania that maybe could estimate fac exposure?

We looked at our books, yes, we have some. We might have excess limits on some restaurants. We might have that potentially there is a coverage for business interruption emanating from even a flood event or -- but nothing that makes me worry about it, knowing what our book of business is. You see it's going to be -- the business interruption issue is going to be more complicated than some people think because is it flood that caused Con Edison to go out or is it wind that caused them to go out and not supply certain areas? And is that restaurant now having contingent business interruption, claiming it on his policy on the basis that he can't get power to run the restaurant? There's going to be -- you've got to look at the contracts, you've got to look at interpretations and that's why I think we'll be talking about the size of this Sandy loss 5 years from today. There are going to be disputes, there are going to be interpretations that need to happen.

And Ian, one other piece of color on that, and Dinos has kind of hit on it earlier. We can't estimate the size of this loss. I mean, the damage ability or the damage assessment is still in process, let alone dollars that will be -- start to be attached to it. But one comment we can make is that in a general way, there's a difference between subject books of business. So whether it's wholesale-driven and in the surplus lines of market or whether it's retail-driven businesses, there is a difference on how it's approached. Most E&S business and property exposures explicitly have storm surge excluded from their definition of wind. And if you want that to be included, it's endorsed on specifically as a flood cover. On some of the retail businesses, because they're good at their jobs, over time, especially in the Northeast since that was viewed as a fire PML and not really leading as cat, but a fire PML basis, terms started to slack off over time and what crept in was the definition of storm surge within the wind definition itself, not to have an explicit flood cover to have it done. So those kinds of morphs will play out. Like Dinos said, "Until you have a contract in front of you, it's very difficult to gauge."

Ian, I can't take the underwriting out of him. I made him CFO, but he still goes back to underwriting. I don't know.

That's good. I like it. Just to follow up on that point, so let's say it ends up being -- power loss being covered as for BI. Does the coverage period end when the power comes back on? Or is -- if the power comes back on but the person can't get up to speed because they have to clean up their flood damage, which wasn't covered under BI, are they still covered the whole time they're out or until only the power comes back on?

No, no, they're covered until they go back on business. So usually, let's say they need an inspection by the food -- by the Public Safety Department, usually, I'm using the restaurant example, you can't just turn on the lights and say I'm open. You've got to call the city and they come and they inspect and then they give you the green light, and it says you can reopen, that's the day you have reopen, that's the day that determines that you're back in business.

Okay. And then just on the PMLs. Northeast has always been a big zone for you, close to your peak or at your peak at times. Can you just talk about sort of what you've liked about the Northeast as a zone? I mean, you seem to have a higher exposure to it than other companies. And I know you said there's a mix of personal and commercial, but anything else you can you help us understand as far as how to think about your exposure in that zone?

No, no. The Northeast I don't think is -- a Gulf is even bigger than the northeast for us. It's just the density of the population and the kind of clients we have, which we write these nationwide contracts. And as they have a higher exposure in the northeast by taking the nationwide program that we make the evaluation as to should we write this client or not based on their nationwide exposure, it might give us a bit more in the Northeast than it might give us in Florida, for example. Because a lot of these nationwide clients have tried to manage and reduce or even eliminate exposure down south, southeast, because they don't -- the frequency of events down there is a problem for them. So -- but we're not uncomfortable with what we have in the Northeast and it wasn't a determination by us that we need to have more in the Northeast. It was more the outcome of evaluating nationwide contracts for nationwide clients and because where the values are, we got a little bit more than we have in Florida.

Okay, got it. I just wasn't sure if you were seeking out sort of a regional-type cover up there or individual risks or stuff like that. It's more of the national kind of covers, then.

Our next question online comes from Jay Cohen from Bank of America Merrill Lynch.

Question -- just a follow-up on an earlier comment you made, Dinos, about the ROE that you kind of -- the underwriting-year ROE you think you're operating that at, 9% to 11%, is that using all your capital? Because you also said you have excess capital. Do you take that out of the equation when you get to that ROE calculation?

It's using the allocated capital. So with this -- the lion's share of it, I mean -- all right, don't forget, it's not rating agency capital, it's also the cushion, we allocate that to the units. So you might take a few hundred million above it and deduct it, but basically, it's an allocated capital that we calculate that.

Our next question comes from Ryan Byrnes from Langen McAlenney.

Just one quick one. I may have missed it , but can you give your overall rate increases on the insurance side? And then quickly, just to get your thoughts to see if -- what your thoughts are, I guess, insurance rates, I guess, post-Sandy? Will that, I guess, materially accelerate increases?

Well, Ryan, for the Insurance Group, it was roughly 4% across-the-board. We kind of alluded to it in our comment. 4% does not mean 4% across-the-board. There's pretty strong variance. Dinos already mentioned that we're anywhere from 0% to 17% in the quarter, depending upon the line and the pocket of business. I think what usually happens, but I can't -- I don't have a perfectly clear crystal ball, is that the property cat rates will move and other things may or may not languish. It's -- it depends what the ultimate cost of this is to the industry and whether or not it's an earnings event or capital event that emerged out of it.

The only color I will say, that's the rates on the Insurance Group for our books of business. On the reinsurance side, especially on some of the new lines that the rate increases are much more significant, especially in the mortgage insurance space. So we don't -- because we make underwriting decisions on contract-by-contract, et cetera, we don't try to pay way too much attention to just rate increases. We're looking at the absolute level of the rate we're charging for a particular book of business and do we believe that gives us an adequate return or not. And that's where we aggregate those calculations to get to the decision point. And that's why I gave you a range, 9% to 11%, because I don't know how my mix might or might not move over the next quarter or 2 quarters, et cetera, because we have geared our underwriting teams to go where they believe they can get an adequate return. So they'd be pushing to the double digit, but I got to hedge a little bit, I'm the CEO, so.

At this time, I show no further questions. Mr. Iordanou, Mr. Lyons, do you have any closing remarks?

No. Thanks, Richard. Thanks, everybody, for attending. We'll see you next quarter.

Thank you, ladies and gentlemen. This concludes today's conference. Thank you for participating. You may now disconnect.