Arch Capital Group Ltd (ACGLN) 2007 Q3 法說會逐字稿

Good morning, ladies and gentlemen, and welcome to your Q3 2007 Arch Capital Group earnings conference call. My name is Mike. I will be your operator today. At this time, all participants are in a listen-only mode. We will be taking questions at the conclusion of today's presentation. (OPERATOR INSTRUCTIONS).

Please be advised, 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 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 results may differ materially from those express or implied. For more information on the risks and other factors that may affect future performance, investors should review periodic reports that are filed by the Company with the SEC from time to time. Additionally, certain statements contained in the call that are not based on historical facts are forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. The Company intends the forward-looking statements in the call to be subject to the Safe Harbor created thereby.

Management also will make reference to some non-GAAP measures of financial performance. The reconciliation to GAAP and definition of operating income can be found in the Company's current report on Form 8-K furnished to the SEC yesterday, which contains the Company's earnings press release and is available on the Company's website. I would now like to turn the presentation over to your host for today's call, Dinos Iordanou and John Vollaro. Gentlemen, please proceed.

Thank you, Michael. Good morning, ladies and gentlemen, and welcome to our third-quarter earnings call. As you can see from the numbers, Arch had another great quarter. Both of our businesses, insurance and reinsurance, performed extremely well, but more importantly, they displayed the underwriting discipline that we expect in the current market.

Our book value per share, including the effects of share repurchases, increased to $51.34, a 17% increase from December of 2006. Our after-tax operating income was $2.89 per share, which represents a 24.2% annualized return on average common equity.

John will give you more detailed commentary on our financials in a few minutes, but before I turn it over to him, let me share with you some comments on general market conditions and how our two units responded.

The market is soft and is following the recent downward trends we have experienced in the past year or two with the trends varying by line of business. Rate reductions range from the low mid single digits to up to low double-digit for certain sectors. Large premium accounts continue to be under more competitive pressures than smaller premium accounts. Overall, our gross written premium declined by 9.6% for the quarter and our net written premium by 7%. By unit, our reinsurance group reduced its premium volume by 22% on a gross basis and 15.2% on a net basis, while our insurance group's premium volume was down by 3.7% on gross and 2.3% on net.

We are pleased with the returns generated by both our operating units and although our focus on maintaining margins resulted in a decline in premium volume, we are confident that the underwriting discipline exhibited by our units will enhance our ability to grow our book value per share at a very attractive rate over time.

Underwriting discipline will differentiate winners and losers over the next few years as we go through the next phase of the cycle. Our original business plans contemplated that, over time, insurance will become a greater part of our business because in times of heightened competition, the insurance sector has an advantage in selecting and pricing risk as insurers are closer to the risk. This anticipated change in business mix took longer than we originally thought, mainly because the cat events of '04 and '05 helped maintain favorable reinsurance market conditions, which we took advantage of. We expect our business mix to continue to shift towards insurance over the near term as we predicted back in our original business plan. Our robust insurance infrastructure and diversified distribution capability allows us to access a broad cross-section of business. As we continue to write more small to medium-sized accounts, our ability to underwrite the business locally is proving to be an advantage. As a matter of fact, our insurance group is planning to expand that capability with the opening of two new offices in early 2008, one in Philadelphia and one in Dallas.

From our beginning, we recognized that the ability to monitor underwriting activity was paramount in building an underwriting culture. Today, we believe we have the best-in-class rate monitoring systems for our renewal business. We also recognize that in a market environment characterized by declining rates, monitoring new business pricing is also critical to ensure that appropriate rates have been charged. Over the past year, we began developing a benchmark price monitoring system for new business. The system is already operational in certain units and is expected to be fully implemented in all of our units at the insurance group by January 1st, 2008.

We believe our price monitoring systems enable us to effectively allocate capital and human resources to the most attractive areas of the business. For example, small account business currently represents a greater portion of our premium volume than it did in '03, '04 and '05, and it was a result of this activity.

In addition to underwriting discipline, appropriate capital management is also a critical part of cycle management as it will not only enhance returns, but it will also reinforce appropriate behavior. In times when we can deploy effectively all of the capital in our businesses, we do so aggressively. While in periods where we accumulate excess capital, we intend to return it to its rightful owners, the shareholders. Given the current market environment, we expect to continue to implement our share repurchasing authorization.

Clearly, we are in a more challenging market, where underwriting and financial discipline will be key to future success.

Before I turn it over to John for a detailed discussion of our financials, let me update you on the level of our cat exposures. As we have mentioned in prior calls, we would like to remind you that in determining our one in 250 event PML, we include all exposure that we believe could accumulate in a given event, not just our property exposures. As of September 30th, 2007, our one in 250 year PML from a single event expressed as a percentage of common equity was approximately 18% and the Tri-County area in Florida continues to be our peak zone with Northeast wind and Gulf wind as our next two significant events. Now let me turn it over to John for a detailed discussion of our financials. John?

Thank you, Dinos. Good morning, everyone. From a financial standpoint, third-quarter results continued our terrific year with an annualized ROE of 24%. I'm going to briefly walk you through the key components of our financial results, starting with the top line, primarily focusing on the quarterly results included in the release. Dinos has commented in some detail on premium volume, but there are a few additional items worth pointing out.

On a trailing 12 months basis, premiums written by the insurance segment represented 64% of our gross volume and 59% of our net volume, while property, marine and aviation business represented approximately 31% of our consolidated net premium volume.

On a consolidated basis, the ratio of net to gross written premium in the 2007 third quarter increased slightly to 69% from 68% in the comparable 2006 quarter. On a reported basis, we ceded $51 million of written premium to Flatiron Re in the third quarter of 2007 compared with $78 million in the comparable 2006 period.

As you may recall, in the third quarter of 2006, the percentage of subject business ceded to Flatiron was increased for a brief period of time. This change was the primary reason for the difference in ceded premium to Flatiron on a quarter-over-quarter basis. Earned premiums ceded under this treaty amounted to $68 million for the third quarter of 2007 in comparison with $52 million ceded during the 2006 third quarter. The override in estimated profit commission recorded on the treaty with Flatiron are reflected as a reduction of the acquisition expenses of the reinsurance segment and improve the expense ratio of the reinsurance segment by 260 basis points in the 2007 quarter, while the impact on the expense ratio in the comparable 2006 period was 240 basis points.

Underwriting income for the third quarter of 2007 amounted to $1.55 per share, roughly equal to the $1.58 per share of underwriting income recorded in the third quarter of last year. On a consolidated basis, the combined ratio was 84.8 in the 2007 quarterly period, which was slightly higher than the comparable 2006 quarter. These outstanding results were primarily due to continuation of excellent underwriting results in the property, marine, and aviation business of the reinsurance segment, as well as the favorable development of prior-year reserves. On a consolidated basis, favorable reserve development, net of related adjustments to acquisition expenses, totaled $49.7 million in the 2007 third quarter compared to $24.4 million recorded in the third quarter of 2006.

The 2007 net favorable development was primarily attributable to short and medium-tail lines, with property, marine and aviation accounting for the most significant part of the development. Development on prior year catastrophe events was modestly favorable in the 2007 third quarter. In general, reported and paid claim activity across most lines of business remained at favorable levels and IBNR and additional case reserves continued to represent approximately 75% of quarter-end loss reserves.

On a consolidated basis, specific cat losses, including losses resulting from the flooding in the UK, amounted to $20 million during the third quarter of 2007 compared to $9 million of cats in the corresponding 2006 period. Pretax net investment income in the 2007 third quarter on a reported basis increased over the comparable 2006 quarter by approximately 11% to $113 million or $1.56 per share. Investment income in the 2007 third quarter was adversely impacted by $4.6 million or $0.06 per share from the marked-to-market effects of several fixed income investments that are accounted for in this manner because of the legal form of the investments. There were no such investments in the comparable 2006 period, while, as we noted on our last conference call, marked-to-market adjustments had a positive impact on our 2007 second-quarter earnings of $3.4 million or $0.04 per share.

The growth in investment income on a quarter-over-quarter basis was primarily due to a higher level of average investable assets. This was primarily generated by cash flow from operations, which amounted to $426 million for the 2007 quarter and approximately $1.5 billion on a trailing 12-months basis.

After reflecting share repurchases, which I will comment on in detail in a moment, investable assets rose to approximately $9.9 billion at quarter end.

The average pretax yield of the fixed-income portfolio, before the effects of the marked-to-market adjustments I just mentioned, in the 2007 quarter was 4.92%. This represented an increase of 14 basis points over the comparable 2006 yields while sequentially on an adjusted basis, the yield increased by 12 basis points.

The increase in investment yield on a quarter-over-quarter basis primarily resulted from higher interest rate levels as the portfolio's average credit quality remained near the AAA level. During the quarter, the duration of the portfolio remained basically unchanged at about 3.3 years. The unrealized gain in the portfolio for the 2007 quarter was primarily attributable to interest rate and currency movements and increased book value per share by $1.53 for the quarter. The portfolio continues to be comprised primarily of high-quality, fixed-income securities with essentially no investments in hedge or private equity funds.

The income tax provision applicable to operating income was roughly 1% for the 2007 quarter. As you may know, the Company's quarterly tax provision is adjusted to reflect changes in the estimated annual effective tax rate. A significant part of the Company's catastrophe exposed business is written by a Bermuda subsidiary. As a result, the Company's effective tax rate is favorably affected in periods with a low level of catastrophe losses and adversely impacted in periods with significant catastrophic claims activity. As a result of a below-average level of cat losses this year, the third quarter of 2007 includes a benefit of approximately $4.5 million or $0.06 per share to reflect the downward adjustment to the annual effective rate. We currently expect that our annual effective tax rate on operating income for the balance of 2007 will fall within a range of 2 to 3%.

Our balance sheet and financial flexibility remain strong with total capital amounting to approximately $4.2 billion and debt and hybrids representing 15% of the total. With respect to capital management during the third quarter, we repurchased approximately 2.2 million common shares at an average price of $67.04. That brought the total number of shares repurchased through the end of the third quarter to approximately 5.8 million shares at an average price of $68.94 or 1.34 times our September 30 book value per share.

At quarter end, we had approximately $600 million remaining available under our current repurchase authorization. On a weighted average basis the share repurchases contributed $0.13 to EPS for the 2007 third quarter. The effect would have been $0.16 per share if full weight were given in the quarter. The repurchases aided ROE for the quarter by approximately 170 basis points. Share repurchases through September 30 reduced book value per share by approximately $1.37 on a net basis.

In summary, our 2007 operating results and capital management initiatives produced an ROE of 24% and our book value per share, despite the short-term effects of the share repurchases noted above, rose 8% during the quarter to $51.34 per share. With that, we will take your questions.

(OPERATOR INSTRUCTIONS). Josh Shanker, Citigroup.

Commenting on the -- you ran through the reinsurance premiums ceded issue with Flatiron pretty quickly. I just wanted to go through it a little bit slower. I noticed that the percent of business ceded declined pretty significantly during the quarter, and we've started to see Flatiron commissions come down. Can we conclude that the peak for commissions from Flatiron has passed?

This is John. No, the answer is, the way the commissions are recognized relate to earned premium. And as you noticed in my remarks, earned premium is continuing to grow. So I wouldn't anticipate that.

The reason for the drop in volume, as I mentioned, was in the third quarter of 2006, there was a brief period of time where we increased the ceding percentage to Flatiron from 45% to 70%. That resulted in an increase only in the third quarter of '06 of the business ceded, but obviously that affects the comparability of the numbers from '06 to '07.

Okay, very good. And the other question relates -- and I missed the first five minutes of the call, but I don't think you mentioned it. During the -- ultimate cat losses for the Tri-County area in California, I assume that's quake. Is there any comments said that are important to make about the California wildfires?

Well, on California, based on what's happening today, very, very minimal exposure. We might get maybe a risk or two out of our insurance group in case some commercial property goes down like a shopping center or a warehouse of some sort that we might be on risk. Nothing reported to us as of yet.

On the reinsurance side, it has to be significantly greater damage. We run numbers to a $4 billion event and our exposure is negligible, so it's got to get to the $10 billion or more for us to be a significant event, and we don't see that based on the information we have today.

And Josh, the peak exposure was Tri-County Florida, not --

Oh, Florida, Florida. Okay.

But I don't want to mislead you. We do have significant exposure in California, primarily to quake.

To quake.

Absolutely. Okay, well, thank you very much. Congratulations on another quarter.

Kevin O'Donoghue, Banc of America Securities.

I was wondering first of all if you could give us a little bit more details on your -- on the write-down you took in the bond portfolio. Was that just related to -- or what was that related to?

It wasn't a write-down. Let me go through this because I think it's worthwhile. Certain investments -- we allocate, usually give mandates to managers. But we've begun, in the beginning of this year for instance, to allocate some of the portfolio to European credit. And we did that by going into a fund. Now, the legal form of the fund, and this is one of several investments, it's not one. The legal form of the fund is a limited partnership. Under the current accounting rules, when you have an investment in a limited partnership, even if all of the underlying securities are fixed income securities just as they are in our regular portfolio, you have to mark everything to market. So in those investments each quarter, the yield plus the change in market value, including the unrealized, gets marked-to-market and runs through investment income.

And I noted in the first quarter that it had a positive effect, was slight. Last quarter I pointed out that it had a positive effect, so those numbers can move around quarter to quarter. It's not a huge part of the portfolio, but on a quarter-to-quarter sequential basis especially, it can have an impact on the comparability of the numbers.

Okay, thanks. So first of all, this was -- just to be clear, this was related to European credit and then --?

And other securities. It could be U.S. fixed income. The key point is these are fixed income securities. If we were holding them -- we had a manager with a mandate, what you would have seen is the unrealized gain or loss on those specific accounts would have run through with the rest of unrealized directly through the balance sheet. But because they are -- the form of the legal investment is a limited partnership, the unrealized gain or loss in a quarter is going to run through the income statement in the investment income line.

Okay. And what would you attribute the change in the value of the securities to?

Change in credit spreads because they are (multiple speakers) they are taking on certain level of credit risk.

Okay, thanks. And then just one other brief question. On the tax rate, you mentioned 2 to 3% for the duration of the year. Given similar levels of cat losses in 2008 as you have this year, I realize this was a good year, but if we were to assume that, would the 2 to 3% be a good number to use for next year as well?

Well there's a lot of moving parts and typically we'd prefer to get to the end of the year before we get a better handle on what we would expect next year. But all things being equal, the answer to that would be yes, assuming all things are equal; they rarely are. But if it happened, the answer to that, that wouldn't be a bad guess, assuming a low level of cat activity.

But if you go back historically, you will see we had years or quarters that our tax rate was as high as 8 or 10% because, or higher, because of cat activity in the U.S. and losses in the U.S. So we can't predict what's going to happen with cat, so very hard to predict what's going to happen with the tax rate.

Okay, appreciate it. Thank you very much.

Matt Heimermann, J.P. Morgan.

Couple of questions. First, Dinos, could you just elaborate maybe a little bit on how you see renewal pricing versus new business pricing at this point and maybe just give us a flavor of how it varies by market?

Well, renewal pricing is -- well, first, you know, we try to keep most of our renewals, as most of our competitors do, you have more knowledge on those accounts and our preference is to maintain as many accounts that we understand the best. And also we can measure the relevant changes year over year. So renewal pricing has been stronger for us than new business pricing. New business pricing is more competitive. And as you can see from our volume, we have more of difficulty in attracting new business by maintaining our underwriting standards. You know, at the end of the day, you can always attract a lot of new business if you don't measure it. But to me if you don't measure what you are writing, you know, you're not managing it. And that's why we put a tremendous amount of effort in trying to create this benchmark system that is outside our underwriting staff. This is run by our pricing actuaries and make sure that we are comparing the new business pricing to our renewal business and see that we are getting equivalent kind of risks for the same kind of exposure as we get the same kind of price.

It is not as trivial as it sounds. It's a complicated process, but we like to do it electronically for as many lines of business as possible because that gives management a better understanding as to the fundamentals of the books of business that we underwrite. And that's where the effort -- but in general, as a comment, the new business is probably -- doesn't have as good margins as renewal business and we see that across all of our business.

Can you quantify what that is or maybe even if it's -- given an example like contractors or [Med Mal], I think you've been a little bit more cautious today relative to the past?

Well, it's not a question of cautious as the more information we give, we find more (technical difficulty)competition on certainly lines. So I -- but in essence, I -- we believe that new business will probably behave a few loss ratio points worse than the renewal business and it varies by line. And what we are trying to do is to make sure that our underwriting departments price the business with the same margin expectation, but I can tell you it's difficult. That's why we're not finding a lot of new business that we like and they have the return characteristics that we have on our renewal book. And I think that's true for most of our competitors too independent of if they talk about it or not, because it's tough to move accounts, even in a market that prices are coming down. Everybody is trying to maintain their renewals.

Okay. And then could you just maybe discuss a little bit, it looked like growth in your European business slowed down or actually declined. And that had been an area where you had been seeing growth in previous quarters. Anything dramatically change there or is this just kind of a maturation of the business?

No, as a matter of fact, actually, our European business had a slight growth for the quarter. There is a way in the way we account for it, and I'll let John get into the details. It's only one deal that we had that we had a participation in a pool that we cease to participate. John, do you want to elaborate on that?

Yes, Matt, I know you got that number essentially by looking at the business written by client location. And the assumption you made, which would not be an unreasonable one, is that all of the business that we said was in Europe was written by our Europe operations. But we did have a participation in the U.S. insurance group in an aviation pool, in France, basically, that we're on from 2002, and we withdrew from that this year. So some of what you are extrapolating as a decrease wasn't in fact a decrease from our European insurance business--

From the London office. It was more what we're doing out of New York.

And what was the underlying growth in the London office then?

Our gross written premium in the London office grew by -- I have it here -- it's $8.5 million on 92, so about 9% or so. And on a net basis, it only grew about 1% because we had changes in the reinsurance structure. So we went from only about $1 million growth on a net basis.

Okay, very helpful. Thank you.

Vinay Misquith, Credit Suisse.

What is the impact of the facultative reinsurance case with Gen Re that recently we heard that Gen Re won? What would be the impact on Arch Capital?

We don't see any impact other than we continue to build that business. And as we said in prior calls, this is a long-term play. It will take time for them to build their business. The order that we have is basically the same as the pre-existing agreement we have with Gen Re that in essence we can't -- we have to abide by existing law and not use any Gen Re trade secrets in the way we build our business.

We never had any of their trade secrets. We never had any intention of utilizing any of that. So basically we continue to be in the marketplace and compete as another competitor in the fac arena. And we are building the business in the same fashion that we were doing a month ago, two months ago or three months ago. And we're pretty happy with what we have accomplished so far.

Okay. And in terms of your Flatiron Re, it's up for renewal soon. What are your thoughts with respect to the renewal? Do you want to renew? Do you have a sense for whether the investors on the other side want to renew it and at what level?

Well, it's premature to get into those discussions and we won't do them on the call. I will tell you what we're doing right now is our cat teams in Bermuda, they are assessing the market. We got to make an assessment. Now the season is over and we're going to see what our assessment is for the upcoming year. That study I haven't seen yet. They working on it; Nicolas Papadopoulo and Marc Grandisson, are responsible for that section of our business. They are in the process of doing that. We're going to meet first internally to review what the prospects are for next year in the next couple of weeks. And after that, I think we will make some determinations about what our risk appetite is, how much additional capability do you have to maintain more of that business ourselves or not. Then at that point in time, we can talk to our partners and see what their appetite is and then we will make determinations. But we are probably two, three weeks ahead. If this call was three weeks from today, I would probably give you more information. But right now it's still up in the air.

Sure. That's great. One last question. Considering the noise about some form of Bermuda taxation, I'm just curious as to what is the dollar value of premium that you cede from your U.S. subsidiaries to Bermuda? Would that be about 1.6 billion roughly? And what is the combined ratio right now on the business that is ceded to Bermuda?

Well, there is -- we have two sections of business in the U.S. One is in the reinsurance section and one is our insurance. On the reinsurance, we choose to write a lot of business out of our U.S. subsidiary and in essence, cede 80% of it to Bermuda, but we don't have to do that. As a matter of fact, every client in our reinsurance group in the U.S. has is a third-party client. So in essence, we have the ability, if there is a change in the law, to move the business and underwrite it directly out of Bermuda.

So you might draw the wrong conclusions if you try to see what we cede today. We do that because we have an interest in maintaining jobs in the U.S. and that allows us to keep those jobs in the U.S. But very clearly, if there is a change we need to adjust to it.

On the insurance group, we see 80% of the net net after inuring reinsurance. Our insurance group probably cedes a little -- a little over $1 billion to Bermuda. And basically, that's the business that probably any change in legislation will affect.

I would like to remind you though that we do pay excise tax and also we do pay a risk pricing transfer tax. Not all of which goes through the tax line; all the excise tax comes as acquisition cost up on the top line. So not knowing what changes might or might not go through Congress, you can't really predict any effect on our tax rate.

Sure. That's fair enough. Since most of the cat business is written abroad in Bermuda, what would you roughly estimate the combined ratio to be on the business you cede? Would it be high 90s, mid '90s do you think?

Remember this -- we don't actually -- you mean the cat business, you're talking about the U.S. business or --?

Yes, U.S. business that you cede, would it be --?

Well, the U.S. business has the same combined ratio as the insurance group. You see that. It's a quotashare across the board. And the reinsurance group will probably have a bit higher combined ratio because the business we write through the reinsurance group that today we cede to Bermuda, their regional property business, which has pretty much similar characteristics and casualty business. And our casualty business has a higher combined ratio.

When you look at the insurance group, you have to remember we write business out of London and a fair amount of the business out of London is property business also. So if you looked at the group overall, the U.S. rate, because it's more casualty oriented, the combined ratio is probably higher than the (multiple speakers) combined ratio as a whole.

Sure. So mid to high 90s would seem about right. Okay. Thank you very much.

(OPERATOR INSTRUCTIONS). Jay Gelb, Lehman Brothers.

Good morning. I was hoping you could talk a little bit about two financial components, the pace of reserve releases, how we should think about that and also the pace of share buybacks.

Let me start with the buybacks. We have an authorization for $1 billion. We utilized $400 million. The cat season is pretty much over. We will continue based on market -- where our share price is. We will continue on a steady state on executing that authorization. So we don't see any change in that. On -- your second question was on --?

Reserve releases.

On reserve releases. It's always driven by the data. No difference in our methodologies. As a matter of fact, in the insurance group, periodically, we will review different lines of business, and based on what the new indications are, we might make adjustments to prior years based on what the new information supports. And we will continue with that process.

This quarter, as you saw, maybe we had a significant release, but most of it came from the reinsurance group and most of it came from our short-tail lines. And basically you had certainty on those, so we had to release them.

And, John, I would like you to comment a little bit on that and also how did it affect as part of it on the expense side because some people might not get that distinction.

That's a great point, Dinos. But as Dinos said, a lot of the releases came on the property side and it was driven mostly -- it was property and it was out of reinsurance. And on the reinsurance side in particular, sometimes the contracts have profit commissions embedded in them. So when you look at the amount of reserve releases -- and we give this information on page 8 of the release; it's right there for you to look at. But essentially you should look at the effect on acquisition expenses because we might reduce the reserves on a treaty from 2002 or 2003 or 2004, and when that happens, that triggers a related acquisition cost. So the net impact on earnings is less. So when you look at that, for instance, in the current quarter, it added about 1.8 percentage points to the consolidated acquisition expense ratio and total expense ratio, and a higher number than that to the reinsurance ratio. Over 4 points of the ceded commission increase in reinsurance was related to the profit commission. So you should -- one thing to keep in mind is always look at the two net numbers. We are going to continue to provide people with that; we recommend that.

And the second thing again, it is data driven. Most of it continues to come out of the early years of the hard market, '02, '03, '04. There is a little bit on the reinsurance side into '05 a little bit, where we have data on contracts for a long time. But as we've said on prior calls, this is a process that we'll continue to evolve as we introduce more of our own experience into the equation. And other than that, that's about all we can really say about that.

Some other Bermuda insurers and reinsurers have started releasing more long-tail lines of liability. Can you talk about where you are in (multiple speakers)?

Yes, our approach there continues to be relatively cautious. We are obviously now -- we're coming up -- this is the end of our sixth year. And we're coming to a point where at least '02 and '03 are beginning to gain more credibility. But again, you can get one or two years worth of experience; in our view, in any event, it's premature to be reacting strongly to an indication because that indication can turn around on you on a future period. So we continue to be cautious about the more recent years.

Of course, as you go through the process, we are looking at loss picks over a continuum when we have a base year, we're continually monitoring the base year. So as prior years get better, we will start to look at the current years once we feel there's a little more credibility, and we roll things forward, we take what we think the base year loss ratio was, add what we think the trend or claims inflation was, take pricing changes into account and then get a new pick. So that, as we go forward, that can start to have an impact. But we're still rather cautious on the current years.

And then finally, can you comment on loss frequency trends in the primary insurance book? We're starting to hear a little bit that those favorable loss frequency trends may be bottoming out industry-wide.

Yes, no disagreement with that. I think early on, in the early years, we had better frequency trends, and those are having --starting to level off. But our book of business is so specialized that it's hard to draw general conclusions by what we do. And also our book of business is starting to shift a bit. As I said in my prepared remarks, I think we are intentionally moving to smaller accounts because that's where we live. Even though it's harder to find them, underwrite them and retain them, they have better margins than the larger accounts.

When you say smaller, can you quantify that a bit premium-wise?

Yes, under $50,000 in premium.

Jay, in a hard market, for instance, when you are using all your resources and you want to make sure you focus on the most attractive opportunities, we tend to have artificial minimum premiums established. So there's nothing magic about whether an account is 45,000 or $52,000. It's just a question of where you want to focus your energies, in the current market where everybody has to filter through a lot more submissions in order to determine what they're going to get. And we've got a wide distribution basis. Dinos mentioned in his remarks, we can access it. We're starting to look at a broader spectrum of risks. that's -- so we've dropped sort of what might be considered sort of artificial definitions of what size premium account we look at.

Yes, but when we say broader spectrum, it's on size. We're still in all the specialty lines, we still are doing what we know and understand and we're sticking to it. But if we had in some divisions a requirement that we told a distribution channel we want accounts that they generate a minimum of $50,000 or more, now we might have dropped that requirement to 25 or even below that. So we are broadening the funnel, because we've got to process a lot more ore to find the golden nuggets, and we are willing to do it. Our people are working very, very hard throughout the country even to find these opportunities.

And the fact that we have built infrastructure that allows us to be close to the distribution channel and close to our insureds, I think it gives us an advantage against those who don't have it. Of course, there's many companies who have it. A lot of the established players, they have robust distribution and branch office structures. And that's basically one of the strengths of ours, as you compare it with some of the newer companies.

Right. Thanks very much for the answers.

(OPERATOR INSTRUCTIONS). Brian Meredith, UBS.

A couple of questions here. First one, I want to focus a little more on the reinsurance business and the drop in volume there. You mentioned on competitive pricing, is it that you are uncomfortable with what's going on with terms and conditions in the reinsurance market? Or is it more the subject premium base that you are having some issues with? Or is it just a combination of them both? Because if I look at some of your reinsurance competitors, they actually had growth in the quarter, particularly some of the larger, more established reinsurers.

Well, it's a good question. First let me remind you that we are predominately a quota share reinsurer. We like that form. We like to be core partners with our cedants. And in essence when you look at our volume split, it's probably close to three-quarters quota share, one-quarter excess of loss. And most of the excess of loss comes when we write the property, property cat business in general.

With that as a background, of course, we do spend quite a bit of time understanding what's happening on the underlying book of business that we're going to quota share with a cedant. And in some cases, we got off treaties because we didn't like the underlying environment. We felt that basically, the rate reductions that we were experiencing based on our audits was beyond what we had a comfort level with. And in some cases, it was the willingness of our cedants to retain more net themselves and, in essence, in order for us to continue, we had to give up ceding commission points. And once that happened, the partnership gets skewed, so you'll be in arbitrage a bit and you have a negative arbitrage because the economics for the cedant will be much better than the economics of the reinsurer. And we had situations that either one or the other caused us to cease participating on certain treaties.

And we said from the beginning in the reinsurance business, where you make forward bets, especially when you write quota share, that any risk that they are going to bind over the treaty term, which is between now and also 12 months from today, we're going to be side by side with them. We're a bit more cautious than most, and we are willing to give up the volume for making the right underwriting decisions.

And a lot of what we have given up is based on those two parameters. Either we felt that the ceding commissions, they were too high for our liking and we were not an equal partnership or we didn't like the underlying pricing of that business, the book of business that we're underwriting.

Great. And then my next question, you mentioned that your net PML is 18% I guess of equity right now. Remind us again where you are comfortable taking that and what is your gross PMLs right now? And how is that going to play into your decisions with respect to Flatiron and renewing the fleet? I mean at what level or even if you do renew the Flatiron, sidecar?

It's the -- there is two components of the net PML. It's what comes from the insurance group and what comes from our reinsurance group. And our reinsurance group has two components. It has the U.S.-generated facility, which is our Morristown facility versus the -- what we do in Bermuda. Our Flatiron agreement is for what we write out of our reinsurance company in Bermuda.

Without getting into -- just to give you a flavor what it is. The way we allocate a PML aggregate dollars is basically -- 20% goes to the insurance group, one in five; and approximately another 20% goes to the U.S. reinsurance group, give and take a few points. So then, you're a good mathematician. You can do the math. 60% of the net PML is Arch Re Ltd. and the relationship between us and Flatiron is 55/45. So take a pencil and paper and just do the math and you get the answer about the gross PML.

Okay. Great, thank you.

At this time there are no further questions. I'll turn it back to management for closing remarks.

Well, thanks, everybody. Another great quarter. We'd like to congratulate our staff for their discipline and we're looking forward to sharing our comments with you in the next quarter. Have a good day.

Ladies and gentlemen, this does conclude today's presentation. You may now disconnect. Thank you very much and have a great afternoon.