CNA Financial Corp (CNA) 2007 Q3 法說會逐字稿

Good day and welcome to the CNA Financial Corporation third-quarter 2007 earnings conference call. Today's call is being recorded.

At this time I would like to turn the conference over Ms. Nancy Bufalino. Please go ahead.

Thank you, Connie, and good morning. Welcome to CNA's third-quarter 2007 financial results conference call. Hopefully everyone has had an opportunity to review the press release and financial supplement, which were released earlier this morning and can be found on CNA's website. With us this morning to discuss our financial results are Steve Lilienthal, Chairman and CEO; Craig Mense, CFO; and Jim Lewis, President and CEO of PNC Operations. Before we get started I'd like to advise everyone that during this call there may be forward-looking statements made and references to non-GAAP financial measures. Please see the section of the earnings release available on CNA's website headed "Forward-Looking Statements" with regard to both. In addition, the forward-looking statements speak only as of today, October 29, 2007. CNA expressly disclaims any obligation to update or revise any forward-looking statements made during this call. This call is being recorded and webcast. During the next week the call may be accessed again on CNA's website at www.CNA.com. Following the conclusion of today's prepared remarks by CNA's senior management, we will be happy to take questions from the investment community.

With that I'll turn the call over to CNA's Chairman and CEO, Steve Lilienthal.

Thank you, Nancy. Good morning, everybody. Thank you for joining us. Welcome to our third quarter earnings call and we're happy to have you with us today. CNA had a solid third quarter. Catastrophe activity for the quarter and year to date has been favorable and below expected levels. But most importantly we continue to operate in a very disciplined fashion, consistent with previous quarters, consistent with the recent upgrade of our Fitch rating to A, consistent with our decision to increase our quarterly common dividend to $0.15.

Some highlights from the quarter. Third quarter net operating income was $212 million versus 283 in the prior-year period. Year-to-date, net operating income was $837 million versus the prior of 822. This reflects continued solid results and a disciplined focus on the fundamentals in our Property & Casualty Operations. Period-over-period comparisons also include the $108 million impact of the previously-disclosed Hancock settlement. Absent that settlement, the third quarter net operating income improved 13%, to $320 million versus $283 million. Year-to-date net operating income improved 15%, $945 million versus $822 million.

Third-quarter net income was $174 million versus $311 million in the prior-year period, which included the impact of $38 million in net realized investment losses in the quarter versus $22 million net realized investment gains in the third quarter 2006. Year-to-date net income was $687 million versus prior-year period, the date of 779, which includes $142 million of net realized losses in 2007 versus $41 million in net realized investment losses for the same period in 2006. Book value per share of $37.23 was up modestly from June 30, '07, and up $2.80 or 8.1% over the same period in 2006. Net investment income of $580 million was down modestly from the third quarter of '06, $580 million versus $600 million. Craig will provide more detail on this shortly and also give you an update on our run-off operations.

As I mentioned our core Property & Casualty Operations continue to demonstrate solid fundamentals and disciplined execution across the board. The combined ratio, with some help from favorable development, improved to 91.6 for the quarter and 93.8 year to date. Excluding the impact of development, the combined ratios were consistent with the past several quarters. Net written premiums were down 9%, in-line with selective underwriting and a softening rate environment and our average rates were down 4% versus the more significant decreases being reported in the external industry rate surveys. Retention of renewal business was down slightly due to market pressures and our new business volume continues to run at a reduced level, which is appropriate for the market and exactly what we told you we would do several years ago.

The Property & Casualty Operations expense ratios continue in the sub-30 category at 29.4% for the quarter, 29.2% for the year, which is consistent with recent quarters. With the topline under pressure, we continue to take a consistent, persistent and a very granular approach to management, just as we have done in the past and we'll have an even more urgent need to do so going forward. Jim Lewis will touch on this in a moment in his comments. Overall I am quite pleased with our third-quarter performance. Our core business delivered another quarter of solid business and we are well positioned to continue on this track in spite of market pressures. Going forward, as it has been in the past our focus will be intelligent, disciplined participation in the market, optimization of our highly-diversified portfolio, specialized segment and claims operations, expense management, balance sheet strength, talent development, and an absolutely relentless and singular focus on bottom-line results.

With that let me turn it over to Craig.

Thanks, Steve. Good morning, everyone. The third-quarter results are reflective of our ongoing efforts to continuously improve the operating and financial fundamentals of CNA. As importantly, our behaviors are consistent with what we said you could expect from us. The core Property & Casualty Operations combined ratio was in the low 90s. Rate retention and new business reflect disciplined participation in the market. Our sustained sub-30 expense ratio remains competitive with our peers. In the run-up operation, while the Hancock settlement did add an asterisk to our results, it removed a long-standing legal contingency and its associated uncertainty. While our earnings and ROE were reduced by the settlement, our third-quarter and year-to-date financials continue to demonstrate the solid earnings power of CNA. Our investment portfolio continued to stand up very well in the current financial market and creates for us an environment of opportunity, not stress.

Now I'd like to give you bit more detail on the financials before turning it over to Jim Lewis. Today we reported third quarter net operating income from continuing operations of $212 million or $0.78 per share compared to $283 million or $1.05 per share in the prior-year period. You will recall that the per-share amounts for '06 include the impact of undeclared preferred stock dividends. These dividends were eliminated with the repurchase of the stock from Loews last August. Net income for the quarter, which includes the impact of discontinued operations as well as realized investment gains and losses, was $174 million or $0.64 per share, compared to $311 million or $1.15 per share in the prior period.

Year-to-date, net operating income from continuing operations was $837 million or $3.08 per share, which compares to net operating income of $822 million or $3 per share in the same period of 2006. Year-to-date net income was $687 million or $2.53 per share versus $779 million or $2.83 per share for the prior-year period. Third quarter operating ROE was 8.6% and year-to-date operating ROE, 11.7%. With respect to book value per share, we ended the third quarter at $37.23, up slightly from the second quarter and as Steve said, up 8.1% year over year. The increase was driven by solid net operating income, which more than offset net realized losses and changes in our unrealized gain loss position.

Property and Casualty Operations produced net operating income of $331 million in the third quarter, which compares to $282 million in the prior-year period, a 17% increase. Our core business benefited from favorable net development and increased investment income, and our lines favorable development was primarily in general liability and property, and was offset in part by unfavorable development in specialty lines related to our large law firm book. Life and group non-core produced a third quarter net operating loss of $131 million. In addition to the $108 million impact of the Hancock settlement, our life group results were affected by a lower net investment income related to the pension deposit business. With respect to the corporate segment, the third quarter net operating income was $12 million, which compares with $16 million in the prior-year period. The decrease was primarily driven by lower net investment income. Pretax net investment income in the third quarter was $580 million, which compares to $600 million in the prior-year period.

We benefited from higher income from our fixed income portfolio and lower interest expense on funds withheld. These positives were offset by lower income from both limited partnership investments and the trading portfolio. Invested asset growth was dampened by the Hancock settlement, but still increased $160 million during the quarter and is up $1.2 billion year over year. Net realized investment losses were $38 million after-tax in the quarter as compared to gains of $22 million in the prior-year period. The decrease was driven by higher after-tax impairment losses, which totaled $122 million, comparing to $30 million in the prior period. Approximately 25% of the impairment write-downs were related to subprime exposures. However, those mark-to-market losses were largely offset by economic hedges held against that segment of our portfolio. We've added an exhibit on page 6 of the financial supplement to provide a profile of our subprime exposures inside our mortgage and asset-backed holdings.

You will recall that last quarter we expanded our disclosures to explain that our interest-sensitive securities are managed to two distinct duration objectives. Those securities that support the Property and Casualty segment, which comprise approximately 77% of all fixed income had an effective duration of 3.5 years at the end of the third quarter. The second group of assets that are segmented and aligned with certain long-term liabilities, principally in the life and group segment, had a duration of 10.5 years. Combined, our fixed income portfolio had a duration of -- an effective duration of 5.1 years. Turning to expenses, the PC Ops third-quarter expense ration was 29.4%. On a par with peers, continues to reflect our sustained focus on CNA's expense structure. Expense management remains a critical component in our strategy for managing through the cycle. In other items of note, I'm sure you saw that Fitch has upgraded our rating. We are pleased, of course, and we view it as validation of the improved operating and financial fundamentals of CNA. Finally, with respect to our dividend, as announced last week, we increased it to $0.15 per share. This represents an increase of $0.05 per share over the $0.10 per share dividend paid last quarter.

With that I'll turn it over to Jim.

Thanks, Craig. Good morning, everyone. Property & Casualty Operations performed well in the third quarter. In a market that keeps getting tougher, we continue to focus on the bottom line. We leveraged our strengths, underwriting discipline, our diversified portfolio and our strong market franchise. We continue to move forward on growth initiatives in the right way, where we can make a profit and deliver a value-driven sale. I have a very healthy respect for the intensifying pressure of today's market, but we handled them well in the third quarter and we're well positioned to continue managing through the market cycle.

Now let's turn to a few of our key operating metrics, starting with premium. Property & Casualty Operations gross written premiums were down approximately 8% to $2.2 billion in the third quarter, and down 2% to $6.9 billion year to date. The underlying gross written premiums were down 8% for the quarter and flat year to date, while specialty lines decreased 7% for the quarter and 5% year to date. On a net written basis, PNC Ops premiums were down 9% for the quarter and 4% year to date. Standard lines in net written premiums decreased 15% for the quarter and 7% year to date, while specialty lines was essentially flat for the quarter and year to date. The differences between our net and gross premium numbers relate primarily to change in volumes in our fully ceded captive program, as well as the reduced use of reinsurance and specialty lines that I mentioned last quarter.

The pressure of the market is evident across our portfolio. As we have said before, we are not going to compromise our discipline in pricing and selecting risk. Our premium volume reflects disciplined participation in the market combined with ongoing portfolio management to improve overall profitability. Profitable top-line growth is more and more challenging. Over the past few years, however, we have become better and better at identifying and pursuing attractive market niches. That capability serves us very well in a soft end market. In small business, selected middle-market segments, technology, wholesale property, and specialty niches, such as healthcare and professional liability, we are building momentum nicely.

Now let's look at rate and retention. For the third quarter, average rates decreased by approximately 4%, a slight deterioration, less than 1%, from our rate experience in the first and second quarter. Standard lines rates were down a little more than 2% while specialty lines decreased a bit less than 5%. Industry-wide pricing surveys are showing average price decreases in the mid teens. We're not going to go there. Besides not wanting to undo five years of hard work, we have built in some resistance to market forces. Our focus on small to mid-sized accounts makes our book somewhat less price sensitive. In addition, our strong franchise gives us a competitive advantage in construction, healthcare professional liability, and other key markets.

As for retention, we are running at approximately 79%, slightly down from the past few quarters. Standard lines retention was 76% during the third quarter with specialty running at 83%. We track the quality of our renewal business using a measure we call renewal efficiency. It's just the spread between the incurred loss ratio of renewed business versus nonrenewed business. Renewal efficiency and other measures tell us that our underwriters continue to retain the right business. Overall, our rate and retention numbers reflect consistent underwriting discipline. It's tough out there. Market pressure goes with the territory. Still, we know what it takes to maintain the quality of our book of business, also what it takes to be smart about growth. More on that in a moment.

Now let's turn to new business. We wrote approximately $289 million of new business in the third quarter. This represents approximately 18% of production, very close to where we have been all year. By comparison, new business in third-quarter '06 was approximately $374 million or 20% of production. Overall, our new business continues to be right where it should be and where we said it would be under current market conditions. Our new business used to be someone else's old business, and with every company working very hard to hold on to their quality renewals, our disciplined, data-driven new business strategy becomes even more important. I already mentioned our renewal efficiency measures to monitor the quality of our renewal book.

We have similar measures for new business quality, such as variance to manual for new and renewal pricing. Quantitative measures like this tell us that our underwriters are writing good new business without giving it away. In addition, cross-sell continues to benefit new business quality and volume. We work very hard at selling additional CNA products to every CNA customer. Not only does cross-sell deepen our client relationships, which helps drive retention, it gives us more data about their risk profile, which helps with pricing and selection. In the third quarter cross-sell generated approximately $118 million of new business. This is good business that might otherwise have been left on the table.

Turning to loss ratio, the third quarter Property & Casualty net [common] year loss ratio was approximately 62%, a four-point improvement from the prior year. Standard and specialty lines both came in at approximately 62%. Our third quarter loss ratio is benefited from approximately four points of favorable net development. Standard lines was seven points favorable, while specialty was unfavorable by two points. The impact of development is fully broken out on pages 7 and 9 of our financial supplement. On a net accident year basis, the 2007 loss ratios for Property & Casualty Operation is approximately 66%, standard lines is 69% while specialty lines is 61%. Each ratio is up approximately two points from 2006, in part reflecting the impact of a declining rate environment.

We view loss ratio as our number-one operational metrics. In the past few quarterly calls, I've mentioned the various measures that protect our loss ratio. First is a very disciplined process of underwriting referrals and audit. Second, we continue to refine our underwriting with predictive modeling tools. Third, our management information system provides underwriting data on a very granular level. Fourth, our segmented specialized claims operation directs the most complex claims to our most experienced adjusters. And finally, we continually adjust our business mix to optimize our portfolio. With these loss ratio protections in place, I feel good about our ability to manage this key performance indicator.

Turning to combined ratios, Property & Casualty Operations came in at slightly less than 92%, three points better than the prior-year period. Standard lines came in at approximately 93%, while specialty lines continues to perform very well with a third quarter combined of just under 90%. Before the impact of development, the third quarter combines are consistent with the solid performance of the past few quarters. In addition, as Craig mentioned, we continue to manage expenses very aggressively. During the quarter we took action to lower our cost structure and improve efficiency. We also expect these changes to sharpen our focus on enterprise-wide production.

Before wrapping up, I'd like to comment briefly on your outlook. For a lot of reasons I am cautious about the market, but confident in our direction. CNA has a very diversified portfolio. Our distribution relationships are very strong and we have a track record of building strong, sustainable, market positions. I feel very good about the momentum of our small business operations. I expect that momentum here will carry over into our custom express strategy, which targets the lower end of middle market. Accounts of this size are now dominated by regionals, but with our local branch structure and our small business products and platform, we are well-positioned for profitable growth. In middle market, we are leaders in many of our construction industry segments and are bringing that same focus to nonconstruction industry programs. In specialty, we're leveraging our strong position in healthcare to related market niches. In short, we know what it takes to grow in the right way and we have the people, tools, and determination to get it done.

In summary, Property & Casualty Operation had a good quarter and is on track to another good year. We continue to focus on underwriting discipline, claims excellence, and aggressive expense management. We continue to leverage a strong market franchise, long-standing distribution partnerships, and a very diversified portfolio. With that I'll turn it back to the operator for the Q&A.

Thank you. (OPERATOR INSTRUCTIONS) We'll take our first question from Jay Cohen from Merrill Lynch.

Thank you. I guess I'd like to get more detail of what happened in the subprime portfolio during the quarter. I wonder if you could break out -- you mentioned there was the realized losses. Obviously, there was some unrealized losses as well. I'm wondering if you could just total up what the impact was to the subprime portfolio, both realized and unrealized, and maybe also go into a bit more information on the economic hedges, which offset some of that?

Jay, one of the things you should also look forward to, we'll be filing the Q early this week -- in the next couple of days, we would anticipate -- and there's some additional detail about all the subprime positions as well as where we are with -- and unrealized loss and other things going forward. If I understood your question correct, you want to know a little bit about the credit -- we hold some credit default swaps, about $60 million of notional value at the end of the quarter, which largely offset that decline and impairment related to subprime. Related to the subprime percentage, as I said, about 25% of the impairments which is $122 million after-tax, is subprime, and the $60 million hedge offset a little bit more than two-thirds of that. And then relative to changes in unrealized, it's a little bit -- really kind of up and down. Not a whole lot of change in unrealized over the -- just the quarter in subprime. Most of that has been reflected in the impairment write-downs.

That's actually helpful. So the 25% on the impairment is basically the hit that you guys took, essentially, on a gross basis, forgetting the hedge, on the subprime portfolio?

That's right. And I think so -- just please don't interrupt as that's the entire write-down, because we did hold a hedge against it and I think deserves some credit for some risk management processes.

Most definitely.

-- having that in place.

Next question, The limited partnership portfolio, do you report that on a one-quarter lag? I can't remember if you do or not?

Yes.

So a one-quarter lag. Okay. And then the premium decline, was there anything --?

Actually, a one-month lag, Jay, not a one-quarter lag.

Oh, okay, that's helpful.

That's in terms of how pricing comes through. And then keep in mind that that portfolio is mark-to-market every month, so that's coming through then -- any changes in market value are coming through net income.

Got it. Operating income as well?

Yes.

Okay. And then the premium decline, was there any comparison issues? Was last year partic -- essentially a strong quarter and the comparison was more difficult, or was last year just a normal quarter? It looked like last year's third quarter had the biggest growth of any of the quarters in 2006. I just can't recall if you guys talked at all that quarter about if this was an especially strong quarter a year ago?

Last quarter -- third quarter of '06 was a very strong quarter. Within third quarter of '06, we had a very large program to the tune of around $50 million and when that came up for renewal, it was moved from a 12-month policy to a 17-month policy. So it actually renews in March of '08. I think the other thing that you're really looking at there, Jay, is just the overall market pressures that we're seeing across our entire portfolio. With the exception of our small part of our portfolio, which is our small business, where rates are flat and also our small program business, such as architects, engineers, accountants, small lawyers and realtors, where rates have been flat and retentions have stayed in the mid-80s, the rest of our portfolio is really under pressure from a rate standpoint and also obviously that's having an impact on our retention numbers because our whole focus is really protecting the bottom line. As you start looking at the middle market part of the portfolio today, that's a place where we're obviously seeing increased pressure. Our overall rates in middle market are down 3%, but that's clearly not reflective of what we're seeing in the marketplace. I think the advantage that we have with the strong distribution relationships is that in a lot of case with our portfolio and the overall value-added services that we provide, our agency plant has not been sending those accounts out to market. I think if those accounts really went out to market, we'd been seeing significantly more rate pressure than what we see in our portfolio.

Great. Thank you.

We'll take our next question from Bob Glasspiegel from Langen McAlenney.

Good morning. I didn't quite follow your answer to Jay on the -- how the hedge positives on subprime worked through the numbers. I assume it wasn't in the impairment part of net income, but it was in the book value; correct?

That's correct. But it also would come through the net realized line. It was an offset to realized losses.

Okay, so it was in the impairment -- it was in the realized --?

Yes, it came through the P&L.

Okay. And did you quantify how much that was?

I believe I did. I said we had -- I said two things. One, that we held about a $60 million notional credit-default hedge against subprime. I said that 25% of the impairment write-downs were subprime. So 25% of $122 million. And I said that largely the credit default swaps offset that approximate 25% of $30 million. So not 100%, but pretty close to it.

Okay, so the realized losses -- I'm microscoping the small number -- were related to other things in subprime?

There were other things in subprime in the realized losses. Yes, that's correct.

So it wasn't a subprime story, it was --.

It was not. No, it's an interest rate and credit spread widening story.

Okay. Well, treasuries rallied, so -- I know you don't own treasuries, but it's a mixed bag in how PC life companies did this quarter. Some companies are having gains and others are having losses, but you're say -- your stuff was more in the BBB -- double -- BBB and above getting hit this quarter, is that what you're saying?

No, that's not what we're saying, Bob. What we're saying is that -- if you think carefully about the marketplace over the summer, you'll recall that the fixed income market was characterized by a significant widening of credit spreads and in most instances, they offset the rally in treasuries. I think then the other thing you've got to remember is the whole -- and I hope we don't get into dominating the call again on impairment factors. Remember last quarter, I led you through the whole process we go through relative to what we might impair and what not, and most of those impairments, a large amount, are just simple interest-related impairments. And you'll recall I also said that we are -- if the market value of the security declines, it's going to show up in stockholder's equity. It's only going to come through unrealized or realized, and we are relatively indifferent to where that shows up and we're more concerned about maintaining our trading flexibility than anything.

Craig, you're totally misconstruing my question and where I'm coming from. Your AOCI didn't swing much at all this quarter. I was just trying to microscope the small delta. We can continue this offline, but I do understand what you're saying, but I don't think you're constructing where I'm coming from correctly. Capital management, where does that stand as a strategy for '07? I assume dividend will be the primary tool of getting capital back to shareholders, but if the top line -- if we are in a shrinking mode on the top line in a competitive market, which I think is what shareholders would certainly hope for, you'll be generating more capital in '08 than you did this year. Is the dividend going to continue to be the primary source of capital management?

Bob, this is Steve. I think the dividend is one part of capital management and all we can say at this point is that we increased it this quarter because we felt it was prudent and appropriate and that we will continue to look at our capital -- our capital and how we want to manage it going forward and maintain as much flexibility as we possible can.

Okay. Thank you, Steve.

And we'll take our next question from Bill Wilt from Morgan Stanley.

Good morning. A few questions, if I could. The first is a numbers. Jim, you'd given the new business from cross-selling at $118 million this quarter. I know you fairly consistently give that. Do you happen to have last year's number handy?

Hold on a second for that one and I'll try to see if we can find the exact number. I can tell you that as we've looked at cross-sell over the last three years, roughly a little over $2 billion of our new business has really come from a cross-sell perspective. And as you look at this quarter and for the first nine months of the year, cross-sell has been running about 38% of our overall new business. I would think last year that it was very similar -- it was a very similar number, but I'll validate that for you.

Okay, that's good. I'll hit you with another one while you do that. I guess similar related topic, retentions down 80 to 76 in standard and forgetting the exact delta in specialty lines. What -- as you look out over the next six, eight, 12 months, what are reasonable expectations for where retentions might go?

Well, we want to hold on to as much as our profitable business as we really can, but to the point where that we still can make a profit on that business. At this point, as we're starting to see pricing pressures, there's some of that business in which we see the pressure and see what kind of returns we would get in order to price our business at such low levels, we've decided to opt out and let that business go. We obviously like our overall loss ratios that we have on this book. We will continue to do the things necessary to retain as much of the profitable portfolio as possible, but it really depends on where the marketplace really goes and how overall pricing pressure continues to drive. We've now had several years of rate reductions. Fortunately, we've been able to keep our overall rate reductions at the low single digit, which is the 3% to 4% level and we're very comfortable at that level, but if all of the sudden rates took a significant nose dive and we're now looking at double digit increases for our portfolio, that would have an impact on the retention. If it continues in the same basis of how we've been able to management it today, with the 3% to 4% overall rate reductions, then I would expect to see similar numbers from a retention standpoint.

That's helpful. So realizing that retention is of secondary concern, I guess -- it's important for us, but perhaps secondary to the bottom line -- is there a point at which the market conditions and your reactions to them would be such that we can be looking at retentions and see a flashing yellow light? Is that -- if they drop down to the low 70s or high 60s, is there a point at which internally there's a flashing yellow or flashing red light as to internal prospects or marketplace -- the marketplace environment?

No, I think the advantage that we have is we have a very broad portfolio and within that portfolio, 40% of it is specialty lines, which has been running combined ratios in the 80's, 60% of it is our standard lines portfolio. We have shifted our overall book from a very large exposure base to a more small-sized accounts with low to moderate hazard exposures itself. Those particular type of accounts are less price sensitive. We manage and look at our pricing and our loss ratios for every line of business, every subsegment within our book. I think right now we've got a broad enough portfolio that we can still manage through the cycle and still hold on to our retentions. But we do look at each one of the subsegments to say what's happening today in that market, where is the pricing -- where is the pricing versus where we should be to at least hit our hurdle rates or at least cover the cost of capital and we're managing that appropriately. There's no question that there are parts of the portfolio in specialty that were given up more rate, but those are also places today where we can give up more rate when I look at where our combineds are.

Thanks for that. I have one more and then I'll requeue. The other one is on professional liability and exposure to -- from professional liability, E&O perspective to the whole subprime debacle, at our financial institutions or mortgage brokers, have you received any claims? Can you quantify your relevance in that market from a professional liability or insurance perspective and just add any color that you can on the prospect for insurance claims from that process?

I would look at it that, as we look at our exposures overall, any exposure that we would have with that, be it at a very high retention level and most of our book is really A-side D&O to date I expect there's very minimal exposure from -- within our overall portfolio to the subprime itself. You never say never that something's going to occur, but based on what we've seen today, we're very comfortable with our book and the fact that we've really shifted a lot of that to the A side and that we really have very high retention rates.

Thank you.

Bill, this is Steve Lilienthal. I just wanted to add something to that. I think Jim made a lot of points there that I just want to repeat. Of the three production metrics that we look at, rate, retention, and new business, we choose rate first, retention second, and new business last. And to the extent we write new business -- which has shrunk -- because of the market conditions we choose to be very selective and then cross-sell the daylights out of that to max out and leverage up on what we consider to be the prime or better accounts. So that's the one point that I think Jim made and that we've been making all along. But we -- we were trying to convey a message of consistency and predictability to you on the Street and I guess I'd want to remind you, this is exactly what we -- Jim and myself said that we would do years ago in terms of our response to softening market conditions. That we would allow our retentions to drop and certainly we have warning lights that we watch as things get below a certain level or our business -- the rates are falling below the levels that we would like to see for the returns that we need.

But we also gave you almost an exact number as to where we said this business would be when the market got to condition that it is in right now, which is in the high teens, exactly what we're doing. I just didn't want that point to get lost in the shuffle, I guess, so one other point I'd make, just to pick up on what Jim was saying, we continue to shrink our risk management portfolio during the course of the year, which has also contributed to the overall shrinkage of our top line. And we consider -- we will continue to shrink that because we consider for it to be a fool's play and probably the most sensitive part of the market right now.

Thanks for that, appreciate it.

I'm sorry, just to add one other point. With respect to the top-line production and the pressures, both Jim and I have referred to expense management and as in the past, we have demonstrated a willingness, an aggressive willingness, to deal with the expense structure of the business and size that structure to the size of the business. So if the top line continues to shrink or shrink faster than we anticipate, we will continue to take out costs and adjust the business to where we think it should be from a cost standpoint.

And we'll take our next --.

To Steve's point, we actually started that process in the third quarter of at least -- in our overall organization, which I think allows us to be much more responsive in this marketplace and also right size in our overall structure to fit there where the revenue is today.

And we'll take our next question from Dan Johnson from Citadel.

Great. Thank you very much. A couple of questions, please. Can you talk a little bit about the size of your construction business and whether or not that falls into -- or which of the two reporting segments? And then I've got some other questions after that.

Our overall --.

-- going on with that book.

Our overall construction book is around $1 billion in total gross written premium and it falls within our standard lines portfolio. I know a lot of our competitors actually show that into specialty, but we show that within our standard portfolio.

What do you see going on in that marketplace over the last couple of quarters?

Our overall pricing levels have held fairly stable from a rate standpoint from quarter to quarter. We do see more competition coming in from regional players that are now, as they're looking to grow, are really moving outside of their appetite of the lower end of middle market and really looking to now write and try to grow in the middle to upper-end of middle market, which is a place when a lot of cases they really do not have the overall service capabilities and/or the claims capability that we have. And we've been able to retain our overall book of business in construction because of the expertise that we provide from an underwriting standpoint, a risk control, and a claims. We know this business extremely well, we know the profit opportunities within this overall book, and our distribution plant really values the expertise and the services that we provide. And even though there's competition coming from regionals, we've still been able to retain our book.

Great. Let me then move on. On the investment side, first with the portfo -- excuse me, the partnership performance in the quarter, can you remind us again of broad construction of that portfolio such as private equities, whether it's hedge funds, et cetera?

No private equity, or, excuse me, very limited amount of private equity, and really it comprises a lot of different strategies across the portfolio. Very little leverage in most of those. They have real assets underlying them and there's going to be some variability quarter to quarter, but over the longer term -- and I mean over a lot of quarters and of a lot of years, they've been very positive contributors of investment income. We'd expect it to be the same.

It sounds like there's more in there than just investment funds, just by the way of your description?

No, it's mainly investment funds.

Okay. Then I'll move on to -- thank you for the disclosure on page 6. Very helpful. A couple definitional questions. You've got -- maybe in the second and third boxes, we've got subprime and then listed below that is other CDO. Can I infer that all of your subprime exposures are in that 903 total, or is there subprime CDOs that are in the line below it?

No, all of the subprime exposure is in the subprime line. So some of our -- so that amount of CDOs that have subprime collateral is a relatively small amount in that subprime line.

Yes. And then just the last question on that is the -- within the other CDO, the biggest piece of that's the single A at about $220 million. Can you just tell us a little bit about the assets that are backing that? Thank you.

It was primarily commercial real estate.

Commercial real estate, okay. And that would be separate from the commercial real estate that I guess I would have thought in the line below it?

No.

Yes, okay, I understand. So the -- right, that's the CDOs of the commercial real estate, whereas the --.

With commercial real estate collateral, right.

I understand. Great. Thanks again for that disclosure.

And we'll take our next question from Charles Gates from Credit Suisse.

Hi, good morning. My first question, approximate -- through this call you've highlighted how the greater pricing pressure is in what you term middle market. Approximately what portion of standard line sales were middle market in the third quarter of last year and the third quarter of this year?

I'll have that number for you in just a second.

I guess --.

If you look at our overall new business for the quarter, it was actually -- $73 million was what we had part of, $289 million so -- so I mentioned the $289 million is our total new business, so $73 million of that was middle market.

That was this year?

I'm sorry?

That was this year. But what would be the --?

Third quarter. If you look at prior, prior middle market overall new business was $92 million, part of $374 million prior.

I'm sorry, I'm confused. So what would have been the --?

(inaudible) new business. New business for the third quarter, our actuals for the third quarter '07 was $73 million. Prior was $92 million in new business.

And that's for middle market?

That's middle market. If I look at total written premium for the third quarter of '07, it was $388 million, prior was $444 million.

$444 million. How do you define middle market?

If you look at it as far as low volume type business that could be handled with our predictive models, which would be handled with less touch by the underwriters, it goes through our overall artificial intelligence system that we have, it's automated, mechanized, more single location. That's more of what we call small business. Anything beyond that would really fall into our middle market with the exception of the larger end of our risk management casualty, which would primarily be accounts in excess of $1 million in premium. And also all of our lost sensitive business really falls into risk management casualty. All the guaranteed cost in the middle market.

My second question, if you go back to the charge that was reported on September the 6th, are there other similar possible charges that the investors should be mindful of?

Well, I can answer that. This is Jon Kantor. We actually -- if you look at note G, which is our litigation note, there's five items there and three of them were settled in this quarter, which will be coming out in the Q in another day or so. So, no, I don't -- and the direct answer to your question is we don't anticipate settlements of anything like the magnitude of the Hancock case in our corporate litigation.

Okay. So you said there were five situations listed there --?

There are five disclosed cases.

And three have been taken care of?

Hancock, long-term care, and wage and hour, all three were settled in the third quarter. That leaves us with two and one of them actually -- the contingent commission antitrust litigation -- was dismissed by the federal district court in September, but we keep it in the litigation note because it's on appeal to the third circuit. That's CNA as a defendant along with something like 30 other industry players so really the note G risk profile is rapidly dwindling down, I would say.

So there are only two left?

Including the contingent commission matter that I mentioned, yes.

My final question what do you think $100 a barrel oil does to the commercial lines property casualty insurance industry?

I'd just say that's anybody's guess. I can't even speculate at this point what that does.

Thank you.

Due to time constraints, this will conclude the question-and-answer session for today's conference. I would like to turn the call back over to Ms. Nancy Bufalino for any closing or additional remarks.

Thank you, Connie, and thank you all for joining us today. Once again I call your attention to the disclosures concerning forward-looking statements and non-GAAP measures in the earnings release. A taped replay of today's conference call will be available for one week immediately following this call until November 5. Please see the earnings release for replay details. We appreciate your participation in today's call and thank you again for joining us this morning.

This concludes today's conference. We appreciate your participation. An immediate replay of this call will be available to you by dialing 888-203-1112 or 719-457-0820 and entering the replay passcode 3629040 followed by the pound sign. You may now disconnect.