The Travelers Companies, Inc. (TRV) Q2 2011 Earnings Report, Transcript and Summary

Good morning, ladies and gentlemen, and welcome to the Second Quarter Results Teleconference for Travelers. [Operator Instructions] As a reminder, this conference is being recorded on Thursday, July 21, 2011. At this time, I would like to turn the call over to Ms. Gabriella Nawi, Senior Vice President of Investor Relations. Ms. Nawi, you may begin.

Thank you, Jennifer. Good morning, and welcome to Travelers' discussion of our second quarter 2011 results. Hopefully, all of you have seen our press release, financial supplement and webcast presentation released earlier this morning. All of these materials can be found on our website at www.travelers.com under the Investor section. Speaking today will be Jay Fishman, Chairman and CEO; Jay Benet, Chief Financial Officer; and Brian MacLean, President and Chief Operating Officer. Other members of senior management are also in the room, available for the question-and-answer period. They will discuss the financial results of our business and the current market environment. They will refer to the webcast presentation as they go through our prepared remarks, and then we will open it up for questions. Before I turn it over to Jay, I would like to draw your attention to the explanatory note included at the end of the webcast. Our presentation today includes forward-looking statements. The company cautions investors that any forward-looking statement involves risks and uncertainties and is not a guarantee of future performance. Actual results may differ materially from those projected in the forward-looking statements due to a variety of factors. These factors are described in our earnings press release and in our most recent 10-Q and 10-K filed with the SEC. We do not undertake any obligation to update forward-looking statements. Also, in our remarks or responses to questions, we may mention some non-GAAP financial measures. Reconciliations are included in our recent earnings press release, financial supplement and other materials that are available in the Investors section on our website. With that, Jay Fishman.

Thank you, Gabby. Good morning, everyone, and thank you for joining us today. Given the fact that we disclosed as much as we did at our June 10 Investor Day, our results shouldn't be much of a surprise. In April and May, it was all weather all the time. Total after-tax catastrophe costs for the quarter were just under $1.1 billion, generating a loss for the quarter. The tornado and hailstorms in those 2 months were unprecedented in their size, scope, geographic dispersion and destructiveness. A loss of life was substantial, and our hearts go out to those who been personally impacted by the events. In that regard, our immediate focus is on delivering outstanding claim service and support to those who have had losses. Our entire claim organization is very much on the case and energized to do this right. We're well aware of the demand that events like this place on the claim organization, and we're thankful for their dedication and professionalism. We're also well aware that events of this magnitude will happen from time-to-time. It is critical that we price our product and manage our exposure thoughtfully. In that regard, we have great confidence in our data, analytical skill and underwriting expertise. Given the hail and tornado we experienced in the last few years, we are gaining a better and more granular understanding of the tail risk associated with these types of storms. Even in these early days following the April and May storms, we are focusing on improving rate and/or terms and conditions in those geographies that may be more exposed to these kinds of storms than history has suggested. Importantly, it is events such as these that reinforce our view that the correct way of defining success in our business is to measure returns over time. While we are certainly not pleased that we had a loss in the quarter, we know we are on the right path in running our business appropriately. There is no lack of enthusiasm or optimism here. In that regard, we are very encouraged that a number of important parameters of our business continue to improve. In Business Insurance, exposure in the quarter improved, and audit premiums, which were negative throughout the financial crisis, turned positive in the first quarter and grew meaningfully this quarter. Perhaps most importantly, we continue to be successful in delivering written rate improvements, particularly in Business Insurance. The overall pure renewal rate gain in Business Insurance, excluding National Accounts, was just over 2% for the quarter, demonstrating the success we are having in driving rate improvement. Retention remains at very strong levels, and all of the anecdotal feedback from our organization remains solid. We are executing on this strategy effectively. As we've said before, we are not waiting for some mythical turn in the P&C cycle to improve our pricing. We are driving rate improvement on a measured basis, particularly for those specific accounts where rate is needed to produce acceptable returns. We are also encouraged that there are signs that economic activity may be improving. While it remains somewhat anecdotal, I was personally encouraged by the article of yesterday's Wall Street Journal on C2, in which a number of banks reported an uptick in commercial lending. Early results from the credit card sector, as well as results from the tech industry over the last few days, are also encouraging. And while it does seem that we are coming down to the wire with respect to a federal debt limit solution, it may be that they are the beginnings of a bipartisan agreement. And it is encouraging that more states are dealing with their own financial situations, building foundations for an improved fiscal outlook. Before I turn it over to Jay, let me make a brief comment about our share repurchase philosophy. Simply put, it remains unchanged. Our premise for the creation of shareholder value is that we will generate top-tier earnings and return excess capital to shareholders through dividends and share repurchases. The fact that we had a loss in the quarter rather than our historical level of earnings impacted our share repurchases, but only in the second quarter. With less internal capital generated, there's less capital to return. On the premise that weather patterns return to more normal levels and our profitability returns to historical levels, and to the extent that the capital can be put back to work in our business for growth opportunities, we will continue buying back shares. Summing up, while the weather was challenging, it is the business we are in. We're doing all the right things and remain on the right track to create superior shareholder value. And with that, let me turn it over to Jay.

Thanks, Jay. There are a few things I'd like to highlight about our operating results in the quarter. First, in early June at our Investor Day conference, we disclosed an expected range of $1.0 billion to $1.05 billion after-tax and reinsurance for cat losses relating to the extraordinary tornado and hailstorm activity that had occurred in April and May. While our current estimates for those storms remain within that range, additional storms that occurred in June brought total cat losses for the quarter up to $1.085 billion after-tax and reinsurance. Second, we continue to see net favorable reserve development in each of our segments, although at a lesser amount this quarter in Business Insurance than we have experienced in recent quarters. Third, x cat in TYG [ph], underwriting margins were impacted by several items in the quarter. The re-estimation of certain first quarter non-cat winter storm losses in Business and Personal Insurance, adverse current quarter non-cat weather-related losses in PI for both homeowners and auto, as well as a larger number of large dollar prior losses in Personal Insurance. Adjusting for all of these items, underlying underwriting margins were generally in line with our expectations, which included as we've previously anticipated and disclosed, modest margin compression in Business Insurance, where earned rate increases were not keeping pace with changes in underlying losses. I'd also like to add that investment portfolio continued to perform very well. Net investment income was down slightly from the prior year quarter as the impacts of lower fixed income reinvestment rates and lower average assets, due to our share repurchase activity during the past 12 months, were mostly offset by strong performance in our non-fixed income investment portfolio. Finally, our net unrealized investment gains net of tax increased by $400 million from the beginning of the quarter from $1.8 billion to $2.2 billion. Book value per share of $59.62, although down less than 0.5 percentage point in the quarter due to the cat losses, has increased 2% since the beginning of the year and 7% in the past 12 months. Year-to-date we repurchased 1.3 billion of common stock including 237 million in the second quarter and paid $330 million in dividends. While in the past 12 months, we repurchased 3.5 billion of common stock and paid $659 million in dividends. Our capital management strategy worked very well in the quarter and remains the same as it has been. Despite the extremely large cat losses, there was no meaningful change in our balance sheet strength between the beginning and end of the quarter. Shareholders' equity and stat surplus was still over $25 billion and $20 billion, respectively. Debt-to-total capital was 22.5%, well within our target range. And holding company liquidity was $2.4 billion, more than twice our target level. And based upon first half results and consistent with our announcement at Investor Day, we expect common share repurchases in the second half of the year to approximate $400 million in excess of second half operating income. Finally, I'd also like to point out that we successfully renewed our cat treaties, keeping essentially the same structure, but at a slightly higher cost. The new treaties are outlined on Pages 19 and 20 of the webcast and are described more fully on our second quarter 10-Q, which was filed earlier today. With that, Brian's now going to provide some further insight into our results.

Thanks, Jay. The obvious headline for our results this quarter was the extraordinary tornadoes and hailstorms that impacted the industry. To give you a perspective on the magnitude of these events, when measured as a percentage of our earned premiums, our cat losses this quarter were 12x our 20-plus year average for second quarter cats and almost 4x larger than our next largest second quarter, which happened to be last year. So clearly, the level of catastrophes we experienced this quarter was extraordinary. And in addition to the weather events that hit the catastrophe disclosure threshold, our second quarter combined ratio was significantly impacted my non-catastrophe weather events and developments on the first quarter winter storm activity. The significant and unpredictable impact of weather, notwithstanding, we remained pleased with the underlying performance of the franchise and continue to see positive signs across our businesses and in the operating environment. Beginning with Business Insurance, we had another very strong quarter for top line. Net written premiums continued their positive trend and were up year-over-year, driven by increases in pricing and exposure, along with modest account growth. Retention and new business levels, although down in the quarter, remained at high levels. Looking at pricing, the momentum we saw in the first quarter accelerated, as the rate was up 2 points compared to 1 point of positive rate in the first quarter of 2011 and negative rate in the second quarter of 2010. We are encouraged that rate was positive for all lines of business, and in fact, increased in all lines of business from last quarter, with workers’ compensation leading the way. As we discussed last quarter, our pricing strategy is to actively manage rate in a thoughtful and targeted manner that seeks to minimize disruption to our agents and customers. We believe our results this quarter demonstrate that this strategy is working. We also see momentum in exposure, up over 1 point in the quarter compared to being essentially flat both last quarter and the second quarter of 2010. This growth in exposure, combined with the audit premiums that were positive for the second consecutive quarter, demonstrates that the improving economic trends we noted in the first quarter have continued. Turning to operating earnings. We were impacted significantly by the cats that I just discussed, as well as by less favorable prior-year development versus the second quarter of 2010. Second quarter results have also been negatively impacted by approximately $40 million after-tax or about 2.2 loss ratio points for Business Insurance, due to the re-estimation of certain first quarter losses, primarily related to the development of non-catastrophe winter weather, and to a lesser extent, workers’ compensation. Overall, if you look at underlying Business Insurance margins on a year-to-date basis, they are about where we expected them to be. Before I go to the other segments, I want to take a few minutes and talk about workers’ compensation because there's a lot of anecdotal marketplace conversation about the line. Comp is one of our long tail lines of business, with a duration of about 8 years on a new block and 11 years for the reserves. We actively manage all our lines of business, and in the case of workers' comp, our conclusion is that in the aggregate, we feel very positive about our book. On the severity side, we've not seen any substantial change in the risk. Medical cost trends continue around, in fact, slightly below our expectations, and lost time activity has remained stable. We are encouraged by the reform activity already approved in numerous states and under consideration in many more. These are obviously aggregate marketplace comments, and there are certainly jurisdictions where we see more challenging environment. And accordingly, we are taking the necessary actions in those states. For example, in one state where we are seeing increased losses along with a more difficult legislative environment, we are achieving rate gain significantly above the average for the line, and in the case of Commercial Accounts, well into the mid-teens. We also feel good on the frequency side. But there's more texture underneath the trend, and I want to share with you how we look at it. The graph on Page 12 of the webcast shows our workers' comp frequency for lost time claims from 2006 to the first quarter of 2011. The green curve is actual claims frequency, or more specifically, the fourth quarter rolling average of our reported lost time claims at 12 months. Through 2006 and 2007, the trend was consistent with the long-term average of about a 2% per annum decline in this measure of claim frequency. The dotted blue line on the graph shows a hypothetical trend of reported claims, if we simply extrapolate the historical 2% per annum reduction. With the beginning of the financial crisis in 2008, the actual trend began to diverge from this historical pattern. And by 2009, comp frequency had dropped to levels that were unusually low, and in our opinion even at that time, were not sustainable for the long term. Consequently in the aggregate, we were not pricing new or renewal policies, assuming this unsustainable level of frequency would continue. Off of these lows, we've seen an increase of claim frequency, and it is happening somewhat sooner than we had anticipated. It's important to note that the frequency is still below the extrapolated pre-crisis trend line, which, again, represents a 2% per annum drop in the measure. And while it's early, the most recent data is beginning to point to a flattening. We believe that the movement of frequency trends in the last 3 years have been attributable to the economy. And in that light, recent uptick could actually be evidence of some economic improvement for the classes of business that we insure. So bottom line, although there's been some movement in frequency, it's still at levels that are consistent with long-term improvement. All in, we believe the prospects to the line are sound, but we're not standing on the side lines relative to pricing. In the aggregate, we are driving for rate. And we are encouraged by the fact that our most significant written rate gains in this quarter are in the workers' comp line. Turning to Financial, Professional & International Insurance segment. Operating income was down modestly from the prior-year quarter, reflecting lower net investment income. Underwriting gain saw a slight decrease, which is the net of a few items, primarily more favorable prior-year development offset by an increase in catastrophe and non-catastrophe weather. Excluding cats and prior-year redevelopment, underlying loss margins were consistent year-over-year, while expense dollars were up modestly due to the investments we continue to make in this business. In International, net written premiums were up year-over-year with strong results in Lloyd's, Canada and the U.K., more than offsetting the impact of the termination of an exclusive distribution relationship in Ireland. In both Bond & Financial Products, surety results continue to be impacted by sluggishness in construction spending. We believe we offer a unique value proposition in this business, and we are very satisfied with the margins and the quality of our portfolio. In management liability, we saw significantly improved production results in the quarter and remain satisfied with our overall returns. Finally, we completed the acquisition of our joint venture interest in Brazil, which gives us an ownership position in the market leader of the Brazilian surety business. We're pleased about the opportunity to leverage our leading U.S. surety franchise, as well as our other franchises to expand beyond the surety business into the growing property and casualty markets in Brazil. Moving to Personal Insurance. The weather in the quarter, both catastrophic and non-catastrophic drove the operating loss. Prior-year development for the segment was up year-over-year, driven largely by the favorable development of 2010 cats. Excluding cats and prior-year development, the Personal Insurance combined ratio increased by approximately 5 points quarter-over-quarter. The majority of the increase was driven by adverse non-cat weather losses in Homeowners and even Auto, some of which related to first quarter storms. A smaller financial impact resulted from a higher number of severe fire losses in Homeowners. Turning to production results. We had another good quarter with both lines of business posting growth in policies in force and net written premiums. Renewal premium change was up year-over-year in both Auto and Homeowners and other. Retention remains near historic highs for both lines and new business is solid, but down slightly from recent quarters. So in summary, while this quarter was a strong reminder of how the unpredictability of weather can drive results, we remain very positive about the underlying strength of the business. We continue to see some improvements in the market environment, and we believe we are well positioned for success going forward. With that, Jennifer, we'll now open it up for questions.

[Operator Instructions] Our first question is from the line of Mark Gelly [ph] from Barclays Capital.

This is Jay Gelb from Barclays Capital. I wanted to touch base on the environmental increase on the reserves, and if that could have any implications for asbestos, given what we saw at the Hartford?

Jay, this is Jay Benet. As it relates to environmental, I would say that look at the environment itself, there really hasn't been a major change in the environment. We experienced the same kind of favorable trends in claim payments that we had been experiencing. As we've said before, we extrapolate information that we have into some formulaic views of what the ultimate loss costs are going to be. And it doesn't take a lot of change in the actual environment to make for some dollar changes in the reserves. So while I would say we saw the same favorable trends, the favorable trends were a little less than we had anticipated in our calculation. So that was the adjustment, nothing major going on. And as it relates to asbestos, we really don't have any comments to make here. I'd say, overall, the environment for asbestos has not changed fundamentally from the way we viewed it last year. And as it relates to where we are, as you know, we do our quarterly procedures on asbestos. The quarterly procedures didn't lead to any kind of change in the reserve. We also do a real detailed overall claim review that we complete annually. We're expecting to complete that in the third quarter. And at that time, we'll see what the asbestos reserves will look.

And then a separate issue on capital management. Am I right in thinking that the debt-to-capital level is more of a constraint in terms of how much stock Travelers can buy that at this point? If it's 22.5% now debt-to-capital, what's the kind of range that the company is comfortable with?

We generally talk about a range of somewhere between 15% and 25% associated with a AA rating. And one of the things I'd point out about the 22.5%, there's about a point of that, that relates to essentially prefunding some future debt. We took advantage of the low interest rate environment last year to do some retirement of our convertible hybrids that we had. But we also had some excess that we just kept on the balance sheet to be able to use to pay off future debt. So as a result, as it relates to that range, we do have lots of flexibility in the range. And we'll continue to manage the overall capital position with that in mind, with the operating company capital being first and foremost, as well as the liquidity at the holding company, which we've already said many, many times has a target level of about $1.1 billion.

Okay. And I just want to clarify that. So does that mean Travelers is okay, going up to 25% debt-to-cap? Or would that be too high to be at that end of the range?

As far as the rating agencies' discussion, if we did it in a way that took advantage of low interest rates. Well I mean, we have flexibility.

Yes, but I wouldn't -- Jay, this is Jay Fishman. The question, "Is Travelers okay at 25%?" implies, or the question may imply that we intend to leverage up for the purpose of buying back more shares. And that's just not how we think about managing our balance sheet where the range, at least, as we understand it, that supports a strong AA rating in our business, which is where we belong, is in that 15% to 25%. We always want to leave some flexibility to deal with unexpected issues. We recognize the kinds of events that we saw this quarter could happen again, and they could happen even worse. And so we don't run our business at the edge. But it's that range that really establishes the notion of that 15% to 25% range, is the range that we understand, is in support of a AA rating.

Our next question is from the line of Matthew Heimermann from JPMorgan. Matthew Heimermann - JP Morgan Chase & Co: I guess I just wanted to follow up on the asbestos question. Can you just remind us, because I don't think you report this anywhere anymore, just what survival ratio looks when you adjust for buyouts and settlements?

Not really a statistic I keep at the top of my head because it's not something we think is terribly relevant. We can calculate -- we get it for you though. And we -- the way of looking at it is a good way of looking at it, so taking out the settlement activity, but we just have to get the data. Matthew Heimermann - JP Morgan Chase & Co: Yes, that's fair. I just wondered. I can do the raw numbers, but I suspect there's probably a material difference. So I just want to make sure I'm thinking about it correctly. I guess the other question, maybe for Brian is just could you give us a little bit more color on the rate increases by kind of core product line? And I guess I was thinking about maybe how work comp contrast with property, and how that contrasts with maybe general liability and then maybe other specialty liability lines?

So broadly, as I said, I think, in a couple of spots, the largest increases are in comp. Positive increases everywhere, but consistently, what we in the market have been saying for probably the last year. Property's probably at the tail end of it. So strongest increases in comp.

When you say property, you're really thinking about Business Insurance.

Yes. It's all about Business Insurance.

Yes.

So on the first party's side, and probably over the last couple of years, the larger the account, the more challenging the pricing environment has been. We're seeing improvements everywhere, so we're encouraged that, again, even on the property side, it's gone from negatives to albeit small, but small positives. We're also seeing in the large property world, a little bit of favorable movement on the terms and conditions rhetoric in the marketplace. And so that's very encouraging in that business and held somewhere in between. Matthew Heimermann - JP Morgan Chase & Co: All right. So if I -- just to restate what you said, if I could kind of think about where comp greater than the 2, pretty much property less than 2, pretty much everything else was probably clustered around the average?

Yes, both , yes.

As it relates to the survival ratio that you're asking about, x the settlement activity, it's probably close to 8%.

Our next question comes from the line of Jay Cohen from Bank of America Merrill Lynch.

A couple of questions. The first is just a numbers question. In the interest and other segment, as you break it out, the G&A number was quite a bit higher than it had been running. And I'm wondering what's behind that. The second question, you kind of alluded to the fact that, we're obviously dealing with a debt ceiling at this point and there's been obviously some noise about a potential downgrade for U.S. Treasury securities. Have you thought about the implications? Or should that happen, hopefully it doesn't, but should it happen, what it might be for you, specifically maybe annuities might get downgraded as well?

On the last question, Jay, I -- we don't know. I mean, that's the simple answer. It's unimaginable, right, unthinkable that a sovereign entity that has a printing press in the basement somehow makes a decision to stop printing. We don't -- and so I simply don't know. We don't know the consequences of what a technical default, a partial default or anything else might be. We'll play it out as it comes.

Fair enough, that's my answer too. I have no clue.

As far as what you are seeing in interest in other, from time-to-time, we have some items that will impact that. As a rule, we allocate costs throughout our business segments to minimize the amount of costs that's sitting in the interest and other. So it's got a very low base. In this particular case, there was a legal matter that was settled and it impacted it. But it's a nonrecurring item, and that's as much as we'd like to share about them.

Our next question comes from the line of Amit Kumar from Macquarie.

Just going back to the discussion on asbestos, one of the other companies recently added to their reserves. And they talked about increase their claim activity against smaller peripheral insurers. Am I understanding this correctly that you are not seeing that trend in your numbers?

I wouldn't say that at all. I'd say we've been seeing that trend a long time, and our reserves are reflecting that.

But you are not seeing any comparative adverse trends?

Not based on what we had expected.

Okay, that's actually helpful. The only other question I have is on the new business. If I look back over the past 4 quarters, it has trended down. And I'm just wondering, is that a function of your attempt to drive pricing? Or is that a function of competition?

I would say it's hard for me to separate those a little bit. But certainly as any carrier begins to drive for some rate improvement, you're going to get some reaction in new business levels and you're going to get some reaction in retention levels. So that's what we've seen on the Business Insurance side. And we're very comfortable with the trade-offs that we're seeing there in the activity.

Why don't you comment on our view of the pricing with new business?

Yes, I think that one of the things we are encouraged by directionally is the last quarter, certainly, maybe 2 quarters throughout this year. That environment, we think, has actually improved slightly. And there's always a delta between, as we've talked about a zillion times, a delta between renewal and new pricing. But that is -- that has shrunk over the last couple of quarters. And so directionally, we feel that's moving in the right place.

We've actually gotten a couple of things going on there. The first is, obviously, as renewal pricing lifts, you're working off a higher base. And then I think importantly, but I wouldn't declare a definitive trend here that's unassailable. But importantly, the spread between new and renewal has narrowed, meaning that the gap between the renewal price and new price was wider. Previously, it's more narrow off a higher number. So Brian's comment about it improving is a complex one and really is impacted by both the level of renewals as well as that spread.

And I didn't know if you were talking about just BI. But on the Personal Insurance side, we think there's some of the movement there as the economy and slowdown in new auto sales and shopping and home sales. And then some of it is probably a little bit of reaction to our pricing strategy there.

Our next question is from the line of Brian Meredith from UBS.

Two questions here. The first one just with respect to workers’ compensation. You made a comment that you were pricing in line with the trend line versus -- extrapolated trend line versus back in '09. I'm curious, do reserving practices follow the same kind of pattern as pricing.

Yes. First...

We're going to [indiscernible]. Go ahead.

Two things. One is we were not pricing fully up on the extrapolated trend line either. We were somewhere in between. And what I think any carrier is going to do is they're going to look at changing trends and they've got to decide, "Do I ignore it? Do I fully incorporated it? Or do I look at someplace in between?" And we were someplace in between, and I -- probably, my comments didn't make that crystal clear. So we didn't fully incorporate the downturn. And now I've -- from a reserving side, yes, we will reserve and price somewhat differently in that. As we look at trends and as they emerge, we should appropriately reflect the actual trends reserving for the business that's written and the accident years that those losses are coming from. So we more incorporate into our reserving the facts that we see. Obviously, we're coming up with an estimate of what we think the ultimate is going to be out of those.

With the exception of Brian's comment in his prepared remarks that the lift off the bottom was somewhat faster or sooner. However, one wants to interpret it, then we had originally assumed. We've actually, so far, done a good job of ultimately picking the developed numbers pretty, pretty close. We've had a series of years of favorable development in workers' comp because our assumption about development was actually not that. It was less than that. And with the exception of how fast we assumed that bottoming out would move back, is the only area where we missed a little. And I do think it's actually missed a little. I mean, many of you would disagree with that.

Yes. The other point I'll emphasize, Brian, is, again, the obvious, but that's just frequency. So on the clarity side, we feel pretty good about the trends.

Terrific. And then the other question, Brian, can you just tell a little bit of, obviously, excluding comps, because you've talked about that, just loss trend, what's going on with loss trend in commercial and then maybe even in personal lines, personal auto?

Yes, on the commercial side, there's really been no dramatic change. I would say on the encouraging side, we continue to feel pretty good about what we see out of the tort environment and how that's been progressing. And otherwise, the trends are pretty consistent with what we've been seeing. Greg, why don't you -- on the personal lines side, Greg Toczydlowski.

Sure. Yes, Brian. Similar story on the Personal Insurance side. Good news on the frequency side, pretty much contained our loss cost trend. We see a little bit of pressure on the severity side. And Brian's comment really talked about that this quarter in terms of the non-cat weather driving that. So we don't believe that's an ongoing trend. We believe that's some fluctuations, so we'll continue looking at the business more over a longer term.

Yes. One place where we see a little difference between the commercial and the personal book, quite frankly, is a little bit on the frequency of the auto side. On commercial auto, we are seeing a little pickup in frequency, and that, we think is pretty directly related to a pickup in economic opportunity for our insured base. And we're not seeing it on the personal lines side because we don't think there's pickup in the miles driven on the personal side. So a little bit of a difference there, but small movement.

Our next question is from the line of Michael Nannizzi from Goldman Sachs.

Just sort of a quick one on workers' comp. I'm sorry to keep hitting this but -- after the changes you made in the first quarter to that book, can you talk about the difference between premium trend and loss trend? And especially given that you've seen that book increased a lot or, at least, written premiums has increased a lot in the second quarter. And I just have one follow-up.

We're going to resist the temptation to make a declaration and try and compare written rate to earned loss trend. We spent some time yesterday trying to understand that issue with real clarity. And candidly, you make so many assumptions about will written trend continue and will -- I'm sorry, will written rate trend continue? And what will happen to earned loss trend, that we're just going to avoid any predictions about whether written trend is running higher than earned loss trend. And I think the numbers, at least for the moment, speak for themselves. I think what we're comfortable saying, and correct me because I'm -- if you think I'm wrong, is not we're not losing ground. I think that's the important comment. We don't believe -- at least at the moment. That can change next week. That can change next month either way. But we feel pretty good that we're not losing ground. And that's, I think about as far as we want to go in an outlook kind of statement about workers' comp.

Okay, that's very fair. One question on Commercial Accounts, I guess, I imagine a lot of the uptick there is also comp. But you're seeing pricing or you're taking pricing higher competitively on those accounts or just in the business you're writing. What's happening? Is it that others are not following? Or is it that people are taking their rates higher? Any notion on what's causing both a raise in rate and also the opportunity to hold business? And I see you retention, but also pickup new business.

Yes, I think it's -- so there's a lot there in your question. We've talked a lot about what we do as a franchise. And we think we've got some real market-leading, distinct competitive advantages in those businesses that, that customers' value and enable us to keep our business and keep it at appropriate rates of return. I'm not sure where the broader market is when -- because I sense a little bit implied in your comment was we're doing distinctly better than the marketplace. I would hope we are. But I don't know exactly what those numbers are.

And just to add to that -- this is Bill Cunningham. I think the other thing you have to keep in mind is we don't approach every account the same. This is not a one-size-fits-all in terms of how we approach each account. We try to be as thoughtful as we can in terms of rate need down to the risk based under the risk characteristics that we're seeing. We manage things differently office-by-office based on what we're seeing. Line approach may vary by state. So it's thoughtful, as weighted as we can be. I think it's having a big difference in terms of [indiscernible].

I think one of the things that we take for granted that we tend not to speak about in these comments is the risk control approach that we take to workers' comp. It's not an open-to-buy book. It's not that we underwrite any comer that happens to walk into an agent's office. There's an active and engaged risk control process, a substantial engineering group that for virtually every new account in Commercial Accounts -- again, if I'm overstating, correct me. But virtually every new account in Commercial Accounts, we've got our own engineers out in the field, actually observing the risk control practices and procedures of the accounts that we underwrite. And I would speculate, and it is just speculation, that what we're ending up with is a better class of workers’ compensation business from a loss management perspective than carriers who don't bring that level of expertise to the market. And there's an interesting phenomenon to talk about, about what happens to those classes of business and their unemployment statistics and their way of dealing with injuries that's different from the average marketplace. I do think that our risk control organization adds real value here and ends up creating a block of business, a book of business that's just different from lots of other people who write more generally. Not everyone, as there are other people who bring that level of expertise also. But I think relative to those who don't, we end up with a very different managed book of business from a loss perspective.

Our next question is from the line of Paul Newsome from Sandler O'Neill. J. Paul Newsome - Sandler O'Neill + Partners, L.P.: I believe, from the presentation we saw at the Analyst Day, that you folks have been having a pretty heavy push into the Midwest in your personalized business. And I'd like to know, given what's happened in the last quarter and we sought that -- a broader sort of secondary question, should we, in general, think of because of the geographic changes in your book, having a larger cat load, overall, in your personalized business?

A couple of questions there, and I, first, I'm not sure that there've been any geographic changes, in particular, in our book with respect to your last question. I'm going to let Greg speak to the distribution. But to the extent that one concludes that the tornado peril is more prevalent than it was previously, it's a challenge to deal with that from a risk selection and underwriting point of view. It's difficult for us to determine the difference between Charlotte and Atlanta and Dallas and Houston and Joplin and frankly, Springfield, Massachusetts in this last quarter. So as we think about how do we manage this more effectively, it really does come down to rate and terms and conditions a lot more than it comes down to geographic selection, relative to trying to manage the peril differently. It's one thing to manage coastal exposure. We know where the coast is. It's another to try and manage broad-based tornado exposure because it is just so random, it is so random but so intense at the point of storm. So it's a real tough one to deal with from the selection standpoint.

So, Paul, let me -- this is Brian. Let me take a shot at I think addressing the broad question you're asking, which is, is our exposure going up here? Is our estimate of cat activity going up? Then I'm going to throw it to Greg on the specific on the geography. I think it's fair to say for every underwriter in this industry, while looking at RMS 11 and looking at the wind hail activity for the last couple of years actually, but certainly, this year in particular, we're looking at higher rather than lower. But we're still in the middle of evaluating that whole process. But it's not fundamentally driven by our geographic shift. It's driven by the obvious concerns with the changes in the weather patterns and how we impact that -- I mean integrate that. So Greg, if can you talk about geography? It moved a little.

Sure, yes. We've -- I think we've talked in the past about our agency expansion strategy. And we have had a disproportionate amount of new agents coming in the Midwest and West just because of our heavy penetration in the Northeast. So I think that's some of the data that we have shared with you. We've been trying to respond to a lot of the rhetoric in the marketplace. Is this a new normal for weather? And we just got to go back a couple of years in 2006, 2007 when there was relatively no catastrophes in the marketplace over those 24 months. And we certainly -- the marketplace wasn't saying that's a new trend at that time also. So we're certainly not in the business of guessing what the weather is going to be in the future. We're being very disciplined and focused around our pricing in terms of [indiscernible] and I feel very comfortable around our footprint.

Our next question comes from the line of Vinay Misquith from Evercore Partners.

Two questions, the first one was on workers’ compensation. I was curious whether some of the adverse prior quarter reserve development came from workers’ compensation?

Just in looking at the overall line, there was no adverse development for workers' comp this quarter.

Prior year.

Yes, prior year than this.

In the prior and prior quarter. Okay, great. The second question is...

No, no, no. That's the prior year. Yes, yes.

No, no, no. Let me just -- because everything before January 1 of this year, in any given quarter, we're doing full analysis of what our reserve needs are. There's always going to be some prior years, we're going to up a little bit, some will go down a little bit. But net-net, there was no prior year development of any magnitude for workers' comp.

But Brian did say in his comments, because again your question, I thought, was quarter-to-quarter.

Yes.

Some of the re-estimation of the...

Across the year, yes.

Some of it was due to the winter storms, and to a lesser extent, workers' comp. What underlies that is a projected increase in the current year loss ratio for comps driven by this movement coming back to trend line faster than we had anticipated. So there was the impact in the second quarter. These, by the way, relative to a company of our size, are relatively small numbers. But what you -- what did was we upped the loss ratio of some a little bit.

For the current year, right.

For the current year. And we obviously had a catch-up in the second quarter relative to the first quarter re-estimation. Mike, did I answer it [ph] right?

Okay.

Fair enough, understood. The second question is on Homeowners. You are taking up rates 9%. Just curious as to what reception you're getting from the regulators, as well as your customers on the rate increases.

Yes. This is Greg again. We're, as you see, we've had 9% for 2 consecutive quarters. And currently, we live in a regulatory environment, and so we need to get those approved, either file a news [ph] or prior to approval. And we've got very good relationships with the regulators and our distributors. And there are always cost-based reasons around why we get the rate level, and we've been successful in implementing those. So we obviously are seeing the results around that particular product line in this quarter and believe they're warranted. And we'll continue to watch that and be prudent with the rate going forward.

Sure. I was just curious as to why you're not seeing sort of a drop-off in the retention ratios in Homeowners because of rate increases? Would that be because most competitors are taking rates also by some of the levels?

Yes. We clearly are operating as a function of the market and we're not the only one feeling these catastrophe losses over the last 2 years. So there clearly is an increase in rate environment on the property side. We watch the marketplace quite a bit in terms of the amount of filings that come through, increases, the decreases and the simple ratio. And it's close to a double-digit ratio right now, what the marketplace is doing in that particular line. So clearly we're not the only one taking some rate in that particular product.

Our next question is from the line of Adam Klauber with William Blair. Adam Klauber - William Blair & Company L.L.C.: You mentioned exposure growth in the Business Insurance is moving positive. How big of an impact are payrolls on exposures increasing?

In comps, it's the whole game. So it's a -- but the biggest driver in our overall results would be GL, which is fundamentally driven by sales and receipts, business receipts. That's the biggest driver. But you can look at our book and you can look at the relationship of comp. And so payrolls have been moving up and are certainly a piece of that.

Projected payrolls.

Yes.

Exposure is a projected number that our insureds provide us at renewal.

Yes.

And then obviously, it gets trued out at audit premium time down the road in the middle market. And so what we're seeing is it's actually there are 2 manifestations of a better environment. First is that insureds are perceiving the environment better, meaning they're anticipating some payroll growth. And then what we're actually seeing in the audit premium numbers in the aggregate. I don't know how the audit premium relates down narrowly to workers comp. But what we're seeing in the audit premium is that in fact, the activity that our insureds, economic activity, ratable activity that they're experiencing over the policy period, is turning out now to be more than they had originally anticipated. So there's 2 elements to it. They're both worth thinking about. Adam Klauber - William Blair & Company L.L.C.: Great, that's helpful. And also in Business Insurance, very positive that you're getting rate on the renewal book of business. Is it a several year gradual process to get rates where you want?

Actually, it's a very good question and I'll ask Bill to chime in as well. We -- the history on this is that we wrestled for a long time as to how to respond to the precipitous decline in investment rates, which obviously form an important part of the profitability of our business. And when investment rates fell in connection with the prices, it obviously impacted the profitability prospectively of the business that we were writing. The notion that we would have a knee-jerk reaction and attempt to increase prices precipitously, dramatically, never really occurred to us. We always have a mindset about making sure our agents are comfortable with understanding what we're doing and why. And that our insureds understand what we're doing and why. And importantly, we weren't smart enough to know when the investment market would return to more normal levels. And as I commented last quarter, we're beginning to contemplate the notion that the investment environment that we're in may indeed be longer rather than shorter. That obviously back in the third quarter, I think, of last year set us on a path of trying to get rate on a measured basis, not -- being mindful of not disrupting agents' business and certainly not disrupting insureds business. But to do so thoughtfully. If you've got a $100,000 account and you got a 2% rate increase, it's $102,000 premium. And hopefully, we bring value enough to both the agent and the insured that, that's a transaction that we can be delivered to the insured without great concern. So far, so good. But it is so far. And I would answer your question just saying, yes, we think about this as a multiyear effort. It's not -- our strategy in the pricing environment remains intact, driven to a great extent by the declining -- by the reduction in investment rates available and just thinking about the aggregate returns in our business. Now having said that, we balance that out against the notion that cost of equity has dropped dramatically over the last 3 or 4 years for most good companies, anyway. We put up a slide in the Investor Day that showed our historical cost of equity. And then we got into an interesting question of, what happens to the long-term cost of equity? Does it stay down at the levels that they're at, that it's at? And therefore, what levels of returns are going to be available in the marketplace before our competitors step in? And so those are all of the variables that we're wrestling with. The conclusion of all of that, though, is to move for rate gains on a measured basis, thoughtfully, not disruptively. Any of those concerns are dealt with by a measured pace of getting rate gains.

Our next question is from the line of Josh Shanker from Deutsche Bank.

On the charts that you showed about renewal rate change, I'm just -- given the fact there was unfavorable development on 1Q 11, does that change your prior opinions on renewal change? Or is that an absolute number or based on loss cost trends?

The renewal rate change is a pure rate number. And I'm looking over at Bill and Fred Donner, a pure rate number calculated at the time of renewal. And so it's independent of loss cost trend. If we had a policy that was $100,000 and it's renewing with exactly the same exposure and exactly the same terms and conditions, and we do our best to analyze them at this level, and it's renewing at the $102,000, that's a 2% rate gain. And so it's calculated, it's expiring premium relative to renewal premium. And it's independent of any loss trend analysis or re-estimation of loss rates.

Okay, then. Can you comment, in general, on where you see your current growth relative to the rate of loss growth?

Yes, I commented on that before. We're going to resist the temptation to make predictions or estimations about whether the rate gains that we're getting, once they are earned, will then offset the loss trend that we are experiencing at that point in time. What we're pretty comfortable saying at the moment is that we're not losing ground. We're not losing ground. There is a comment, as there always is, in the Outlook section of the 10-Q that we filed earlier today, that speaks to what we expect underwriting margins to be. Now remember, that's earned. What's in the 10-Q is purely earned. So it has nothing to do, obviously, has something to do. But it's not a direct function of written rate gains right now. But there is a comment in the Q about what we anticipate underwriting margins to be. But the rate change moves each quarter. The loss trend moves each quarter. And so any predictions that we make are going to force us to constantly go back and revisit whether we were right or wrong. And we're going to just resist the temptation.

In my comments, I said that if we look at underlying Business Insurance margins on a year-to-date basis, they are about where we expected. There's always little movements. But we're talking real little movements here. So rate might be a little bit better than we expected. Loss trend might be slightly worse. But it's netting out to about where we thought.

And are there any other unusual items besides the catastrophes in the auto combined ratio number? Or it was somewhat higher this quarter?

It was really all weather-related, either catastrophe or non-catastrophe had a weather impact on it.

Partly, just to reinforce that, it's -- I do want to say it's unusual. But most people don't think about the fact that a hailstorm that doesn't qualify as a cat, could, in fact, generate auto losses. So a small hailstorm can, in fact, produce non-cat weather Auto losses. And we've had experiences in this quarter of non-cat weather producing auto losses.

And can you reiterate your threshold on what's the -- where weather then becomes a cat?

I don't believe we...

Yes, we've never...

We've never -- I don't believe that we've disclosed that.

Our next question is from the line of Larry Greenberg from Langen McAlenney.

I'm just wondering if you might be able to talk qualitatively about Business Insurance reserve development, which is positive, but down from prior-period levels. And I know the environmental was a portion of that. But I guess more specifically, I'm wondering if workers' comp might be playing a pretty significant role in that, given that the trend line was favorable for a while. Now you're saying that it's -- loss trend is moving up a bit more rapidly. Is that a meaningful piece of the lower favorable development? And maybe anything else you might be able to add on that front.

First of all, I'd go back to what we were saying earlier. We look at our reserves in a very full-some way each quarter. And we evaluate movements and the net result of it is what you see this quarter. I wouldn't read too much into it in relation to their level this quarter versus prior quarters. As it relates specifically to things like workers' comp or general liability or the other components that make it up, if you go back and read through our quarterly 10-Qs, which I'm not urging you to do. But if you did it, you would see that each quarter, there are different elements contributing to the net favorable development. So it's not like you look at workers' comp this quarter versus last quarter versus the fourth quarter versus the first quarter this year. You're going to see any kind of pattern, that say, one particular line has been favorable, favorable, favorable and suddenly it's not. It's a process that we go through that just looks at what's developing in the data. And we make adjustments when the data suggests we should be making the adjustments.

Our next question is from the line of Greg Locraft from Morgan Stanley. [Technical Difficulty]

[Operator Instructions] Our next question is from the line of Mark Dwelle from RBC Capital Markets.

A lot of my questions have already been asked and answered. But the one topic I wanted to touch on briefly, in both the rate and the volume environment on commercial, are you seeing any big disparity between relatively large accounts, mid-sized accounts and small accounts? We've heard a lot of commentary from brokers that there's that type of a divergence.

Yes, and the answer is yes, but nothing new in that regard. These are comments you can go back over the last several years, and you'll see us saying the same thing. The smaller account business has been less subject to rate volatility than the larger account business, and the middle market business falls right in between. What I always try to reinforce to people, because I do think that there's any number of folks who kind of misperceive us, what it is we do, we are not broadly speaking a large account casualty writer in particular. We are a small and middle market company that comprises a substantial percentage of our Business Insurance business to the extent we do business with large accounts, it really falls into 2 areas. One is National Property, where as a result of our reinsurance programs, our actual net exposure is not what one would think of relative to kind of large accounts. And then we do fee-for-service business in our National Accounts business, where we're handling claims, workers' comp claims for a fee. And the nature of the risk -- it's not guaranteed costs. It's loss-sensitive business, has a very different characteristic. But most certainly, when you look at our middle market business, which is larger obviously, that our small commercial business, you do we see a more price competitive environment and as we watch broadly speaking into the large account arena, which, again, we don't participate in a dramatic way, in a meaningful way, it is indeed more competitive at that end as well.

So that's probably the best explanation as to why the National Accounts' premiums would be, I'll say, the weakest amongst those 3, whereas, the Commercial Accounts segment would really be the strongest and amongst those 3.

Yes, but part of it, but be careful to read into that because as you look at the loss response of -- this is Bill Cunningham, that's also a function of a losses under management and largely workers’ compensation. So over the course of the last year as frequency has dropped and the overall losses have dropped, as we look at those renewals prospectively, that has an impact on the premium we charge and the fee that we charge.

I wouldn't leap into the conclusion because it's a complicated answer that our National Accounts business, to the extent it shows up as premiums, a lot of its guaranteed costs that's showing up as premiums. And it's not subject to the -- what's the premium basis. Bill is expressing that as the loss content is projected to go down, so do our premiums. They go right along with it. But by definition, the profit margin on that business may not go down in the same way. So the nature of our National Accounts business is such that it wouldn't make a lot of sense. It's not guaranteed costs and it would be -- it's just the -- you get the incorrect assumptions. So look at our National Accounts business at the changes and perceive that it's a function of the competitive environment. There are carriers that have real expertise in dealing with large accounts. And we're just not one of them. And I'm speaking on a guaranteed cost basis. It's just not what we do.

And our last question is from the line of the Ian Gutterman with Adage Capital.

I'll be brief. I just had one quick question on Commercial Accounts. I'm trying to understand, but it seems there's some can contrasting trends there. I was hoping to get some insights. The retention pressures seem greatest in that line. And I'm guessing maybe part of that was through the rate increases, and at the same time, your new business is probably one of the better lines there. And the premium growth was obviously very strong, up double digits. So can you help me understand why retention will be getting worse at the same time, new business and overall premiums will be getting better?

I -- your statement is factually correct, but the interpretation is a little off, Ian. We -- the fact is retention in the second quarter of 2011 was 84%. From a historical perspective, that's extraordinarily strong. We were coming off levels of 86% and 87% in the second and third quarters of last year, which is candidly one of the factors that led us to conclude that there might be opportunities to gain rate while still improving the overall quality and profitability of the business that we do. So you're right, it's dropped from 87% or 86% in the second quarter of this year to 84%. But that's still at levels that are solid, strong, great, terrific whatever you want -- which one to put on it. And so at the moment while it's dropped the tiniest bit, we don't perceive it at the moment, unless you see it differently than I do, that we've taken a meaningful dip in retention relative to what long-term profits -- relative to the way where we've managed the business to produce returns.

On the growth side of the question, there's a number of things playing through there. First, in addition to having continued strong retention, you've got the impact of exposure being positive on renewals. But you've also got the audit additional premium coming through those numbers as well.

Yes.

And then their rates. So you've got rate gain, exposure gain at renewal, positive audit premiums and those are the factors playing through this that are impacting the growth.

And the positive audit premium compared to negative audit premium last year. So that's going to have an impact on that percentage growth.

Your question is still a thoughtful and substantive one. We will always watch this as we go down this path slowly, thoughtfully. Hopefully, our concept is that we can execute this without trading off substantial amounts of retention. If, in fact, we overshoot and retention drops in a precipitous rate, we'll obviously re-evaluate our view about how we're approaching pricing. But the question's right. I think you just have to put it into a historical perspective to look at it and say, still on track.

There are no further questions at this time. I would now like to turn the call back to you, Ms. Nawi. Please go ahead.

Thank you. Thank you for your -- all, for joining us today. As usual, both myself and Andrew Hersom will be available to take any additional questions. Thank you, and have a great day.

Thank you. Ladies and gentlemen, that does conclude the conference call for today. We thank you all for your participation, and we ask that you please disconnect your lines. Thank you and have a good day.