Markel Group Inc (MKL) 2005 Q3 法說會逐字稿

Good morning, ladies and gentlemen. Welcome to the Markel Corporation earnings conference call. At this time, all participants are in a listen-only mode. A brief question-and-answer session will follow the formal presentation. (OPERATOR INSTRUCTIONS). As a reminder, this conference's being recorded.

It is now my pleasure to introduce your host, Mr. Steve Markel, Vice Chairman of Markel Corporation. Thank you, Mr. Markel; you may begin.

Thank you very much. Before we begin, I would like to call your attention to our Safe Harbor and cautionary statements set forth in our 10-Q and 10-K. Our discussion today could be affected by the matters described in those statements, and we encourage you to read those statements very carefully. After I make a few brief introductory remarks, Richie Whitt, our Chief Financial Officer, will review the financial results for the quarter and year to date. Tony Markel will discuss our operations -- Tom Gayner, our investment results. And then finally, I will make some additional comments and moderate the question-and-answer period.

Markel's third quarter and 2005 results are clearly marked by the hurricanes. Our net loss for the quarter is $111 million. And for the 9 months, we have net income of $25 million. One of Markel's real strengths is our diversified specialty insurance business. Our catastrophe-exposed property businesses only represent something less than one-third of our total premium volume. And while the impact of these storms is substantial, the other two-thirds of our business is doing quite well.

With a solid mix of business and a very strong capital base, Markel is well-positioned for the future. As Richie and Tony will discuss, the storm losses were larger than we would have expected. And as a result, Markel will be adjusting our underwriting and risk management processes. Our financial goal continues to be to compound book value at a high rate and over a long-term. In earnings, an appropriate rate of return on capital for our catastrophe-exposed business is very much a part of plan, and I will talk about that a bit more later as well.

Without going further, Richie, why don't you jump into the results?

Thank you, Steve. Good morning, everyone. I do want to point out that our 10-Q was filed yesterday, as Steve said. If you have not seen it, we clearly encourage you to review that thoroughly.

As usual, I will focus my comments on the year-to-date results. Our 9-month results were heavily impacted by hurricane losses. 9 months 2005, we incurred 254 million of hurricane losses versus 80 million of hurricane losses in 2004. Giving a bit more detail about our 2005 hurricane losses, our estimated loss on Hurricane Katrina is 206 million pretax; that's 134 million after tax. We have not changed our original estimate for Katrina. And while it is still early, all indications are that it is set conservatively.

Our estimated loss on Rita is 48 million pretax; and that is 31 million after tax. While it is even earlier than it is for Katrina at this point, we do believe our estimate is set conservatively for Rita as well.

Tony will go into more details regarding our hurricane losses in a few moments. I'll now take us through some of the captions on our profit and loss statement. Gross written premium was down 7% for the 9 months of 2005 for several reasons. The first is our sale of Corifrance in January of this year, which led to a decrease in gross written premium. Secondly, our re-underwriting and discontinuance of certain lines of business in our investors' underwriting management division has also contributed to the decrease. Certain discontinued lines at our London Market operation, most notably aviation, is contributing to this decrease. And then finally across all our lines of business, increased competition in the property casualty market led to a decrease in our gross written premium in the first 9 months of the year.

Net written premium is also down 7% in the first 9 months of the year. A large portion of this is due to 54 million related to the hurricane losses in 2005. The remainder of the decrease is due to lower gross premium volume. It is worth pointing out; our retentions of gross written premium actually are up before considering the impact of hurricanes. The increase is 80 to 85% retention from an 82% retention for the 9 months of last year. That increase in retention is primarily to due to our purchasing of less reinsurance in our Excess and Surplus Lines segment in 2005 versus 2004. Earned premiums were also down in line with the decrease in gross and net written premium. This again is due to the reinsurance reinstatement costs associated with the hurricanes and the lower gross and net written premium volume over the past several quarters.

Adding it all up, our combined ratio for the 9 months was 109 combined versus 97 combined in 2004. The hurricanes added 18 points to our 2005 combined ratio compared to 5 points in 2005. The improvements before hurricanes to 91% combined from 92% combined in 2004 was primarily due to improved underwriting results in our London Market operation. I think this also demonstrates what Steve was saying in that the diversification of our business across over 100 product lines that the lines that are not exposed to property and cat business are performing very nicely.

Our 2005 results also included a charge for asbestos of 31 million versus the fact that we had no charge for asbestos last year. We completed our annual asbestos review in the third quarter of this year. We noted in that review increased severity in asbestos claims and as a result responded to that in the third quarter.

Turning to our investment results, average invested assets increased to 6.4 billion from 5.6 billion in 2004. Our total investment return was 2.2% on an annualized basis versus 4.0% on an annualized basis in 2004. This decrease was primarily due to 93 million of unrealized losses on our investment portfolio. This 93 million unrealized loss was approximately equally divided between our equity and fixed income portfolios. This decrease in our unrealized gains in terms of total return was partially offset by a larger portfolio and higher yields on our portfolio.

Operating cash flows for the first 9 months of this year were 429 million versus 495 million last year. This decrease was primarily due to some commutations that we reported in the second quarter and some additional ones we did in the third quarter and the timing of tax payments in 2005 versus 2004. Tom will go into greater detail in our investing results right after Tony.

Adding up our underwriting and investment results, net income for the 9 months was 24.8 million versus 115 million last year. While we are certainly disappointed to incur 254 million of hurricane losses in 2005, we believe being able to withstand and remain profitable is a testament to the strength of our organization. We reported a comprehensive loss of 46 million for the 9 months of 2005 versus a comprehensive income of 122 million in 2004. This decrease is due to obviously the hurricanes and the unrealized losses in our portfolio previously mentioned. Book value per share was approximately 163 million at the end of September versus 168 million at the end of December '04.

At this point, I'd like to turn it over to Tony to discuss our underwriting operations. Tony?

Thanks, Richie. As Richie indicated, the hurricanes have emasculated what would have otherwise been a banner year from an operational perspective. As we reported, the quarter and 9 months produced combined ratios of 149% and 109% respectively as opposed to 92 and 91 had it not been for the effects of Katrina and Rita. Simply put, these two events turned in what looked like $130 million pre-hurricane underwriting profit into $124 million underwriting loss. Clearly, a defining series of events that deserve elaboration, so I'll try to anticipate your questions in this report.

Let me start off -- number one, how confident are we regarding our loss estimates for the two storms? Although at this stage, no one can ever be 100% confident, we deliberately waited to report our Katrina and Rita losses until we had a reasonable basis for making an estimate. We are very comfortable with our approach, having done a policy-by-policy analysis on every exposed segment except for the smaller non-stationary properties like motorcycle and watercraft. That said obviously however, it is still early particularly as to Rita. So we continue to monitor the incoming claims on a day-to-day basis.

Second question -- what went wrong? How could you have sustained a net loss of the magnitude of the Katrina loss? First of all, let me assure you that we did not set up that program from a reinsurance standpoint to digest that type of loss knowingly. Our losses far exceeded our targeted worst-case scenario, which was set at roughly 75 to $100 million from any one event. In spite of a disciplined, comprehensive adherence to the fundamentals necessary to use the cat model that we employed effectively and the resulting purchase of cat cover that we thought was well north of conservative loss estimates from the model and recognition of the obvious potential fallibility of the model, several things went wrong. One, we naively relied far too heavily on one governor, the cat model, which unfortunately dramatically underestimated some idiosyncratic and systemic issues, including the expense flooding of New Orleans. And we provided some limited, select coverage on some policies, the strength and magnitude of the storm surge, which was clearly underestimated by the models, to the full extent -- and the full extent of the resulting expected demand surge, given materials and supplies necessary to make the resulting repairs.

Thirdly, third question. What changes are being made in your cat modeling techniques and oversight to avoid this type of surprise in the future? Well first, we are working with our current cat modeling vendor to determine whether their models can improve. And obviously, they are running around feverishly at this stage. Second and much more importantly, we're in the process of building our own conservative models using the comprehensive data that we already capture to act as an additional point of stress testing. Third, although we thought we were buying cat coverage extremely conservatively, the lessons of Katrina dictate the need for even more cushion between the expected loss and the purchase protection.

Next question. What effect will these losses have on marketplace as well as on Markel's response? Well, it's too early for us to predict the full impact on the market, but we think that the market will provide an extremely favorable platform for appropriate risk selection changes, reduced line sizes and dramatically higher prices. Although our underwriters are not waiting for word from us and are reacting appropriately, we are in the throes of finishing a product-by-product assessment that will lead to a corporate-wide mandate for aggressive change. Although still speculative, we anticipate the following -- reduced property and marine line sizes; substantial pricing increases on all wind-affected properties; large but less substantial price increases on wind-exposed properties in unaffected areas; moderate increases on all non-cat properties or at the very least a stoppage of the recent rate erosion; moderate increases on earthquake-exposed properties, given the obvious questions surrounding cat model adequacy; significant rate increases on marine and energy exposures. And on the casualty side, we expect at the very least a flattening of rates and hopefully some bounce back.

Although we think the market will prove receptive and provide the ability to achieve the new rate levels, there has to be one cautionary aside. In spite of the early indications of substantial increases in saber rattling in that regard, the market needs to be realistically challenged against not only the severity of Katrina but the frequency of the serious storms in the last 2 years. In other words, as we work on these issues to come up with adequate pricing to digest it and thus to project the expected storm activity for 2006, the driving question has to be -- how much is truly enough? And I can assure you that we will be looking at it from that aspect.

The next question -- what does Wilma look like? As noted earlier, we have been deliberately cautious in reporting estimates until we have reasonable information, and the same is true for Wilma. We have exposure, but it's too early to guess at our estimated losses. I will point out, however, that industry publications have estimated the overall loss from Wilma could be as high as $12 billion. And any loss of 10 billion or more would make Wilma the third most-costly hurricane in history behind Katrina and Andrew. This clearly reinforces my earlier point about the need for the industry to seriously examine and challenge assumptions about the frequency and severity of catastrophes going forward.

Obviously, there is still a lot of work to do in the aftermath. But we are extremely bullish about the resulting marketplace and the potential to make up the loss. In spite of the obvious and justifiable attention being paid to property cat issues, life -- as Steve indicated -- and the rest of Markel goes on. And the news is extremely good, as reflected by my earlier reference to the projected combined ratios that we were headed for before storm season.

In that regard, we recently announced a very exciting new corporate initiative, which is the hiring of Steve Cullen, through an experienced marine underwriter and manager to build a new U.S. marine and energy team and provide technical oversights and the growth opportunities in our existing marine operations in London and Richmond. We already have substantial presence in that marine and energy arena, but we think the opportunities for growth and even increased profitability have never been better. And Steve's addition puts us in a particularly unique position to take advantage of the current expected marketplace.

Last but not least on the positive side as well as we reported earlier, we were in the process of adding some expansion branches to our international operation. And we recently opened the three new retail branches in the UK -- Bristol and Cambridge, England and Edinburgh, Scotland, which followed previously-discussed openings in Madrid, Spain and Toronto, Canada -- both of which have already begun to write business. So we're dealing with the hurricanes in retrospect on every single one of the platforms. The rest of the business seems to be performing extremely well.

Obviously, I would be more than happy to answer any questions in the Q&A. With that, let me turn it over to Tom to talk about the investment side.

Thank you, Tony. Good morning. Today, I would like to spend a few more minutes than usual to cover three points about our investment activity. First, as Richie stated earlier, investment returns in 2005 have been minimal. I call it a year of hugging zero. Year to date, our total returns after adjusting for foreign exchange and investment expenses were 1%. Equities declined 0.2%, and fixed income gained 1.8%. Although these results are below our long-term results and future expectations, I am very excited about some of the steps we've taken in our investment activities this year and the outlook for the future.

To review, over the longer and more meaningful time horizon of 5 years, equities returned 10.9% annually and fixed income 6.4%. After foreign exchange and investment expenses, the total return per year for the portfolio was 7.1%. Given the difficult and tumultuous nature of the investment markets during the last 5 years, I am and I hope you are very happy with these results. Coupled with Markel's normal investment leverage and underwriting profitability, these sorts of investment results clearly contribute meaningfully to the net growth in book value goals that we have as an organization.

Equity markets in 2005 have been largely dominated by energy and commodity firms. These are not normal areas of interest for us as investors. And clearly, we suffer from the lack of exposure to them in 2005. Over the longer term of 5 and 10-year periods, our focus on areas of human and intellectual capital as opposed to physical capital has served us well. Over the last 5 years, the S&P 500 is actually down 1.7% per year, while we are up 10.9% per year for this timeframe. That degree of out-performance is usually good, and I would not look for that in the future. But I think it validates our basic investment discipline and approach.

I'm quite optimistic that the values we are finding in the equity markets today will produce good returns over the next several years. The list of dominant companies with wonderful economic positions selling at the lowest-priced earnings ratios that they have in years is long enough to give us plenty of attractive investment opportunities. I'm confident we will reap the benefits of our steady investment in these areas over the coming years.

Secondly, I'm also confident that the overall economic environment will continue to improve as time goes on. Every day, I drink my coffee from a mug which says, "Doom and gloom sells, and optimism pays." The headlines and media coverage of our economy and the catastrophic storm losses is more negative than history suggests it should be. In a list of the deadliest natural catastrophes over the last 100 years, there is not one single entry for a hurricane or earthquake in North America or the UK. Our size, population dispersion, building codes and techniques, technological advancements and sheer good luck have caused our tragedies to take far fewer lives than what has occurred in some other parts of the world.

We value individual lives highly in these societies and the catastrophe statistic partially reflects this underlying value. Economically, the historical record shows that true capital does not reside in buildings or bridges or oil refineries or pipelines or natural resources but in the human mind. As such, I'm confident that we will recover from the current set of hurricanes and higher energy prices with greater efficiency and more economic vigor than before; that is the long-term record of our economy. And I don't see any reason to believe that it will be different in the future.

Third and finally, I'm pleased to report to you that Markel entered into an agreement to acquire a 40% interest in First Market Bank recently. First Market is primarily a supermarket bank with a focus on consumers and small businesses in the Richmond area. The remaining 60% of First Market Bank is owned by Ukrop's, a Richmond-based, family-owned grocery chain. The Ukrop family is dedicated to customer service and quality and has a long history of success. We think the match in values and financial performance is a wonderful opportunity for Markel.

Private equity and alternative investments are very popular subjects in the investment world these days, perhaps too popular. While we have made a few investments in private equity over the years, all of our previous commitments have been in the insurance area. The First Market investment marks a measured step into different realms, and we will look for additional opportunities of this sort in the future. We will use the same four-point criteria as we use in public investments, i.e., profitable businesses with honest and talented managements with reinvestment opportunities and capital discipline at fair prices. We'll also be consistent with our public practice of investing directly and largely in-house rather than through intermediaries to maintain very low costs.

We manage our total investment operation for a single digit number of basis points per year, and we expect this efficiency to continue in the future. When we compare our overhead expenses to the management fees and expenses associated with more traditional private equity in alternative investment programs, we are optimistic that this will be a fruitful arena for us over time. We will take measured and disciplined steps to further explore direct private investments, and we think that more opportunities will develop as the current high level of interest in the field fades during a different investment environment.

In summary, while 2005 results appear calm on the surface, we are excited about the underlying economic activities occurring in the investment department and we are excited about our return prospects in coming years. With that, I will turn it back to Steve.

Thank you, Tom. A few final comments before we open the floor to questions and answers. If you remember, it was all the way back in 1992 when Hurricane Andrew hit Florida and caused the insurance industries some $15.5 billion or 21 billion in today's terms. Over the next 12 years, hurricanes that have impacted the United States were in hindsight extremely modest. Undoubtedly though, during those 12 years as time passed and more and more people forgot the impact of Hurricane Andrew, insurance prices relative to exposures undoubtedly declined. In spite of whatever you may have heard about price increases, I'm confident of that statement. Clearly, the events of 2004 and now Katrina, Rita and Wilma in 2005 remind us of what the true exposures really are.

Undoubtedly now as we remember the events of 2004 and 2005, there will be significant price increases in the industry and Markel will certainly enjoy part of that. However, for Markel, we will do something in addition and something differently. We will make sure that we establish appropriate risk management processes and internal controls to assure that we will earn an appropriate rate of return on our capital over the long-term. I'm not certain that all of the new money raised in response to these events will do the same. And I think the recent evidence of capital raised in 2001 and 2002 proved that point. Hopefully, we will all learn something from these experiences and be better for it.

With that, I'd like to open the floor to your questions.

(OPERATOR INSTRUCTIONS). David Lewis, SunTrust Robinson Humphrey.

Actually, this is Eric Jackson (ph) calling in for David. You answered most of my questions. But one I had for you -- as you review your model and assumptions for cat exposure going forward, if adequate pricing can't be achieved, is it likely that premium growth could further be pressured due to the seizing of business?

No doubt about it. If we can't get the right price, we will write zero business.

Now some have speculated that increased hurricane activity in the Atlantic is now more the norm. Do you believe that is the case as you approach your cat modeling assumptions?

We at this stage, not tried to predict a change in frequency, and I'm certain all the resources at Markel would not prove to be accurate at trying to forecast future hurricanes. We'll just take a very, very conservative view when we set up our models and do our pricing them.

Beth Malone, Advest.

Could you comment on what role you think the rating agencies may play in causing pricing changes or demands on capital from the reinsurers in other markets and how Markel may react to those issues?

Markel is I think uniquely and in a very strong position because of the diversified product lines that we have and because of our strong capital base. But I think clearly, the rating agencies as do the buyers -- and it is really more important for the buyers to come to grip with this I think than the rating agencies -- but I think clearly the model of mono-line property writers, who operate on the old Lloyd's mentality that if you blow your capital away, you can replace it -- is not something that makes much sense today. I don't think it's reasonable to assume at least as a buyer of reinsurance that if our reinsurer loses money that the capital markets will bail them out. And so I think it's appropriate that we all -- rating agencies and buyers both -- reexamine those issues and those questions. Clearly, understanding your reinsurance companies' risk management processes could be part of that. How the rating agencies deal with it? You'd be better off asking them than asking me.

Initially though, the market seems to have been pretty receptive in recapitalizing or raising capital. There are several new reinsurers apparently that have been funded since Katrina.

You know Beth, as you know, people also flock to Las Vegas and Atlantic City and slot machines every single day, and that will continue.

In terms of the overall health of the property casualty and Markel's role in it, it sounds like you're planning to be a leader in terms of what you are willing to -- how you're willing to price your product regardless of what kind of less discipline may be demonstrated in the overall market?

Well, I wouldn't want to try to guess what others will do appear, but clearly our models will be based upon our ability to realize return on our capital. And if that means there are product lines we're not competitive in, so be it. The losses for the 2005 events are certainly north of $50 billion. And the new capital raised is a small fraction of that.

If you were to look at the exposed premiums for example, the total premium volume in all of the insurance -- all the property casualty insurance premiums in all of Louisiana and all of Mississippi is only $11 billion. If you were to look at just the coastal areas probably is only 2 or $3 billion worth of insurance premiums that were being paid to cover the coastline properties in Mississippi and Louisiana. And if Katrina's $34 billion loss or $40 billion loss -- pick your number -- clearly, a 50% rate increase would not come close to supporting a thesis that in the next 20 years, another Katrina could hit that same place.

So you can do the arithmetic and look at the numbers as easily as anybody. Fortunately, you can spread those risks over larger books of business, and that's the key is to make sure you have the appropriate diversification.

Steve, I might add, clearly, we will not allow the marketplace to force us into something that we're not comfortable with post-Katrina, Rita and Wilma relative to pricing. One of the strong advantages that we have and it's already been mentioned a couple of times earlier is the diversification of our product line, and it gives us the ability to frankly be pretty cavalier with regard to what we think is the appropriate pricing and walk away from market segments should we not be able to achieve them.

Dreyfus Noonan, Morningstar.

Tough quarter. I'd like to explore three things sequentially if I may, starting off with the risk on your books. Given that you said the loss models under-predicted the actual losses and you also said in the Q that you exceeded your reinsurance protection, does that mean that Markel now has a bunch of risks on its books so the data wasn't paid for or wasn't paid enough for?

I don't think that's quite the right way to interpret that. But clearly, the good news is the hurricane season is now behind us, and our reinsurance continues until next August -- is our scheduled renewal. So we will not have to deal with a renewal in January. Having said that, we had now three events, and some portions of that coverage is used. And so, as there is further development either in the existing losses or if there are new events, our coverage is different. We are closely monitoring those situations, and we're reasonably comfortable with where it is given the circumstances.

I think we also feel pretty good because in setting up our loss reserves, we tend to try to be reasonably conservative. And while additional development would have a negative impact and may not get the full benefit of further reinsurance coverages, likewise, reductions in losses and benefits that we might receive would fall completely to the bottom line.

So if I move on then and ask a related question about the reinsurance then, is this going to have any effect on your plans to increase your retention? I'm especially wondering if it means that the increases in reinsurance pricing might change that. And also, I noticed that I think you've had about 150 million in premium through your reinsurers and you're expecting to get about $570 million back. Do you think you're going to have any problems getting the reinsurers to pay up?

Actually, we have looked at the reinsurance accounts over the last 5 and 10 years. In reality, our reinsurers continue to be over the long-term in pretty good shape because of the nature of the way our coverages have been placed and the fact that prior to 2005, in 2004, we really did not have meaningful losses to our reinsurers despite the fact that there were four hurricanes. And so the reinsurance market is clearly taking some big shots, and there will be significant rate increases I think. But on a relative basis, I think Markel is in fairly good shape.

More importantly, what I think we've learned and we still have a little more work to do on this is that I think our reaction is more likely to treat our property catastrophe writings more like we treat our other lines of business, where we write smaller lines but keep a larger share of it for own account. And so the property account where last year, we may have written 100 million in premium and ceded $75 to our reinsurers, keeping only 25; next year, our total writings might drop from 100 to 50 or 75. But we might increase our share of it by having more smaller risks in a different dispersion so that we're actually depending less on reinsurance. And I would guess that that might well be the way we would move our property exposures.

Steve, I might add in that regard that I don't think we're going to change that posture. Our posture has always been to be the least dependent upon reinsurance as we can practically be and to push the retentions to the fullest extent that we're comfortable with, and we will continue that. I think as I mentioned earlier and Steve just sort of alluded to it, I think the resulting market after these hurricanes is going to create smaller lines, which is going to necessitate less reinsurance which plays right into our strategy. So the net result is, I think we will be less dependent on reinsurance but no more vigilant about retaining more and ceding less than we have been in the past.

So I will take that as more of the same on the reinsurance front. And if I can squeeze in one more question, I am looking at the combined ratios for the London business. And I noticed that even after -- excluding the hurricanes, those businesses still aren't generating underwriting profits. Is there any reason for that?

I'll let Richie speak to that.

I think what's really happening there and we have talked about it in the past as well, I think we're very comfortable with the business that we are currently putting on the books. We believe it is at an underwriting profit. Clearly, that comment is before the cat-exposed business and what happened last quarter. But we continue to be conservative. We continue to try to establish a margin of safety. And as a result, we haven't gotten below that magical 100 combined level yet.

Gail Golightly, Wachovia Securities.

This is sort of the season for Lloyd's capacity decisions, and I was wondering if you could share with us what your thinking is about capacity in Lloyd's, particularly relative to the international expansion that you have been building out.

Our capacity at this point is already set with Lloyd's; the business plans had to go in a few weeks ago. And we have based on what we knew at the time set capacity flat. Obviously, we will see how the market shakes out. And depending on which way it goes, we will adjust accordingly. But at this point, capacity is basically the same as last year.

Richie, you might also -- I might also add that half of that business in the international operation is placed in the Markel International Insurance Company as opposed to Lloyd's. So Lloyd's only, there are only roughly 50% of that business written over the season is actually going into the Lloyd's platform.

In the vein about your reinsurers and the whole issues about whether or not some of these companies have been stressed, have you been looking at your recoverable situation, which is up because of the problem in terms of collateralization, security? Are you comfortable with that position?

We are comfortable, Gail. We did look at it very, very closely in the context of these storms and continue to look at it and monitor it. And we are very, very comfortable.

My last question is, have you had any discussion with the rating agencies at this stage and have preliminary take from them on the size of the hurricane losses?

Richie, I'll let you --

We are constantly talking with the rating agencies, as I think most people are. I think at this point, they are aware of our losses on Hurricane Katrina and Hurricane Rita and I think are comfortable.

(OPERATOR INSTRUCTIONS). Charles Gold (ph), Scott & Stringfellow.

Thank you for a very informative and candid call today; it is very much appreciated. My question comes from a gentleman farmer in Stanton, Virginia, who is a shareholder and long-term shareholder. Richie, you said that when you set the reserves for Katrina and Rita, they were set conservatively. And then Tony said he was very comfortable with the approach that you took. And then Steve, you came back with -- we hope we've set them at a higher number than actually transpires.

So the question is -- last year, you had hurricane losses and I don't have the 10-Q in front of me but I just remember something like $80 million. Has enough time transpired to tell us how that number worked out in actuality?

Pretty close, Charles. I think there has been a very, very modest amount of slippage but not a meaningful amount. I think there's still some open claims as well. But the slippage has been almost insignificant.

The other thing I would say, Charles, is obviously we learned lessons last year in how we performed in those hurricanes. So we took some of those lessons learned as we evaluated these hurricanes, and I think probably put a better shot together in terms of coming up with our estimates.

So in estimating, you've made these numbers even more conservative than the model you would have used last year to make the numbers (multiple speakers)?

Well, it's hard to say that. But I would suggest -- we took what we learned last year and the sorts of things we saw coming out and the surprises we had, and we tried to learn from them.

But let me underscore, we did a pretty good job last year. Steve's comments concerning moderate slippage is right on point.

And seeing some companies lose 100 or 200% of their equity -- I see today that Montpelier is getting hit, even though I know that it's a different category -- to have roughly 100% of your equity intact at the end of the year with a hopefully 1 in 100 year storm, you all deserve kudos for keeping our equity intact. And we appreciate all your hard work. Thank you.

Well, thank you. Our goal though is to comp down at a high rate over a long period of time. So we need to do better.

Johnny Walker Blue believes in you.

Chuck Aiker (ph), Aiker (ph) Capital Management.

I am a little confused about one issue, and that is the discussion about choosing not to increase your reinsurance coverage beyond what you have done with the reinstatement on the one hand and on the other hand, having this expressed need to make sure that you are perhaps more adequately covered than you were this go around and all within the language of the discussion that your losses exceeded your reinsurance coverage. So I'm just a little confused about that whole issue.

Well, it's a confusing subject, so I'm not totally surprised. We tried to clarify it as best I can. I'm not sure exactly how to go about doing that. But I think there are two or three different issues. First, where we stand today is that as it relates to our primary exposures, which would be hurricanes in the Southeast -- the hurricane season is over. And between now and next June, we will very diligently re-examine all of our exposures and in the meantime be changing them and pricing differently and moderating them so that as we enter next year's hurricane season, we will be well-protected and we will have a model in place, where we will earn appropriate rates of return on our capital relative to the exposures we have.

As it relates to the longer-term -- and this is a little speculative at this point in time -- but my sense is that we will write a larger number of smaller risks and retain a larger percentages of them. So we will become less if you will reinsurance dependent in our property lines -- very much like we have been in our casualty lines. And the economics of that will make more sense than trying to arbitrage and leverage the reinsurance market. Time will tell. And it varies a lot from product to product to product. Each of our products will be under the microscope, and we will need to look at that.

In the meantime, we will enjoy significant price increases both in real terms and in terms of per unit of risk. We will continue to look for the appropriate geographic dispersion, which I think quite frankly we already have a good example of. Over time, I think dependence on reinsurance will be less, not more.

I have a follow-on. The original estimates of the loss, you began your conversation today by saying have been in the neighborhood of 75 to 100 million. Certainly, a conversation that you and I had 2 or 3 weeks ago indicated that you have revised those estimates to something that was in order of 150 million or thereabouts. And I noticed that that number reasonably coincides with what you described today as the after-tax loss for Katrina and Rita. So are we still talking about the same number then?

Again, I'm not 100% sure I followed your question but --

2 or 3 weeks ago, sort of what I'd call the first revised estimate of losses raised them from the $75 million range generally to a number that was twice that -- around 150 million. And today's losses -- an announcement of estimate of losses today is on a gross basis, pretax basis, significantly above that but on an after-tax basis about in that neighborhood. What I'm asking is, is today's estimate yet another change? Or is it consistent with the revised number from about 3 weeks ago?

I'm going to let both Richie and Tom make a stab at this. Because I think they understand better what you are driving at than me. But first, let me make a comment. In the big picture, we have said in the past that losing one quarter's worth of earnings would be an appropriate kind of expectation for sort of a normal run-of-the-mill catastrophe. That would equate to roughly 5% if you will of our capital base or on an after-tax basis somewhere in that 75 to 150 million that we have been talking about.

But the Katrina loss after tax, we estimated when they made the report a month ago between 125 and $150 million. And that number was after tax, and that's consistent with where it's reported today. We have not changed that estimate at all. It is after tax still $134 million, which is right in the middle of that range. In fact, we think that number is very, very solid. So the problems are that we have had more than one event in one quarter, and that Katrina was a storm that we would have thought would not have been anywhere near as big as it turned out to be. But having said that, let me both Tom and Richie go throw in their two cents worth. Or Tom I guess you had --

As you added that comment, you addressed what I was going to say. And that is -- I cannot remember the timeline of number of weeks. But clearly before all this hit, as Steve said, we would design a program such that we expose roughly one quarter's worth of earnings. So that was the estimate that all of us had in our minds before we had any data. The first number that we put out and the after-tax numbers that Steve and Richie referred to, those have indeed helped. So nothing is changing on the development from the initial estimate that we put out once we put out an initial estimate.

I think going forward, the standard will be very similar. The methods of measuring it and hopefully the future results when we have future events will be more consistent with what we forecast. But I think going forward, we will continue to feel like we can and it's appropriate if we earn the right rates of return be willing to expose 5% of our capital to a very significant event.

Chuck, Steve, I might just add -- maybe my comment earlier relative to our continuing target pride (ph) of things to have a no single event hit us for more than 75 to $100 million net before tax -- has brought something into this discussion. And that was our -- we thought we had our reinsurance program, given the modeling on a very conservative basis using a 1 in 250-year event that would have kept the worst-case scenario in any single event below 100 million, and that was what I was alluding to in terms of the way we manage our catastrophe exposures. And of course, it also singles relatively the surprise that the numbers on Katrina have given us compared to what we would have modeled in a worst-case scenario.

Right, and so for Markel because just like others in many businesses, you've had an event here which in effect fell outside the models.

(multiple speakers) That's correct.

And you know, we've seen that happen in all kinds of other businesses with much more drastic consequences. And fortunately, Markel manages its balance sheet so that the result that's let's say twice as bad as your model would suggest the worst could be, still presumably doesn't affect your need to raise capital or your need to -- or the action by rating agencies to downgrade you, and I think that is terrific.

It does raise the question though, as we've had encountered -- I go back to New York contractors for example, where over a period of years, you kept having a fallout of experience outside the balance of what was expected. And now, we have a fallout of experience outside the bounds of what was expected, and it goes to the issue of how difficult it is to manage this business and to me reinforces the wisdom of running it with a very conservative balance sheet. I applaud you in that regard and wish you the best.

Thank you, Chuck, and I think your point is absolutely valid. We were in 1992, amongst those companies who entered the property catastrophe market, taking advantage of what was perceived to be higher prices in the wake of Hurricane Andrew. And we have not on a first-hand basis experienced such an event since then. And I can guarantee that Markel for one will learn from the experience. And I suspect there are others out there, who will require the future experience to be their learning experience.

Nicolas Prendergast, BB&T Capital Markets.

I'm actually calling in for Kenneth Billingsley. I've got a couple questions here for you. The first one relates to the A&E review. When I am reading through the 10-Q, it says there's an increase in allowance for uncollectible reinsurance due to the potential for disputes with the reinsurers. I was just wondering, is there currently a dispute? Should we be expecting one in the future?

No, no. It more relates to the fact when we're talking about asbestos reserves, you are talking about contracts that were in place '86 and prior. A lot of those companies either had merged into other companies or in runoff in a lot of these instances. As a result, the reinsurance you might have placed back in those days is to some extent impaired at this point. So your loss will as a result have a component that relates to reinsurance you just won't be able to collect.

Also one more question. Given the current events, do you expect a slowdown in the share re-buybacks?

We have not repurchased any additional shares since the hurricanes and since really waiting for this earnings release. We in the next 24, 48 hours will have an open door to reexamine those issues. My sense today is that we will get a better handle on the impact of these storms on our forecast for 2006 before we aggressively buy back stock. But if we find that our price levels are going to be pricing us out of the market, if we are more aggressive in pricing and underwriting than others and find that we cannot get the right price for these products, the result would be that we would probably have excess capital.

If in fact we find there to be interesting and meaningful market opportunities, hopefully, you will see our premium volume grow and that issue will become less relevant.

Tom Cholnoky, Goldman Sachs.

Just two questions. I guess one, given that you are going to be obviously looking at your cat exposures and the fact that I might believe something a little bit differently in that maybe Katrina was not a 1 in a 100 or 200-year event; we had a very significant reinsurer basically say that this was a 1 in 20-year event actually and that we are in a period of heightened catastrophes. Given what you are doing and let's say another Katrina were to happen next year, how aggressively can you reduce your exposures and what kind of loss would you potentially have if another Katrina happened next year? And then I have another question.

You know, we have a fairly good period of time between now and the next hurricane season. So I would suggest that whatever the exposures are today, we could easily reduce them by 50% in the next 6 months. That might not be 100% accurate, but given the fact that there is 7 months between now and the start of next hurricane season, that's not a -- really the 8 months, that's not an unrealistic expectation. So I'm not worried about being prepared for the next year's hurricane season at all. We will be prepared, and we will have our portfolio in a very good position.

Steve, Tom's comment concerning somebody alluding to this thing being a 1 in 20-year event speaks exactly to the comments that you and I both made about not necessarily getting euphoric about some of the rate increases that are banging around until you sit down and analyze what you really think realistically could happen in the '06 wind season.

You know Tom -- and your point is right on. Andrew and I don't have the exact data with regard to category and wind speed, but my guess is that if Andrew had hit Miami instead of Homestead, instead of it being a $20 billion in today's terms, it would be equal to Katrina or maybe worse. I don't have the data to be 100% certain about it, but I would say that it could very easily be, given we had Andrew 12 years ago or 13 years ago, it is a 1 in 13-year event.

Yes, I agree.

So I agree with you. The difficulty though, is you do have a couple 1,000 miles worth of coastline and you don't know in which 100 or 200-mile segment it is going to hit. And as we know with Wilma, if it hits 100 miles on one side of Florida and exits on the 100 miles on the other side of Florida. So I'm not sure what the appropriate banding is anymore either. But clearly, the arithmetic could be different. And the model that you can lose 100% of your capital and go to the capital markets and replace it, clearly is suspect. And likewise, I think the model of operating in a tax-free environment is suspect.

In Markel's case -- and we're talking at some length in a minute ago with Chuck Cholnoky about this -- but the tax benefits or the tax impact on Markel is real. And the non-taxable environment of Bermuda, the arithmetic is different. They have a benefit in a no-loss year because they can compound the money tax free. But it sure is the Dickens (ph) are not getting tax-affected benefits 2 years later.

Steve, your comment concerning correcting the book is a very valid one. We are in the final stages of coming up with our corporate mandates relative to response post-hurricane. And we expect to get that implemented pretty quickly, and we have now got realistically 10 months between now and the start of next year's hurricane season. So implementing that relatively soon will frankly mitigate and minimize being caught with exposures we don't think have been adequately selected and price based on our knowledge from this season.

And then just a question on investments, Tom Gayner. We are continually facing a fad that continues to tighten. The 10 year is stubbornly stuck; seems to me doesn't want to move higher. Ultimately, it will probably pop up. In that kind of environment where you continue to face greater pressures from unrealized losses in your bond portfolio, how should we think about your ability to continue to generate the kinds of returns that you have been able to generate with that kind of headwind?

Clearly, this year that headwind has already started. So far, it's been relatively gentle. And the fixed-income return is less than the coupon because you're giving it back on price a little bit. And I kind of think that's going to continue for a while. We have been short relative to our book of business for 24 months because we've been worried about this happening. It is starting to happen; we continue to be short. And that's all you can do. The returns are what they are.

When you have a high-quality, minimal credit risk, kind of fixed income portfolio and I think the mistake would be to try to override that force of nature and stretch for yields. The only way you can do that is going down the ladder on credit quality, and we are not going to do that. We would rather just fight the headwind until the rates stop going up, then stretch for yields by accepting lower credit quality in the fixed-income portfolio.

So how should we think of is a reasonable target for your overall returns over the next 2 or 3 years?

Over the next 2 or 3 years, they sort of normalize out. If you talk about 4.5, 4.60 on the 10-year and you add a little bit for buying some immunities that have tax efficiency and a little bit of corporates mixed in there, maybe you get to a 5% kind of number on fixed income. And that seems eminently reasonable to me.

Brent Barton (ph), Evergreen Investments.

Have you guys had any contacts with the rating agencies or a need for additional capital after these losses?

I'm sorry; would you repeat that?

Have you had any contact with the rating agencies or the need for additional contact -- additional capital to maintain your ratings?

No, have not.

Dreyfus Noonan, Morningstar.

I just realized no one has asked you about our friends in Toronto. I see they raised another 300 million this quarter. Did Markel participate again this year and on the same terms as last year?

No, we did not buy additional stock in that transaction.

I also just noticed that you say in your Q that you need to make some capital contributions to a subsidiary. And it looks like from reading the Q, that's not going to cause any liquidity problems at the holding company. Is that correct?

That is correct. It will cause no liquidity issues.

Ladies and gentlemen, there are no questions at this time. I will now turn the conference back over to your host to conclude.

Thank you very much. We appreciate your long-term support. And if you have any further questions or comments, we'll be happy to hear from you. Thank you very much.

Thank you. This concludes today's conference. Thank you all for your participation.