Arch Capital Group Ltd (ACGLO) 2011 Q2 法說會逐字稿

Good day ladies and gentlemen. Welcome to the second quarter 2011 Arch Capital Group Limited earnings conference call. My name is Modesta and I will be our coordinator for today.

At this time, all participants are in listen-only mode. Later we will conduct a question-and-answer session.

(Operator Instructions)

As a reminder, this 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 assessments and assumptions, and are subject to a number of risks and uncertainties. Consequently, actual result 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 filed by the Company with the SEC from time-to-time.

Additionally, certain forward-looking 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 attends the forward-looking statements in the call to be subject to the Safe Harbor created there by. Management also will make reference to some non-GAAP measures of financial performance. The reconciliation to GAAP and definition 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 Web site. I would now like to turn the conference over to your host for today Mr Dinos Iordanou and Mr John Hele. Please proceed.

Thank you, Modesta. Good morning everyone and thank you for joining us today. In the second quarter we continue to see significant catastrophic activity, this time emanating primary from severe weather events that occurred in the United States in April and May. The level and intensity of these storms were unprecedented and a departure from historic patterns, and serves as another reminder that in the insurance business significant natural catastrophes do occur and need to be priced for.

I continue to be pleased with our underwriting approach and the discipline which our insurance and reinsurance underwriting teams are executing our overall underwriting strategy. As the soft cycle has evolved, we have continually adjusted our book of business to lines and classes that are priced appropriately on a risk adjusted basis in order for us to achieve the required return relative to the risk assumed. As a result, of executing this strategy, on a trailing 12 month basis, our short tail exposures have grown to be about half of our premium. Our medium tail is roughly 25% of premium volume and our long tail has been reduced to approximately 25%. This is a significantly different book of business than we had several years ago.

Total net catastrophe losses for the quarter were $95 million with North America weather related losses contributing $79 million and the balance coming from the re-estimation of our first quarter catastrophe losses and the second quarter New Zealand earthquake. Our annualized return on average common equity was 6.1% on a reported basis, which was affected by the above mentioned cat losses. We exceeded our quarterly cat load.

By our own estimation, we continue to believe that in the current underwriting and investment environments we operate in, on an expected basis, we were still able to achieve approximately 9% ROE on an underwriting year basis. This is essentially the same level we estimated for the entire 2010 underwriting year.

Our investment performance for the quarter was good with a total return of 1.7%, aided by a slight improvement in treasury rates. As a result of our investment performance and despite the cat losses, we were able to increase book value per share to be $31.00, which is an increase of 13.3% from a year ago, and 2.2% from a quarter ago. From an underwriting point of view, we recorded 100% calendar quarter combined ratio, which is an acceptable result in a period marked by above average catastrophic losses. Cash flow from operations remains solid at $222 million.

The broad market environment continues to be competitive with most long tail product lines having plenty of capacity available and which are experiencing slight price declines. For us, these classes represent about 20% of our quarterly volume, and 25% of our trailing 12 month premium volume.

In the property and property cat areas, the environment improved with the best increases today reflected in international cat exposed business. As noted earlier, our short tail book of business total was roughly 54% of net premiums for the quarter, and approximately half of our volume on a trailing 12 month basis. For medium tail lines the environment is stable, for some classes, and improving for others. Our medium tail business was approximately 26% of this quarter's volume, and 25% of the trailing 12 months volume.

From a premium production point of view, our gross written premium was up 12% and our net written premium were up 13%. Most of this production increase emanated from our reinsurance operations which reported growth of 36% in gross written premium and 33% in net written premium. Part of this growth is attributable to one account that was renewed on a two year basis, and adjusting for this account, the percentage increases would have been 25% and 21% respectively, still a very strong performance. Most of the balance of the increase in written premium was in short tail and medium tail lines.

Our insurance operations were up 3%, on a gross written premium basis and 4% on a net written basis. During the second quarter, we saw no significant change in market conditions, with rate levels for all lines of business either improving slightly, or staying the same as in the prior quarter. The areas under the most pressure continue to be executive assurance and health care, with mid single digit rate declines. All other lines report a flattish or slightly positive rate movements. In our reinsurance business, we saw good improvement in the property cat area, with average rate increases of 8%, in the United States, and 28% on average in international business where increases range from as little as plus 5% to as high as 70%.

Across our business segments, even in the most difficult markets, we always look for opportunities to find acceptable books of business to underwrite, without sacrificing underwriting standards. We are starting to see some of these opportunities from both our existing product offerings, as well as from our new product initiatives which have helped stabilize premium levels. Our insurance group continues to emphasize and move their book of business to smaller accounts and to relatively less volatile lines with an emphasis on reducing their exposure to the US casualty business. The area with significant volume reductions in the second quarter was professional liability. This reduction in volume primarily resulted from the larger account sector of this business.

As we mentioned on last quarter's call, we continue to monitor the primary casualty E&S sector in the US. Although we continue to see price increases in isolated areas, we remain in a defensive posture in this line as we still believe that the rate improvements today will not produce an adequate return.

I would like to reiterate that our capital management philosophy has not changed. We have consistently and will continue to return excess capital to our shareholders when we cannot profitably deploy it in our business. As is customary for us, during the hurricane season, we will temper our share repurchase appetite. Of course, should we find opportunities to buy our shares at very attractive prices because of market dislocations, we will continue to do so.

Before I turn it over to John for more commentary on our financial results, let me update you on our cat PML aggregate management. As of July 1, 2011, under RMS 10, our one in 250 PML from a single event was $740 million, or 18% of common equity. This represents roughly the same level of exposure and percent of common equity as of last quarter's end.

We are continuing a process of reviewing RMS 11 and its various changes and have not yet finalized our examination at this point. However, under RMS 11, based on the current iteration of the model, our preliminary estimates of PML for one in 250 year events, indicate that these exposures estimates in comparison with the RMS 10, will produce increases which range between 10% to 30%, depending upon the zone with all such preliminary amounts remaining within risk management limits.

With that I'm going to turn it over to John for further commentary or financials.

John?

Thank you, Dinos. Good morning. Regarding our premium income, on a consolidated basis, the ratio of net premium to gross premium was 77%, about the same as a year ago. Our overall operating results for the quarter reflected a combined ratio of 99.6% compared to 90% for the same period in 2010.

The 2011 second quarter included $95 million, or 14.8 points of current accident year cat activity net of reinsurance and reinstatement premiums, compared to $7 million or 1.1 points in the 2010 second quarter. The 2011 second quarter US storms and New Zealand earthquake had a gross impact of $98 million and a net impact of $82 million. The re-estimation of the 2001 first quarter events of the Australian floods and Cyclone Yasi, the New Zealand earthquake and the Japan earthquake and tsunami added $21 million gross and $13 million net to the second quarter provisions for the 2011 cat events. The re-estimation was mainly driven by further claims development and contingent business interruption expected claims from the Japanese event.

The 2011 second quarter combined ratio reflected 8.9 points, or $58 million, of estimated favorable prior year reserve development. Net of related adjustments compared to 5.2 points or $33 million, in the 2010 second quarter. The prior year development in the second quarter of 2011 reflected net favorable development primarily in property and other short and medium tail lines, as well as in the reinsurance segment casualty business mainly from the 2002 to 2005 underwriting years. And also included a $5 million reduction in the provision for the 2010 cat events. More over, excluding the cat activity in the quarter, we generally experienced better than expected claims emergence on most lines.

The 2011 second quarter current accident year combined ratio, excluding large cat events and net favorable development, was 101.8% in the insurance segment, and 79.7% in the reinsurance segment, both consistent with results as of a year ago. The 2011 second quarter expense ratio, 32.5%, was 0.8 points higher than the 2010 second quarter resulting from higher costs in part from billing the insurance accident health business and foreign currency adjustments as well as higher incentive composition accruals.

On a per share basis, pre-tax net of investment income rose to $0.63 in the 2011 second quarter compared to $0.57 for the same period a year ago and $0.63 in the first quarter of 2011. Our embedded pre-tax book yield before expenses was 3.23% at the end of the 2011 second quarter, down from 3.52% at year end which reflects lower reinvestment rates and a short duration.

The total return of the investment portfolio was 165 basis points in the 2011 second quarter, compared to 150 basis points in the 2011 first quarter. Excluding foreign exchange, it was 154 basis points in the quarter. The total return in the second quarter benefited from good returns on fixed income assets, partially offset by negative returns on equities and some alternative assets.

We recorded net foreign exchange losses of $18 million in the 2011 second quarter, due to the weakening of US dollar against the Euro and other currencies. These losses resulted from revaluing our net insurance liabilities required to be settled in foreign currencies at each balance sheet date. However, this should be compared to the 11 basis points contribution to total return from foreign exchange on our investment portfolio which offsets some of this income statement loss in the equity section of the balance sheet.

Our allocation to equities was approximately a net 3.5% of our investable assets including cash. At quarter end, while alternative investment and equity method investments were consistent at 6.3% of our investable assets. We continue to maintain the vast majority of our investable assets in very high quality fixed income investment portfolio, with an average credit rating of AA plus. The duration of the investment portfolio was 2.87, about the same as 2.83 at year end. We continue to maintain a shorter duration of our assets than our liabilities, which are approximately 3.5, due to the continued global uncertainty and potential volatility of interest rates.

For the 2011 first half, our effective tax rate on pre-tax operating income was a benefit of 2.7% and 1.7% on pre-tax net income. The cat activity this first half year and low investment returns have resulted in a beneficial net tax position.

Our balance sheet continues to be conservatively positioned with total capital at $4.8 billion at June 30, 2011, up from $4.7 billion at March. In the quarter, we purchased 0.9 million shares for $29.6 million, at average price per share of $33.51, at a ratio of 1.09 to the average book value during the quarter. Which fits in to our expected return of under 3 years at this profitability level. Our debt plus hybrids represent 15% of our total capital, well below any rating agency limit for our targeted rating. Our book value per share ended the quarter at $31.00, up 2.2% from last quarter and 13% from a year ago.

Our excess capital position, which we define as the rating agency actual capital in excess of the acquired for an A plus rating plus a buffer, was estimated to be approximately $300 million at the end of the 2011 second quarter under RMS 10 and including AOCI. Depending upon the final implementation of RMS 11, we would expect our excess capital to be reduced by an amount yet to be determined. We are comfortable with our capital position heading into the usual cat season. With these comments, we are pleased to take your questions.

Modesta, we are ready for questions.

(Operator Instructions) Your first question today comes from the line of Keith Walsh with Citi. Please proceed.

Good morning everybody. Dinos, you mentioned underwriting ROE's remaining in the 9% range, I was curious why that was. I would assume with the property cat and especially growth that you were writing that well above the 9% you cited. It was a pretty sizeable chunk this quarter. If you can touch on that.

Yes. The -- you got to look at the entire book of business. You are absolutely correct, I think the prospects of better returns on that portion of our business is there, but you have to also see where we are losing ground in the executive assurance and health care, even though arrest much smaller part of our business, we are still giving rate reductions, so you are losing grounds. Then the rest of the lines they haven't really -- even though we are seeing a little bit of up tick on rates, flattish, maybe slightly up, I don't think that keeping up with trend is still positive, so when you put all that in put the positives and negatives we didn't see any move in the profitability I think that we are achieving on a underwriting basis. And 9% is not really our target, but at the end of the day, it is what we can achieve in this market.

Okay. The other part is, on the long tail lines, profitability is somewhat dependant upon the new money yields, so as you saw from John's commentary, our embedded yield on new money invested is coming down because as we reinvest the funds new cash flow it's going down. So, that affects the ROE too because even though the duration of liabilities on the long tail is still the same, you are earning a lot less on the float.

Okay. Just thinking about new business growth and the property cat area, is this -- are you winning because of wounded competitors? Is that a major reason you are seeing new flow?

I wouldn't characterize it as such. I think in some parts of the world, and even in the United States, we never really push the envelope to our upper limit of our tolerance, we always have room and we continue to have room to write more. So, where we found the opportunity is independent if there were opportunities overseas or opportunities in the United States, we did take advantage of it. At the end, we let the market determine where we have the best chances to get a good return and that's where we go.

Thanks a lot.

Your next question comes from the line of Jay Gelb with Barclays Capital. Please proceed.

Thanks, and good morning. Dinos, with the excess capital position potentially coming down a bit more with the implementation of RMS 11, does that mean Arch could be more geared towards premium growth as opposed to share buy backs going forward?

I think we are going to do both. If we find opportunities for premium growth, we will take them. That's our number one priority. We are in business to produce underwriting profits for shareholders. Where we have an opportunity to expand our business through underwriting, we are going to take those opportunities. Having said that, though, we are not going to keep excess capital on the balance sheet, so depending where the market goes, how this hurricane season fairs for the industry and us, we are going to re-evaluate where our capital position and continue with share repurchases if we can deploy the capital in the market place.

Jay, because we are under levered, if the market does turn more and it becomes quite hard we have great capacity to increase writings if we see the ROE is there.

Right.

Flexibility is -- plenty of flexibility on the balance sheet. From a directional point of view, I don't think there is any change, would be steady as we go. In prior years we kind of wait and see on the third quarter, maybe a share repurchases being light is usually in the third quarter because we want to see where the hurricane season takes us. But we had significant activity in the fourth and first quarter. That pattern has been around for you guys for what, three or four years now - since we started a buy-back program.

Right. Okay. On the reinsurance segment, the 36% growth as reported and a big portion of that clearly was renewal of a multi-year deal and you probably also had some reinstatement premiums coming in. On a normalized basis, what do you think that growth rate was in 2Q?

Well, on a normalized basis, I think it was in the 20% range. Reinstatement premiums weren't that significant, one large deal is I gave you the delta, it was like 11%. 36% versus 25%. That's a deal that has always been on a two year basis. For year-over-year comparison it throws off our numbers a bit for you guys. But you've known it for years that we skip a year and then it comes in. On a earned basis, it makes no difference because you are earning the premium over 24 months anyway.

Just in the follow up on the reinsurance segment, what do you think is a reasonable expectation for growth going forward? My guess is it wouldn't be in the 20s.

Yes. Listen, if I was that good I could predict the future I would be in Vegas betting on the tables. It's whatever the market shows us. We don't shy away from underwriting and taking risk as long as we believe that there is a profit in the transaction. We found opportunities in the first and second quarter of this year, we took advantage of those. If those opportunities continue to be presented to us, you will see growth. If not, we are not ashamed to go back to our defensive underwriting pasture and have no growth or even negative growth.

Thank you.

You're welcome.

Your next question comes from the line of Josh Shanker with Deutsche Bank. Please proceed.

Thank you. Hope all is well. I was wondering if you guys could detail a little bit on loss trends by line, we've had a few conference calls here and not getting a lot of detail, but everyone agrees that lost costs are up and maybe you could add a little bit of detail to that.

I can give you the ranges, Josh. We've seen as low as 2.5%, in some lines, to as high as 6.1%. We don't -- we don't change trend projections on a quarterly basis. One set of data on another diagonal, it's not going to change those projections. Our actuaries take a longer term view and they look at both frequency and severity and they do it by what we call IBNR family or product line and then we use that in pricing our business. The range is as low as 2.5% and as high as 6.1%. That's what we have by line of business. Of course, the highest is in the medical care area, executive assurance has a high trend too because of severity.

That's a great answer, thank you.

You're welcome.

Your next question comes from the line of Greg Locraft with Morgan Stanley. Please proceed.

Thanks. Wanted to actually just take your temperature on, again, the capital side. I'm trying to understand -- I know you have 300 today and, obviously, we are entering an RMS 11 world, if it's just a normal hurricane season, how will things look in to year end and what impact does that have into the marketplace from pricing perspective in to 2012 and beyond?

To give me a little leeway to answer your question is what is a normal hurricane season? Would you define that for me?

$18 billion is the average ten year losses.

So, having one hurricane with industry losses of $18 billion. For our book of business it would be contained in normal cat loads, so it wouldn't have any affect on our capital.

Okay, great. So, then it is sort of -- you are trying to navigate from an RMS 10 to an RMS 11 world and you'd feel pretty good about where you are sitting.

Don't forget, RMS 10 and 11, is just another recalculation on the same set of exposures, right. We have been operating with RMS 10 for a couple of years. We have an idea as to what our exposure, how you measure them, et cetera. That's why we are very, very careful before we fully implement RMS 11 that we do understand all the changes that they have made to the model and we agree with them.

Don't forget, even with RMS 9 or 10 or going back, we always made our own customary modifications ourselves and if you look at our cat management over the years, I think we get pretty good marks. I'm pretty confident in the ability of our teams to not only understand the external models, but also how they internally make tweaks to them in order to serve our underwriting needs and that's the process we are going through right now, that's why I didn't want to just plug in RMS 11 without having a full understanding of what modifications they made, do we agree with them and all that. Does the empirical data that we have jives with that or do we have to make some of our own changes. John, do you want to -- you spent a lot of time with the cat teams too.

Greg, I think what is it important when you look at us and also think of other people in the industry as well, is when we say RMS 10, it's, as Dinos mentioned, how we have implemented RMS 10. We have already done a lot of backtracking of various storms and how that mapped the model and had increased some exposures in certain areas, and likewise with RMS 11 we are looking at the exact same way, we are looking at their changes and their empirical research also from other vendors and looking at their research before we finalize and zero in on our final numbers. The percentages may be different from us, from other people, it also depends on where the locations are. The RMS 11 versus 10 raw models have changed quite a bit by regions. It's going to depend a lot upon your book of business and this takes some time to make sure you are comfortable with. Overall, we think we have sufficient capital for all this and we will continue looking to write some good business.

There is no restriction in our ability to respond to market opportunities. We will stay within our risk management tolerance. We are not going to increase that. We are comfortable at 25% of common equity, we are going to manage the book within that and we are comfortable where we are today and RMS11 will be implemented within the Company shortly. We are in the process of doing it right now.

Finally, Greg, just one point. When we say excess capital, that's from our target and our target number is A plus, plus a buffer. The buffer is there to get through big storms or bad events in the world and that's how we run the Company. Excess is really an amount above A plus and above a buffer. It's quite up there.

Okay. Great. Just again on the capital side, how do you think of the impact of two events in the fixed income markets? One is higher interest rates, just let's say they go up 100 basis points or 200 basis points, and secondly would be sovereign downgrades on the US side. How would that impact your thoughts around your excess capital position?

So, the duration of our entire portfolio, which includes everything, is 2.87. If interest rates went up around the world, because we have business in other currencies, it would be about $300 million. If you take our total investment portfolio and multiply that out by the 2.87. That, though, is a very big move from where we are on the whole yield curve across the board. More likely you will see this in Europe, it's not the whole curve moving for these countries, it's shorter term is moving up more, it's hard to predict. I think we feel with our buffers we have versus our capital position that we can withstand some pretty big shocks here, that happened around the world.

When it comes to sovereign downgrades, we don't have an exposure to the picks or the risks in Europe. It's very hard to predict secondary impacts on what would happen around the world to these things and the US downgrade is also hard to predict what might happen. You may think rates may go up instantly. History has showed that when Japan was downgraded from AAA to AA plus in 2001, the 10 year Japanese bond actually went down 24 basis points, in the subsequent time. We are in a very bit of unchartered territory here and by being short is really one of the best protections of being conservative on the credit quality.

We do own a muni portfolio which is a little over $1 billion in the US, but the muni story is specific to each bond. We do a lot of the credit analysis within our worlds, we just don't depends on outside advisors and we feel pretty comfortable where we are with our muni portfolio because even in some states that there might be a lot of stress, et cetera and there might be some municipal default, when we run that against what we own, we feel comfortable with the positions we have on the muni side.

Thank you very much.

You are welcome.

(Operator Instructions) Your next question comes from the line of Matthew Heimermann with JPMorgan. Please proceed.

Sorry to beat the dead horse here on --

If it's not dead, we will kill it.

Let's work towards that. So, just in terms of -- I was curious on how RMS might impact a couple of things, specifically just in terms of what is driving the increase, is that likely to be -- is there any specific region driving the increase and I think last quarter, historically southeast has been your biggest zone, is that potentially going to change under -- once you implement the new model?

Let me give you some preliminary color and we haven't finished all the analysis. It seems to us that it has less of an effect in Florida for us because of the changes we were doing to how we were viewing Florida risks. It has more of an impact in Gulf states meaning Louisiana, Texas, et cetera, and most of that is coming from the changes in storm surge. And also inland damage ability going in 30, 40, 50, 50 miles. So, when I gave you those ranges of 15% to 30%, to be more precise it was like 14% something to 29%, the 29% was mostly for the Gulf states, not for Florida, but I want to reemphasize again, these are preliminary numbers because we like to be very thorough and our teams are spending a lot of time and a lot of effort on, not only implement the model but understanding it and we don't do things that we don't agree with. We got to get agreement as to what we are going to implement by our own cat teams.

Matt, there are three major areas that everyone has to analyze with the RMS 11, the first one is the shorter term frequency. The model has a higher frequency assumed in the short-term due to ocean warming, the cyclical ocean warming cycle, so there is a long-term average of frequency over 100 years or more but there is a shorter term cycle. If assigned, do you agree with that and how do you want to implement that? The second is the vulnerability, or as Dinos mentioned, the further inland, how far the storms come in, and what the damages are. And the third is the storm surge. Getting comfortable with all those takes some time, we are analyzing it, we are running it, we are backtracking it versus prior storms and also reading all the empirical research and speaking with RMS, I do truly understand where they are at.

Just taking this a little bit further, any -- is insurance or reinsurance or is it similar between the two that's disproportionately responsible for whatever increase we will see?

Well, to a great extent on the reinsurance side, we are -- we take 100% net. We do buy some retro and some ILW's to protect different parts of our book and we do that. That's more, I will call that managing PML aggregates. On the insurance side, we buy a lot of cat cover, so there is scenarios that you cycle over, that say you take $75 million deductible or net retention on each storm and you buy $325 million above it. It's only when you go over the $400 million that -- and there is some scenarios that in our insurance group, there is potential for us to be cycling over.

Now, there is two ways to deal with that. Either you reduce exposure end or you buy more protection at a cost. You can always go and buy another $50 million or $100 million of protection. That is the kind of evaluations we are going through analyzing our book of business, and getting to an agreement. Don't forget, we are taking the risk, we have to agree to it, working with RMS and their scientists in explaining everything that they have done to the model and then we are going to make determinations.

Right now, we are very comfortable with the exposures that we have and where we are at. But, this is -- we are spending a lot of time and it's going to evolve. It's going to evolve over the next month or two. It has not changed what we have done in the cat business. Because that season is pretty much over, after July 1, even at the worst possible escalation, we did 30% across the board, you are still within, well within, our tolerance, so from that perspective it didn't give us any -- it won't become an issue again until January 1. We have plenty of time to analyze and refine and make determinations on the model.

Okay. That's helpful. If we translate this back to the capital discussion, how should we think about the proportionality of the increase in the PML relative to how much excess capital you have. Simplistically, one could come from it. This is why I'm asking. You could come to the approach and say excess capital is the difference between where ever your one in 250 PML is relative to common equity versus what the required equity would be at 25, but I think in reality the models are more complicated. Could you give us a sense of how would that 300 look if it was up 10 versus up 30 or some kind of range of outcomes like that?

Let me -- let me reemphasize what John said before, because there is rating agency required capital. Then there is Arch Management required capital. We always calculate our excess capital on the Arch Management required capital, which is a bigger number than what the rating agencies require us for our current ratings. The reason we have that buffer, as we call it, is because we always try to operate with a simple principle that I want to begin the year and end the year with rating agencies required capital by having a enough of a buffer that if I have one in 250 year event or if I have a 200 basis points parallel movement on the yield curve, my balance sheet can absorb that shock, $500 million, $600 million, in between the earnings of the year, potential earnings and the shock, I begin the year and end the year with a rating agencies required capital. That's a very, very conservative position.

Now, as the models become more robust, you're never going to eliminate the variability of the models around the mean. As they become more robust and we get more and more confidence, that means we might not have to keep as much cushion as we have in the past. That's our own internal determination. That's why we are spending a lot of time in trying to understand the changes in the model and what these numbers mean and how much cushion do we need in order for us to be feeling comfortable. Right now, I can tell you based on preliminary numbers, I feel that maybe we won't need as much cushion to bridge us between what the rating agencies want from us and what our Arch Management with a cushion needs. We haven't made those determinations yet. We are still in the process of analyzing.

The principle, though, we are not going to change. We want, independent of all these stress events, independent if it's a financial event, with a significant movement on the yield curve, 200 points, parallel move on the yield curve or $50 billion or $60 billion or $100 billion storm, we want to be able to begin that year and end the year with rating agencies require capital so our rating doesn't suffer. We will be the company that is going to be the go to company by clients because financially we continue to be very extremely strong.

We aren't in a position yet to give you any guidance on how the excess capital rating agency model works, because they work both with single and also aggregate and how all these add up across the board and how we implement RMS 11 can have an impact on that. That's why it's still too early for us.

That's fair. I appreciate the color, Dinos. I think what you are saying is very fair and very thoughtful in terms of how to think about capital. There is definitely some yings and yangs to how you implement this and it shapes your total view of capital, but I guess what we are struggling on this side is just to think about -- and this might not be the right way to think about it, but for a reference point, getting a sense of what the potential change in PML does to capital on a static basis and if you held your buffer the same and let's say it would help things constant and the PML went up, under RMS 10, you wrote more business, that might not be fair. If -- how does that 300 change?

It is a simple calculation, Matt. It is a simple calculation. If you take it in a simplistic way, as you put it, you take your worse case scenario, 30% on $700 million is $210 million, the excess goes from 300 to 100.

Okay.

But it's not as simple as that. I don't want you to leave with the idea that it is as simple as that. That's why we do spend time on analyzing and understanding what we have. Don't forget, putting -- we don't run the Company because we want to be right that we have this excess capital, this short. We run the Company on the basis taking good exposures, appropriate for the size of balance sheet we have, and independent of major events, financial events, catastrophic events, et cetera, we stand as strong or stronger than most of our competitors. That's the thought process.

Because there is -- the day after and how you feel the day after a major event that is critical to us, we want to always be the company that is first able and willing to satisfy the needs of their existing customers because they continue to have capacity and also acquiring new -- and it's all interrelated. It's our thought process about what kind of balance sheet we have, how much room we have on the balance sheet, what the financial leverages of the balance sheet, it's all interrelated.

We don't think this constrains our ability to grow this implementation of RMS 11 from the preliminary work we are doing. Because, as I mentioned earlier, if we see great business around the world, we can issue some more debt and grow as dramatically. There would be a constraint, a local constraint, we may cap out to be at our 25% of common equity in one zone and cap that, because we will never take too much risk in any one place no matter how good the pricing looks. Overall, we don't think this is a business constraining implementation.

That's fair. The color is good. Thanks.

Your next question comes from the line of Jay Cohen with Bank of America, Merrill Lynch. Please proceed.

Thank you. I have three questions. I think they should be relatively short. On the investment side, if rates -- if interest rates in the market stay where they are, when would you think that the book yield would equal the new money yield?

We have a duration of three. It will roll over within -- it's been rolling down. We are not only investing in treasuries, we are investing in a series of investments around the world as well, both in the US and other places. It would be next year to the year after sometimes.

Okay. Second one, this is a short one. On the reinsurance side the growth in the cat business, did some of that growth come from retro deals because I know there was obviously some demand in the market for retro cover?

No. We are not fond of retro, we don't write but a tiny little piece of retro. Sometimes we are buyers of retro, but not sellers. It came from traditional excess of loss opportunities, both internationally and domestically. There was also some reshifting on our domestic book to what part of the tower we participate so our guys always look to see based on their own curves as to where they think they are getting the best rate in relationship to the capacity they put out. That's why the changes were. Of course, we wrote some medium tail business that we thought it was attractive and gave us good ROE characteristics.

Great. Thank you. The last question, you mentioned claims trends, that 2.5% to roughly 6%, I'm assuming that's what your assumption is for claims trends when your pricing and reserving the business, but I guess a question I had was are you seeing any changes in the claims frequency? Anything in this quarter that suggests the trend might be changing even if you are not changing your view from a pricing stand point, does data suggest the world is changing at all?

Jay, you are conflicting me because you say trend and then you want me to -- trend to me is something that takes over a longer period of time. One quarter. As John said, I think in his prepared remarks he touched upon that our expected claim emergence was lighter than what we --

Cat events.

Yes. Put the cat aside. But on the regular book, both on an incurred basis and on a paid basis, but it's only one quarter. So, do you change your view for everything you do? We don't. At the end of the day, we talk about trend is more longer term and looking at a lot more than just one quarter's data.

I probably misused the word trend. But it sounds as if there was no data, no new information this quarter which suggests a change in the environment, in the claims environment, frequency specifically.

I can only -- we do the trend analysis by, as I said, by line of business, usually once a year. It's not always at the same time. One quarter we will do one product line, next quarter they are going to do another product line. They cycle over, because you got to divvy up the work. What John said, though, we always look at, but we don't pay a lot of attention to it. Yes, we have an expectation, every quarter as to how much emergence we are expecting and how much, both on a paid and on an incurred basis, and it was lighter than usual.

On the other hand, I had other surprises. My attritional losses on property, they were a little higher than what I expected. That happens, one quarter you got attrition going up, another quarter you have attrition going down. It's not always coming smooth. That's the reason we don't take one quarter -- an indication of one quarter and we say that's a long-term trend and everything is going to change and it's going to be based on that.

Got it. Thanks a lot, I appreciate it.

Your next question comes from the line of Scott Frost with Bank of America, Merrill Lynch. Please proceed.

You talked about you didn't have exposure to Euro zone issues, no peripherals, is that mainly in the sovereign portfolio or is it also in the corporate portfolio, maybe go over your exposure to Euro zone credits and your investment portfolio?

In contracts that we have, we don't usually hedge positions. We have what we call a natural hedge. If I write contracts that I have to pay in Euros and I get payed in Euros and premium, we try to invest the reserves we have or the obligations that they are going to come in that same currency. Most of that we do by buying German bonds, some corporate and we buy some gelds in the UK for the British pound.

We don't have Italian banks, Greek banks.

We have no Spanish or Italian or Portuguese debt or anything of that sort. Not on our book.

No exposure to Santander, for example, I saw it in your Q, must have been in a special structured issue?

We had a covered bond from Santander which is backed by paying mortgages.

I guess that's the thing. Is there some sort of breakout of issuer exposure in I guess Euro zone, especially peripherals, by however characterized? Is that something that you guys have provided or can give some color on?

We said, in these areas we don't have any.

Zero. All right, thank you.

Your next question comes from the line of Ian Gutterman with Adage Capital. Please proceed.

This time you are not last.

I was too fast punching in, I'm sorry. A comment you made to Jay a minute or so ago about the attritional losses being up in fac. Does that mean that actually that you had improvement in your accident year, ex the fac?

I said attritional losses in property that were a bit up this quarter.

Is that in the insurance or reinsurance?

It was in insurance.

You reported a flat accident year ex-cat, but if the attritional fac was worse this year, then there was actually an underlying --

Also you got to look at change in mix too. The change in mix -- don't forget, our casualty business is coming down, the highest loss ratio is associated with long tail casualty. When you go into writing less of that, and then you are writing more in small accounts and short tail, which has a lower loss ratio. You can go from one to the other unless you had a static book. You've got to look at it by component.

Understood. I just wanted to make sure that's what was driving it, it wasn't something new. Also, your comment at the beginning about the 9% accident year return across the book, how does that differ insurance versus reinsurance, I assume it's higher in reinsurance?

Yes.

Can you give some magnitude. Are we double digit reinsurance?

I don't want to embarrass my guys.

I guess I am wondering if insurance was pulling its weight.

Insurance is not a business you can navigate as well as reinsurance. Reinsurance, you can be a bit more opportunistic, and fluctuate volume in more significant ways than insurance. I can tell you -- and also the reinsurance business has the cat business which is probably the best performing sector in the market which the insurance business doesn't have the benefit of it. It's a bit unfair to try to compare one with the other. One strategy is working, but I would say reinsurance folks are in the double digits and our insurance folks are a bit less than the 9%. But, we will look at combined company and those things will change depending on where the market it.

Understood. My last one is, I think John, when you were commenting about RMS 11 and the different drivers, I guess what surprised me is some data I saw recently. It seems most of the companies are talking about the storm surge and the inland wind being the main drivers. What I had seen was that the actual biggest driver was this change in the short-term environmental impact, which I guess surprised me because I didn't think there was anything new in the science that would have made that worse since they introduced that five years ago. Is that consistent with what you guys see when you look at it and if so, is that where your biggest concern is?

It's not really a concern on the frequency. AIR had already published documents and papers on this and RMS is now agreeing with the magnitude. I think people agree that the warmer seas surface temperatures, which is a cyclical fact, there is different reasons why people think it's happening but it has happened and the data is quite clear that there is a higher frequency when the sea temperatures are warmer. There are other factors that can also drive it as well, the El Nino effect and things like that. It's really the magnitude, I think, by which this frequency, where you come down versus the long-term trend and that's been subject to some interesting discussions and debate on that piece. I think our teams are coming down that we think we will be a more frequent event of these storms while the sea temperatures are higher in the cyclical time. But you've got to go backtrack against all of your data and really come to grips with it. That's what is going to be -- all three of these are going to be interesting as companies across the industry analyze this and implement this.

Fair enough. I certainly agree that there should be some component for it. It just surprised me that they really hiked it up a lot. It doesn't seem that anything has really changed from what we learned in '06, and if anything, from what I understand, some of the scientists they work with have now refuted what they put out in '06. Just surprising that was the main driver and it seems the most flimsy. There is not the data to support it like there is on the wind speed or the storm surge, is that fair?

Ian, you will never get enough empirical data to feel 100% comfortable, that's why you've got to make judgements on these things. You can get very smart people on both sides of the table and they can have very good arguments and they both, by definition they both cannot be right, but the arguments can be very logical. That's why it's taking us some time to understand it and implement. It will have implications as to what happens to us as a company and what happens to the marketplace.

I would not say that one is dramatically that much more important than the other. It depends upon where you are and where your coverage is at. Where are your exposures. These can have different impacts depending upon where you are.

Got it. That makes sense. I think that's all I had.

I would just like to clarify on the prior question about our exposures and peripheral European countries, we do have that -- we still have the $35 million Santander covered bonds in our portfolio.

Your next question comes from the line of Vinay Misquith with Evercore Partner. Please proceed.

Question on the property cat business, since capital is rising 10% to 30%, those are the requirements, and pricing is up maybe 8% to 10%. How do you look at the profitability of the business, has the ROE actually improved and how does it impact your decision to write less or more business?

Well, there is two areas that you've got to look at, one is your absolute tolerance. That is internal tolerance that says is 25% of equity capital. That might only be affecting one or two or three zones. This is a big world. The second part is where do you see price improvements, especially in areas that you did not have significant participation in the past and have nothing to do about your capacity from a balance sheet point of view, to engage those areas but you just didn't like the price. There is a lot of room for us.

Depending what happens with pricing, in a lot of zones, Japanese quake, South American quake, New Zealand, Australia, European wind, et cetera. In all those zones we are still significantly under leverage, or under weight from the areas that we getting close to our risk tolerance, is mostly in the United States. Northeast, Florida, maybe the Gulf coast. You've got to look at it from that perspective too. Sometimes individual pricing gave us more opportunity to write in parts of the world that we were not a significant participant in the past, and we took advantage of it.

Sure, had some of the increase in the property cat reinsurance this quarter been from Japanese covers and non-US covers?

Some of it was from international covers, some of it was from the US, but some of it was from international covers, yes.

Okay. The second question is on more general pricing, where do you think we are in the cycle right now? And your retentions are rising, just curious what is happening on the competitive landscape.

Well, no different than what is in the prepare remarks. I think we have pretty much bottomed out. I don't think there is significant pressure for further deterioration with the exception of a couple of lines. Having said that, in some lines, getting 1% or 2% increases doesn't even keep up with trends. You are losing ground on that basis. Our retentions for us is because -- not because we like the business more, you've seen a change in our retentions because in the shift on the book of business. The more we write in small accounts, which we keep 100% net, you will have an effect into the net to gross numbers. As for where we buy the most reinsurance, it is on the large accounts the Fortune thousand type of accounts and as we write less of that, you see the changes in the retentions. Part of the retention is not a change in strategy to buy reinsurance in a different fashion, but is more as the outcome of the shift in the book of business that we write.

That's great. Thank you.

You're welcome.

Your next question comes from the line of Ron Bobman with Capital Returns. Please proceed.

Hi. Evercore's new star stole one of my questions. I have a question about Japan. The severity of the quake I'm told was not part of sort the data set that the modelers had previously put forth, and I guess sort of the knock on affect of it actually being offshore quake, yielding a tsunami, and the nature of the losses, is there any sort of lessons learned or just observations that sort of your take away as a result of the Japanese quake and how losses are unfolding and the types of companies that are incurring losses at the insurance and I guess reinsurance level? Thanks a lot.

There are lessons learned, I think everybody, including us, was surprised about the devastation a tsunami can have. You have seen the pictures and you have seen the force, et cetera. I think at some point in time all this will get recalibrated into the model. It never really changed our view that we always view Japanese cat pricing to be inadequate. For that reason -- we were patient for many years because in the years you had no events, any price looks adequate. You take in premium and you have no losses so it looks adequate. It's only when the event happens. I think we benefited of having way below our norm tolerance for risk in Japan because of pricing.

Has the pricing come up to the level that it makes us want to go in with both feet? Not yet. We have very good corrections in the right direction, but also I think not everybody has digested the total loss that is going to emanate from that because the business interruption and contingent business interruption issue we still don't believe it is a big issue, but it still has not been settled. We are going to find out in the next two to three years as it evolves. I think you will see over time readjustments as to how those covers get priced. And if it gets to the point that we think is attractive, we have plenty of capacity to commit. Right now, I think our appetite for Japan is limited.

Thanks, another broader quake related question. With the -- some degree of probability of aftershocks producing meaningful insurance losses after an initial quake, how does anyone really get comfortable -- how does Arch get comfortable even entertaining participating on a program renewal that suffered a quake loss during its prior 12 month contract life?

Well, it's a complicated question. You do anticipate aftershocks. It is not proven to us that the aftershocks do significantly more damage to the existing damaged buildings. Usually there is a lot of damage that happens when the quake happens, and when a structure has significant damage and aftershock is not going to change your loss potential. A lot of these buildings, they get declared total losses and have to be demolished and have to be rebuilt. The aftershock doesn't have a major effect.

Our underwriters have empirical data and statistics, and we know less about quakes than we know about wind. Wind is a much better area, but in essence, you've got to go with the long historical averages. People who say that I wrote this cover for half and I don't want to write it for 2x because I'm afraid of an aftershock, it sounds illogical to us. Unless they have significant scientific evidence that an aftershock is going to happen and it is going to be significant and it will damage new exposures.

Thanks a lot. Great job as you've always been doing.

Ladies and gentlemen, that does conclude our question and answer portion of today's call. I would now like to turn it back over to Mr Dinos Iordanou for closing remarks.

Thank you all for listening to us. We are looking forward to talking to you three months from today.

Ladies and gentlemen, that concludes today's conference. Thank you for your participation. You may now disconnect. Have a great day.