Arch Capital Group Ltd (ACGL) 2012 Q4 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the fourth quarter 2012 Arch Capital Group earnings conference call. My name is Keith and I will be your operator for today. At this time all participants are in a listen-only mode. Later on we will conduct a question-and-answer session.

(Operator Instructions)

As a reminder today's conference is being recorded for replay purposes.

Before the Company gets started with its update, management wants to first 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 management's current assessment and assumptions and are subject to 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 facts are forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. The Company intends the forward-looking statements in the call to be subject to the safe harbor created thereby. Management will also 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 to the SEC yesterday, which contains the Company's earnings press release and is available on the Company's website.

And with that I would now like to turn the conference over to your hosts, Mr. Dinos Iordanou and Mr. Mark Lyons. Please go, gentlemen.

Thank you, Keith. Good morning, everyone, and thank you for joining us today. The fourth quarter from an earnings point of view was negatively impacted by the catastrophic activity of Superstorm Sandy for which we recorded approximately $200 million of net losses. On a more positive note, as we announced last week, we have entered into is a definitive agreement to purchase CMGMI and the operating assets of PMI. We will comment on this transaction and on Sandy later in this call. On an operating basis, we lost $0.18 per share for the quarter, bringing our annual operating earnings per share to $2.54 for the year, which produced a 7.7% return on equity. Reported underwriting results were significantly affected by Sandy as we recorded a combined ratio of 112.4%, with Sandy contributing 26 points to that combined ratio, while our investment performance for the quarter, including the effects of foreign exchange, was a total return of 80 basis points. For the year, our total return on investments inclusive of foreign exchange was excellent at 5.88%. Our cash flow for the quarter was good at $190 million, an increase of $80 million over the year-ago quarter.

Our book value per common share decreased slightly to $36.19, mainly as a result of the impact of Sandy and the temporary effects of share repurchases. On a year-over-year basis our book value per share increased by 14%, an excellent result. We also had very strong premium growth for the quarter, which I will comment in a few minutes. On the market, the market environment continues on a path of modest improvement across the board. We estimate that for our insurance business the rate increases in the quarter exceeded our estimated loss trend by approximately 230 basis points in the aggregate. These improvements on a line-by-line basis ranged from minus 100 basis points to plus 1,200 basis points. The movement of business from the admitted markets back to the EMS markets is continuing. This provides us with additional opportunities. However, on an absolute basis, most long tail casualty business, based on the current interest rates environment, still require more rate improvements to meet our return requirements. On a consolidated basis, gross premiums were up 16.3% and net premiums were up 20%.

Looking at growth by segment, the Insurance Group's premium rose by approximately 5.6% on a gross basis and 7.2% on a net basis. And similar to last quarter, slightly over 0.5 of the increase in gross return premium was due to increased rates and the balance was due to additional exposures. The Reinsurance Group premium volume was up 51.5% on a gross basis and 50.6% on a net basis. These increases primarily resulted from several opportunities that we capitalized on during the past three years. Namely, UK motor reinsurance and mortgage insurance, as we discussed last quarter, and increased activity in our property cat and property FAC operations. Superstorm Sandy had a localized effect on rates, primarily on loss-driven accounts in the Northeast, but had little other effect elsewhere around the country. Group-wide, on an expected basis, the return on equity on the business we wrote this year will produce an underwriting year return on equity in the range of 10% to 12%. The underwriting margin improvement that I mentioned earlier will not affect the expected return on equity as the improvement in expected underwriting results has been largely offset by the reduction in expected investment yield.

During the fourth quarter, we repurchased 3.9 million shares at a cost of $172 million. We remain in an excess capital position and we expect to still be in an excess capital position after the mortgage transaction closes. As always, our position is to return excess capital to shareholders unless we can deploy it effectively in our business. As a result, we expect to continue with our share repurchases at a level approximately equal to our quarterly earnings for the first half of the year and will follow the same approach to share repurchases in the coming quarters as we have done historically. Now let me update you on our cat PML aggregates. As of January 1, 2013, our largest 250-year PMLs for a single event were basically unchanged with $867 million in the Northeast or approximately 18% of shareholders' equity, $862 million in the Gulf where Florida Tri-County PML now stands at $602 million.

Before I turn it over to Mark for more commentary on our financial results, as I indicated earlier, I have a few additional comments on our recent announcement. During the quarter after an extensive due diligence process we entered into definitive agreements with Cuna Mutual and PMI for the purchase of CMGMI and the operating assets and infrastructure of PMI. The transaction is expected to close in the latter part of this year as it is subject to obtaining approvals from insurance regulators, the Arizona receivership court, and the GSEs, as well as customary closing conditions. This transaction will provide us with an approved operating platform and nationwide licenses that would allow us to participate in the mortgage insurance space on a direct basis.

This new operation will complement our existing reinsurance capabilities and our European-based mortgage capabilities. We expect to continue to support our value customers in the mortgage insurance space as a stable long-term reinsurer. And, of course, we would enter the mortgage insurance business on a direct basis in the United States following the closing of the transaction. We're looking forward to partnering with Cuna Mutual, who will continue to service their credit union customers on our behalf. We're also pleased with the opportunity to hire the existing employee work force of PMI. We intend to create a stable employment environment for them for many years to come. We believe that the combination of Arch executive management and PMI's senior management, together with a strong capital position and disciplined approach to underwriting, will create an enduring enterprise in the mortgage insurance space. Although the transaction is not expected to be accretive to earnings for the first two years subsequent to closing, we believe that the mortgage insurance business will be profitable over the long term and we expect that this transaction will meet or exceed our target return levels and will create long-term value to our shareholders.

With that I'll now turn it over to Mark to comment further on our financial results. Mark?

Great. Thank you, Dinos. Good morning, all. The consolidated combined ratio for this quarter was 112.4% with 25.8 points of current accident year cat-related events that have reinsurance and reinstatement premiums, compared to the 2011 fourth-quarter combined ratio of 89.7%, which reflected 10.5 points of cat-related events, also net of reinsurance and reinstatement premiums. The 2012 fourth quarter reflected 26.1 combined ratio points from Superstorm Sandy with the balance emanating from minor adjustments to other 2012 cat events, such as the first-quarter of 2012 US storms and first-quarter southeast Australian floods. Our best estimate of the Sandy loss is consistent with our December 19 press release and reflects an approximately 60%/40% mix for our reinsurance and insurance segments, respectively.

The 2012 fourth-quarter consolidated combined ratio also reflected 7 points of prior-year net favorable development, also net of reinsurance and reinstatement premiums and acquisition-related expenses, compared to 15 points of prior-year favorable developments in the 2011 fourth quarter. This resulted in 93.6% current accident year combined ratio excluding cats for the fourth quarter of 2012 compared to 94.4% in the fourth quarter of 2011. Over 90% of the net favorable development in this quarter was from the reinsurance segment, with approximately two-thirds of that due to favorable development on short tailed lines concentrated in the more recent underwriting years, roughly 8% from medium tailed lines spaced throughout many underwriting years, and about 25% due to favorable development on longer tailed lines, primarily from the 2004 to 2008 underwriting years. The remaining net favorable development comes from the insurance segment and was mainly driven by shorter tailed lines predominantly from the more recent accident years.

Similar to prior periods, approximately 69% of our total net reserves for loss and loss adjustment expense or $7.1 billion are IBNR or Additional Case Reserves, which is a fairly consistent ratio across both the reinsurance and insurance segments. On a consolidated basis, the 2012 calendar year produced a 95.4% combined ratio on a reported basis compared to 98.3% for 2011, representing $99 million improvement in underwriting income and reflecting a lower level of catastrophic activity than in 2011. The 2012 expense ratio improvement reflects in part the expense control initiatives instituted over the last few years, mostly in the 50 bps improvement in the operating expense component. The full accident year 2012 combined ratio excluding cap was 94% compared to the full accident year of 2011 combined ratio of 93.6%.

Overall, on a consolidated basis, the full 2002 year saw $433 million of gross written premium growth or 12.6% and $375 million or 14.2% on a net basis. About two-thirds of this growth in gross written premiums came from the reinsurance segment and one-third from the insurance segment. On a consolidated bases the ratio of net premium to gross premium in the quarter was 75.3% versus 73.1% a year ago. In the reinsurance segment the net to gross ratio was approximately equal to one year ago at 92%, and the insurance segment was a 68% ratio compared to 67% a year ago as a result of our strategy to grow the lesser volatile smaller account businesses and reduce our exposure in the higher severity businesses.

In the reinsurance segment, the 2012 accident quarter combined ratio excluding caps was 83.9% compared to 83% even in the 2011 fourth quarter. The reinsurance segment results this quarter reflect, as Dinos mentioned, changes in the mix of business on a written basis with a higher contribution from mortgage reinsurance lines, UK motor and property insured tailed businesses than in the fourth quarter of 2011. Additionally, this quarter saw some escalation in recorded reinsurance crop losses for accident year 2012, which the company does not classify as cat losses, based on additional reports from cedent plus our own additional case reserves. Lastly, this quarter had 1.9 points of bonus accrual adjustments stemming from favorable prior-year development. The restated accident quarter combined ratio without this effect was 82%.

The 2012 fourth quarter results for the Reinsurance Group, as has been true in the last two quarters, includes the effects of the April 2012 acquisition of the international credit and surety operations of Ariel RE based in Zurich, Switzerland. This acquisition accounted -- under purchase accounting rules contained an [unearned] premium reserve that was not included in the premiums written but is reflected in premiums earned on an ongoing basis. Net premiums earned for this quarter included approximately $17 million from this acquisition, with the remaining December 31 on a premium reserve of approximately $34 million to be about 50% earned over the next two quarters and the balance substantially earned within one year after that. In the insurance segment, the 2012 accident quarter combined ratio excluding caps was 100.4% compared to 100.9% a year ago. The fourth quarter of 2012, however, contained one expense ratio point of bonus accrual adjustments as a result of prior-period favorable reserve development.

The accident quarter combined ratio is 99.4% when this effect is removed. The insurance segment had net premium growth predominantly emanating from the US operations. This US growth came from programs, executive assurance, and lender product units with a continued reduction in US casualty lines and declines in the tech risk line across all geographies. The program's unit saw growth primarily from two large existing programs along with growth from a newer program which we regained back in the first quarter of 2012. Net investment income in the 2012 fourth quarter was $0.53 per share, substantially unchanged from the $0.53 per share in both the 2012 third and second quarters. Our embedded pre-tax book yield before expenses was 2.6% as of December 31 compared to 2.8% as of September 30. The duration of the portfolio lengthened slightly to 2.06 years from 2.9 years as of September 30.

As Dinos mentioned, the total return on the portfolio was 80 basis points in the 2012 fourth quarter with emerging market investments and some high-yield corporate bonds augmenting the returns in our core investment grade fixed income portfolio. Excluding foreign exchange, total return was 67 basis points. Our exposure to Eurozone countries is listed in the financial supplement with minimal exposures to counties currently undergoing severe economic hardship. The 2012 year total return on the portfolio was 5.88%, including the effects of foreign exchange, compared to 3.81% in the 2011 year. Excluding foreign exchange, the full year return was 5.59% compared to 4.1% in the 2011 year.

Our effective tax rate on pre-tax operating income for the full 2012 year was a benefit of 3.8% compared to an identical benefit of 3.8% in the full 2011 year. The impact of Sandy produced underwriting losses thereby reducing our full year effective tax rate as a result. Fluctuations in the effective tax rate result from variability in the relative mix of income or loss reported by jurisdiction. Current quarter preferred dividend expense of $5.5 million was identical to the third quarter of 2012 but less than the $6.5 million of preferred dividend expense occurring in the 2011 fourth quarter. As discussed on last quarter's call, this reduction relative to the fourth quarter of 2011 is due to the refinancing of our preferred shares by retiring our series A and B classes and replacing them with the more cost-effective series C class. The $5.5 million is the true quarterly run rate.

Our total capital was $5.57 billion at the end of the 2012 fourth quarter, compared to $5.75 billion at the end of the third quarter and $5 billion at year end 2011 with the decrease in the current quarter largely due to the effects of $172 million of share repurchases. This represents an 11.6% increase in capital from year end 2011, and our debt to capital ratio remains a low 7.2% and debt plus hybrids represent only 13% of our total capital, giving us significant financial flexibility. At the end of 2012, we continue to estimate having excess over our targeted capital position. Book value per share decreased $0.60 to $36.19 versus $36.79 in the fourth quarter, reflecting primarily operating results and the short-term effects of share repurchases. Brook value per share, however, has increased by 13.9% relative to year end 2011.

With these introductory comments, we are now pleased to take your questions.

(Operator Instructions).

Vinay Misquith with Evercore Partners.

First from the mortgage insurance perspective. What are the long-term ROAs that you're looking in this? Is it 15% or is it higher than that?

Well, it's-- I would say is 15%. It can be better than that. You know, let's keep it at 15% and we are happy with that.

Sure. And you're saying that the accretion would only happen in maybe 2015 if this transaction goes through?

The accretion will not happen until actually 2016. We estimate this to close by probably third, fourth quarter this year. There is a process we have to go through. So that first full year of operation will be 2013. Second year would be 2014. So those are-- those two years we don't expect any accretion. And then after that we expect to be accretive in 2016. Mark, you want to add anything to that?

I agree. Just a clarification that the first full-- two full years are 2014 and 2015.

Right.

So 2014 and 2015. No income and 2016 will be okay. And also looking at the amount of capital that you plan to invest. You've planned to spend about $300 million. Do you think that you can up-size that based on the opportunities out there? Or is it too early right now?

It will depend on the opportunities. Traditionally as a company we always want to operate all of our units with a strong capital position. That's a commitment we make to customers. That's a commitment we make to the rating agencies. That's a commitment to make to our employees. So depending how well that business and how good our penetration is in that market, we will make sure that we have adequate capital supporting that business.

Second question is on growth. Your top-line growth in the primary insurance segment was lower than some of your periods has been. Just curious as to your perspective as to where pricing is right now and where you would like it to be so that it could significantly ramp up the top line.

I think where we differ with some of our competitors is on the very long tail lines. When you factor in new money yield through the duration of the liabilities and you do those calculations, even though some of these lines are getting actually the bigger rate increases, we still don't see them needing our returns. And for that reason, we're holding back our units in writing more. And to be more specific is some of the primary casualty lines, some of the excess and umbrella lines we're being very cautious about it. And it's excess workers comp, the workers comp line we'll be more cautious about these lines because of where interest rates are. Mark, you want to add to it?

Yes, Dinos. I would say, just echoing what Dinos said, the specialty casualty line over the last couple quarters haves gotten 12% to 14% to 15% increases. The way we look at it though, as we stated before, it's what's the absolute return those increases give us, not just the fact that they're increases. We still think that has a long way to go. Just to give you a little insight, our specialty casualty unit in the month that just went by, a little look forward, only wrote a couple hundred thousand dollars of new business. Why? Because we don't (yet) forgetting those kind of increases, but the absolute margin isn't there yet. So we could make the same comment on excess work comp. We could make the same comment on other longer tail lines of business.

Sure.

But I don't want to mislead you. I think the market is improving, so we are not disagreeing with some of our competitors' comments. It's just maybe it's our view on the absolute profitability of certain classes that differs. We have been always a patient Company because-- on premium revenue because, to tell you the truth, we love to grow high percentages if the market allows us. But we will not hesitate to shrink if we don't think we're getting the right rates. So it's a mixed bag right now. We got segments we're bullish and we're letting the units grow. And then we have segments we're not bullish and we actually shrinking.

The combination of that, though, is giving us a growth which is not-- I'm not disappointed. I think 7% on the net and about 5% on a gross on the Insurance Group is-- in this market environment is not a bad result. And from-- the first quarter I can talk about January 1. I think our growth in January 1 business was even a little higher than that. But one month doesn't make a quarter over year. So, in essence the momentum is going in the right direction as far as we're concerned.

One follow-up on that. Just curious as to what do you think the tradeoff is by being more conservative now, versus some of your peers saying we want to go write the business now and on the next renewal we will just take pricing up?

That's a strategy that sometimes it works and sometimes it doesn't. If you like to be a prophet to see what's going to happen, sometimes things happen to you because you might be reading the obituary pages and you find your name on it. (laughter) So I don't like to predict the future. What we see, we instruct our units this is the market that you're operating. What you see today, you take actions today. And if that changes tomorrow, you have the right to change your approach. And that's why we're a Company pays very little attention to premium budgets. Our units, they get capital allocation on a quarterly basis. We reevaluate that based on market conditions and we expect that agility from our operating units to be able to adjust and reflect what's happening in the market. So that formula has worked for us for 10 years and I'm not going to change it.

And that strategy that you just talked about isn't just a forward look in the insurance market. It's a forward look in the economy. So the idea that the economy would have recovered enough that customers would be willing and able to make-- and pay a big rate increase a year hence is also a bet we don't quite believe in yet.

Amit Kamar with Macquarie Group.

Just a couple of follow-ups on the prior discussion of-- on the USMI business. If you factor in the earn out, what would the deal equate on a price-to-book basis?

We want to pay true book value. So the earn out will give a fair price based on how the book evolves over the next two, three years. And that's what we negotiated.

Yes. And that has to-- the proof's in the pudding on that. As the in force of business is tracked over that period of time it will indicate an additional payment or no chance there.

And it could be a lot more. It could be a lot less. But we'll let the book evolve. If it's more, we don't mind paying it. At the end of the day we are getting the value out of it.

That leads to the related question. Does that tell us something about the state of the market with so many companies trading below book in the insurance land? If you are willing to look at other segments is that sort of overreaching what the state of the cycle is?

It has nothing to do with the cycle. We've been in this space now since 2009 and Andrew Rippert which is the subject matter more senior expert in our operations was hired over three plus years ago originally was for us to see if we have reinsurance opportunities and later on to evolve into insurance opportunities both here in the United States and also in other parts around the world, European Union, Australia, and other parts of the world. So this is-- this was a strategic move by us. But in our way we try do it as we acquire talent and as we find opportunities to acquire additional talent licenses, et cetera.

So it goes in that continuum. It has nothing-- we view this a long-term play. Mortgage insurance has been around for a long time. And with the exception of four or five years that, unfortunately, bad underwriting decisions were made by most in the space, if you examine the history of the product it's been a profitable line as long as you remain disciplined in underwriting.

Makes sense. Final question, and then this goes back to your remarks regarding capital. How do you think about the capital needs for this business? What I'm trying to ask is the deal is done and then you sort of ramp up. At that point of time how much capital-- is that a range of capital would you need and how would that play against capital management, let's say, for 2014 and 2015? I am trying to think of some sort of relative proposition.

Well, you have to, I think, fundamentally, you have to think about capital a little differently. First off any final capital decisions is going to be a function of two things really. It's the implied business plan and our growth and the opportunities that we see and also discussions we have with GSEs and perhaps other regulators. But it's not really a premium to surplus ratio or reserve to surplus ratio. It's a view of what's called really risk in force so-- to capital. And the industry has generally been operating around a 20 to 1 ratio. There is variations on that for different kinds of businesses and so forth.

But 20 to 25 to 1 is what's going on right now. A lot of the key competitors in that space are 23 to 25 to 1. That's the ratio to the risk to the capital. So 20 to 1 would be 5%, 25 to 1 it would be 4%.But again that's going to be a buildup over time directly as a function of our business plan.

So if I simply add up the risk in force numbers I can usually come up with the capital number?

That's correct.

Michael Nannizzi with Goldman Sachs.

One question on the reinsurance side. We heard other carriers talk about just low level of non-Sandy large losses in the fourth quarter. Did you see that in-- on the reinsurance side at all in the fourth quarter?

Yes. I think the quarter was good other than Sandy. We saw that.

If we were to back out the crop losses that were in the underlying and the impact of a benign 4Q on that front, would they generally offset each other? Just trying to think about what's underneath this sort of noise, the underlying sort of look like in reinsurance.

Yes, I would think that's a reasonable approach, backing out the crop. The crop is the really the outlier in this quarter out of the reinsurance segment. We could always talk about-- there is always going to be a few large claims that come through here and there. I would say that's a fairly reasonable approach.

One question on the portfolio. It looks like the turnover is pretty high. I am guessing it's not the entire portfolio that is turning over but maybe a subset. Could you give us a little bit of color on what the strategy is there and whether the source-- whether this activity is the source of the realized gains that you have seen over the last few quarters?

A lot of that strategy is-- first, the subset is generally in the treasury auction space. So the treasury auctions would come up and we would participate perhaps to varying degrees. So, yes, that would be a source of some of the realized gains. But it's really done for liquidity purposes as well. Over time the view of the yields of it when you're holding it for a while, it changes in the marketplace. But to your core question, it's really the treasury auction space is the core source of turnover.

So--

And let me add to that. Our philosophy is for total return. So we are not going to hold some of this to maturity just for the simple reason that we can be clipping the coupons all the time. If we see opportunities that we can make a trade, our investment department will make that trade. And we don't-- I know you guys like investment income versus realized gains. But at the end of the day we don't really care too much. As long as our book value per share keeps going up, eventually what you consider core earnings versus non-core earnings. Earnings are earnings. At the end of the day we like that approach.

I guess the question is how much turnover-- so it looks like if I just add all the cash that was up for the year, it looks like you've got about twice the total portfolio turning over in a given year. So roughly every six months. I am guessing that's not right. I am guessing there is a small sleeve of the portfolio that's turning over maybe three or four or five times. Is that what-- is that it that level of turnover?

Yes, there is some portions of the portfolio-- actually, they have turned over six times.

Josh Shanker with Deutsche Bank.

Following up on Mike's question. The reinvestment of the TALF assets, what's that going to do to yield in the coming year?

The TALF had-- first talk about what it does by not being there as opposed to what it's reinvested in. There is a two-pronged impact. You do have reduction in net investment income but you also have reduction on interest expense because this was a levered investment. It had some debt associated with it. There was roughly a reduction of a $1 million and change on the interest expense and it was about $2 million and change, I think, on the net investment income on a run rate quarter. Where that gets reinvested it's-- you should I guess really look at it as a distribution across all our asset classes. We can't say the TALF numbers went specifically to this investment or specifically to a different investment.

We have an asset allocation. It goes to the different buckets. And they get funded by either things that they mature, things that we trade or things that they come-- do. So we don't specifically say we're going to take the TALF asset and we got to put it into this place. It gets into the mix and then the investment department would say where do we have obligations? Like we-- in our alternative asset classes we have probably close to $0.5 billion of unfunded yet commitments.

So when we get the calls from these managers, then we have an obligation to fund. So sometimes we will go to alternatives. And if there is no calls for alternatives, we will put them into the regular portfolio.

Well, I guess more specifically in numbers, the-- I'm trying to figure out what the impact will be in the coming year. I assume the TALF assets were generally high yielding compared to other stuff in the portfolio? Or maybe that's incorrect.

It was high yielding, yes. And so the nature of it was--

So can we talk about the difference between the (bet) yield and what is available to you in the market right now I suppose?

Well, it depends which sector, right? The TALF yield was in the low double digits. I don't know if it was 12% or 13%, thereabouts. But we have high-yield investments in alternative space, which is exactly that. So I don't know if we're going to utilize all of it in there or not. I -- every month when we sit with the investment committee and our Chief Investment Officer we decide where, based on the asset allocation we have in the different asset classes, where they are going to put-- so it's very hard, Josh, to tell you is this going to reduce the investment yield by a point-- I mean, by 10 basis points or 5 or 3, or it will be neutral. My guess will probably be neutral.

Yes. I think, Josh, the one thing that you perhaps should not do is look exclusively backwards at some of the net investment income or even on some of the total returns because that asset allocation mix has changed over time. It's not necessarily reflected immediately. And as Dinos said, there are still unfunded commitments that clearly wouldn't be reflected in the financial statement. It has got to be with a view towards a changing mix as approved by our finance committee that gets into some of these alternatives.

The other question, I'm looking at the numbers. Maybe it's just the crop. But it feels to me of the (accent) your numbers in 2012 were a higher combined ratio than 2011 when you adjust for the luck of the good weather. Is that wrong? How do you view the two years--?

You've got to see the mix as to there is that equation. And also you got to see as to where we choose to book the accident year. We're not anxious to be aggressively booking good accident years even though the data indicates that we can if we wanted to. We just-- we feel nobody's sticking their hand in our pocket taking the money out. If it comes in the future, it will come in the future.

Meyer Shields with Stifel Nicolaus.

When you talk about the excess capacity or excess capital that you believe you have, does that allow--

We know we have. We don't believe. We know.

I am sorry. I didn't mean to make any such implication. Does that allow you to make riskier higher yielding investments in general?

Well, no. Our approach to investments is the investment department gets a capital allocation depending on the asset classes that-- and the amount of assets that we allocate in each one of asset class. So, yes. In general, in theory, the more capital you have-- excess capital, it allows you to be in more-- in more asset classes that require more capital allocation. But we sit down with the investment department and we make those allocations. Then we know what the capital grow will be because at the end of the day we're very, very fond of our ratings and we don't want to do anything to jeopardize those.

And just for the sake of clarify, assuming nothing else changed, you'd wind up with the excess-- if nothing else changed other than the shifts to some of these other investments, the level of excess capital would drop because of the distribution of those assets getting a higher capital charge. So as we look at our excess capital, it's fully adjusted for forays into riskier investments.

That makes perfect sense. When you look at the return potential for casualty lines, the ones where you are pulling back, is that anticipating a risk-free investment income-- investment return? Or --

Yes, that is our approach. Our approach is if you have a five-year duration or a seven-year duration business, some of these lines they get there, some of them they get excess [com]. It might even be 12-, 14-year duration. We apply the risk-free rate to the calculation. Yes.

Thanks for clarifying that. Can you comment at all on the year-over-year decline in fourth quarter reinsurance segment reserve releases?

I didn't view it as a decline. At the end of the day we do it ground up. A lot their releases that were short tail-- I think two-thirds of it, it was all short tail. And we haven't looked very hard on the long tail in this particular quarter. So I wouldn't read anything into it.

And also you're probably comparing it to the fourth quarter of last year whereas I think the reserve releases over the course of 2012 have been probably relatively consistent. I will have to go back and look.

Yes. I was looking year-over-year. But it's certainly basically flat with prior quarters.

Yes. But, you know --

It's whatever comes out of the analysis.

Right.

And then you make judgments along the way. Yes.

Jay Gelb with Barclays.

Can you talk about the ROE potential for the new differentiated types of programs you're putting on? You talked about the purchase commitment for the mortgage insurance business, for example. I know you have gotten-- you have had that business in Europe already and then of course the UK motor reinsurance. I am trying to get a sense of what that-- what the ROE profile of those programs are relative to the rest of your book.

Don't forget, in the mortgage insurance space we believe that it's at 15% our ROE business. And it can be better. It depends on your underwriting discipline and selections, et cetera. Having said that, the UK motor is not quite as good as that, but it's double digits. And as we said in prior calls, if it drops below that we would be willing to walk away from it if we can't get double-digit returns. So that's where those businesses are.

Now, we have high ROE business in other segments. Small amounts, but for example we just launched a facility on the internet to sell travel accident insurance directly. And if you go on the site it's called RoamRight. You will be able to see our offerings. Of course, it might take years before it becomes significant and adds to-- but those businesses have pretty high ROEs and we're excited about it. These things that we usually have in incubation for some time until we put them out. I think we did put a press release on that.

Yes, we did. Yes, we did.

On RoamRight.

And this had 27,000 hits so far. And they are extending the reach into other aggregator's. But Dinos is right. With an internet play there is some cost efficiencies.

I had to jump on the call a little late. I don't know if you talked about the outlook for share buy-back's in the second half of 2013 or if that was just waiting to see what catastrophe activity?

We commented on that so I don't want to bore everybody. If you read my prepared remarks, you will see it.

Ian Gutterman with Adage Capital.

Good morning.

Always in the back of the class, huh?

I try to get in the front but Vinay and Amit had taken up all the seats. They roped them off, so I couldn't get there.

Okay. (laughter)

A few clarifications. First, the MI, why is it not accretive the first two years? Is there an amort that's higher initially and then drops off? Or is it just the way the growth comes in? What's the math there?

I think it's pretty straightforward. You-- it's a labor-intensive business. It's-- there is a lot of infrastructure that's needed. That's what we're purchasing. And in addition to that with the employee forces in order to utilize that infrastructure. So it's-- just think of it as having over a short period of time an employee force that outstrips the revenue generation. And then that-- and then it reverses itself.

Even though you're acquiring a few years of in force?

Well, that's correct. But what you look at is the first couple of years you have more of the in force runoff, I'll call it.

Okay.

While the other one-- the new business is ramping up. So one kind of helps offset the other but doesn't entirely offset it.

And then a couple of numbers ones. Do you have-- I think you mentioned the crop impact. I didn't catch the number. About how much was that impact?

In the quarter it was-- on underwriting gain or loss it was almost $8 million.

$8 million. Okay. Do you have the reinstatements for this quarter and last year's Q4?

Last year's? I'd have to get back to you. This year the reinstatements-- are you talking Sandy reinstatements?

Exactly.

It was about 10. Between 10 and 10.5 in this quarter.

And the last one just to address the TALF again. I got a little confused. I think, Mark, you said it was about $2 million or so quarter of investment income. And then Dinos said it was a 12% yield and it looks like the balance is around $200 million. So $200 million at 12% yield would be $6 million a quarter, so what am I missing?

Well, you have to get into the details. But it wasn't the entire TALF. We, over the course of the year, depopulated some of that, mostly in the second half. So it wasn't going to zero. Even in the fourth quarter it would have had a partial reflection of it.

As deals that were coming off, it was throughout the year. So your calculation-- your math is correct but incorrect in the degradation of those amounts. We can give you the details as to how it happened over the year. It wasn't all done in the fourth quarter.

Got it. That makes sense. Okay. That's all I had.

(Operator Instructions).

Scott Frost with Bank of America Merrill Lynch.

I want to go back to your risk to capital assumptions for the CMG. Were you saying that you were going to run this at over 20 risk to capital?

Yes. 20 or less.

So we're planning to go back down to what the industry leaders were doing pre-crisis in the mid teens. Is that right?

Yes. It could be a little higher than that. It will be a conservative capital position. I think that's what that business needs. It needs some conservative underwriting and good capital allocation.

A follow-up from Meyer Shield with Stifel.

I was wondering, do you know when you are going to start breaking out mortgage insurance as a third segment. Is that first quarter or after the deal closes?

It's after the deal closes, we will-- once the deal closes, then the immediate the following quarter we are going to break it out.

Which in all likelihood would be the first quarter of 2014.

It can be the fourth quarter if we close by the third quarter. But I don't know. First quarter 2014 I think is a safe bet.

Okay. So I can procrastinate on that one.

Yes. And I didn't get into the names and all that. But Mark Grandison will be the executive in charge of the segment. He already runs all of our reinsurance. And then of course I di mention Andrew Rippert, which is the most senior subject matter expert responsible for all of what we do in that space. Then David Ginsberg will be our CEO at the PMI and CMGMI facilities.

Is there a good rule of thumb that we can use in terms of if there is an extra $100 million of catastrophes in the quarter what the impact would be on variable compensation?

Not-- the $100 million, if it's coming from cats, it doesn't have a huge impact because our compensation system, even though it's purely ROE based, it spreads cat earnings and cat losses on a rolling five-year average. We don't want anybody to either get hurt or benefit. I don't want them to bet their lunch for one year and then the next year I have the major loss. So if our entire book of business deteriorates by $100 million, I can give you the math surrounding that because our incentive compensation is pretty straightforward. We earn 8% or better. That's the cliff. At 15% they get 100%. At 23% of ROE they get 200% and they get 400% if we get into the high 30%s. And then they get nothing if we are below 8%. So what $100 million on the entire book does to ROE, it will give you the answer.

Just to augment that a little bit, Dinos' description was for the majority of-- within the group of cat, which is in the Reinsurance Group, the minority, let's say 20%, in rough terms. The Reinsurance Group doesn't smooth because they already have smoothing, called the [enuring] covers and the cat program that they have.

This is on the insurance side.

On the insurance side.

Right.

So depending on where that $100 million got felt in the group, it's either smoothed over five years on the reinsurance side or if it's really hit, the cat program responds and, therefore, limits the impact on the group.

A follow-up from the line of Michael Nannizzi with Goldman Sachs.

One quick one. I think, Mark, you talked-- or Dinos I guess talked about the PMLs and the net change from last year. I was curious. Property cat premiums in reinsurance and in also non-cat property were a bit higher in 2012 versus 2011, 4Q versus 4Q. How should we think about the cat load in reinsurance, if that's a fair question? And does that change a lot from what we might have thought it would have been before that?

As we said many-- we allocate on the basis of the insurance versus reinsurances. If I have $0.5 of cat, I give $0.4 to the Reinsurance Group and $0.1 to the Insurance Group. So that's our allocation. And then we leave it up to the cat teams to determine which zones how much. As long as we don't exceed the 25% to capital ratio that we superimpose from corporate down to the operating units. So as you see, sometimes you will see some changes like the Gulf was our higher zone a quarter ago. Now it's the Northeast. We're a little underweight in Florida. We had only $602 million in some-- in Florida.

Sometimes we were even higher on a lesser capital a couple years back. But those are the determinations of our cat underwriting teams as they see what comes in and do they like the rates, et cetera. And in a lot of other zones we are not totally utilizing our capacity either because the available business is not there and/or we don't like the rates. I'm talking about Japanese quake or wind or South American quake or Australia and New Zealand, whatever other part of the world would be very, very underweight because we're-- in Europe for [wind] because we don't like the rating environment. And that's individual Company-- we're not thinking that we're smarter than anybody else. We're just a different opinion, you know. Other people who provide the capacity think they're getting adequate rates. Our guys don't feel they are getting adequate rates, so we don't utilize a lot of that capacity.

Ladies and gentlemen, we have no other questions so I'll turn the call over to Mr. Iordanou for closing remarks

Thank you all for your patience. Looking forward to speaking to you three months from today. Have a wonderful day.

Ladies and gentlemen, that will conclude today's conference. Thank you very much for joining us. You may now disconnect. Have a good day.