The Travelers Companies, Inc. (TRV) Q3 2011 Earnings Report, Transcript and Summary

Greg Toczydlowski - President of Personal Insurance Jay S. Fishman - Chairman, Chief Executive Officer and Chairman of Executive Committee Alan D. Schnitzer - Vice Chairman, Chief Legal Officer and Executive Vice President of Financial, Professional & International Insurance Jay S. Benet - Vice Chairman and Chief Financial Officer Gabriella Nawi - Senior Vice President of Investor Relations and Member of Operating Committee William E. Cunningham - Executive Vice President of Business Insurance William H. Heyman - Vice Chairman and Chief Investment Officer Brian W. MacLean - President and Chief Operating Officer

Matthew G. Heimermann - JP Morgan Chase & Co, Research Division Jay Gelb - Barclays Capital, Research Division Michael Nannizzi - Goldman Sachs Group Inc., Research Division Randy Binner - FBR Capital Markets & Co., Research Division Cliff Gallant - Keefe, Bruyette, & Woods, Inc., Research Division Larry Greenberg - Langen McAlenney Joshua D. Shanker - Deutsche Bank AG, Research Division Gregory Locraft - Morgan Stanley, Research Division Brian Meredith - UBS Investment Bank, Research Division Jay A. Cohen - BofA Merrill Lynch, Research Division Keith F. Walsh - Citigroup Inc, Research Division

Good morning, ladies and gentlemen, and welcome to the Third Quarter Results Teleconference for Travelers. [Operator Instructions] As a reminder, this conference is being recorded on Wednesday, October 19, 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, Sussie. Good morning, and welcome to Travelers' discussion of our third quarter 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, and then we will open it 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. And now, Jay Fishman.

Thank you, Gabby. And good morning, everyone, and thank you for joining us today. You, of course, all know that this was another quarter marked by significant weather. In that context, our business is strong enough to have produced $332 million in operating income and an operating return on equity of 5.9%. For the full year, we're at $781 million of operating income and an operating return on equity of 4.5%. While this year has certainly placed considerable demands on our claim department, we're very pleased that our claim teams were prepared and positioned to customers with first-rate speed. I also couldn't be more pleased with our catastrophe claim management strategy upon which we embarked in 2005. Our ability to respond to claimants with almost entirely internal staffs sets us apart from many other insurance companies and allows us to process claims more effectively and efficiently, as well as with a higher level of customer satisfaction. Turning to our operations, our strategy is quite focused. We are seeking improved pricing, as well as improved terms and conditions on the insurance products we sell. We are doing so not only because interest rates may remain low for some time, but also given the possibility that the more active weather patterns, such as we have experienced over the last few years, may continue. And the progress we are making pursuing the strategy is already evident in our Business Insurance segment. In the quarter, we achieved pricing gains across all product lines, led by workers compensation and property. The overall pure renewal rate gain in the quarter for Business Insurance, excluding National Accounts, was just over 3% and just over 4% for the month of September, demonstrating our success in driving rate improvement. And while we are prepared to trade volume for improve margins and profitability, retention remains at strong levels and forms a basis for our optimism that we can continue to improve pricing in our Business Insurance segment. In that regard, we are particularly pleased with our results in our middle market businesses, and Brian will have more to say about that later. We also note that exposure in our Business Insurance segment remains positive and continues to improve. In addition, audit premiums have also turned meaningfully positive. The combination of higher exposure in auto premiums, as well as improved rate are important factors, which allowed us to generate a significant 7% growth in net written premiums in Business Insurance this quarter. While our strategy is the same in Personal Insurance, our renewal premium change for property has been in the 8% to 9% range for some time, and our renewal premium change for auto has been in the 3% to 4% range. And the regulatory environment is such that it will take some time before our more recent actions will show in our results. With respect to capital management, we repurchased $375 million of stock in the quarter, and our guidance with respect to targeted fourth quarter repurchases is considerably higher. Jay Benet will share that with you during his comments. Finally, we're also pleased that Standard & Poor's upgraded our financial strength ratings to AA, although we are not reliant on the commercial paper market to fund our operations, our short-term ratings are now A1/P1. In summary, the weather was again challenging, but we remain bullish on our ability to deliver rate improvements and superior returns over time. With that, let me turn it over to Jay.

Thanks, Jay. There are a few things I'd like to discuss with you this morning, so let me start with the weather. Third quarter catastrophe losses, which were driven by Hurricane Irene and Tropical Storm Lee, with $394 million after-tax, much higher than what we were consider "normal" and more than 5x our cap losses in last year's third quarter. Losses related to Hurricane Irene were $253 million after-tax. It represented only 1% of shareholders' equity. The losses from Tropical Storm Lee was $74 million after tax. The cats were not the only weather story in the quarter. As Brian will discuss later, both Business Insurance and Personal Insurance experienced significantly higher levels of non-cat weather-related losses that, when combined with the expected impact of earned pricing and loss cost trends, drove a 4.5 point quarter-over-quarter increase in our consolidated GAAP combined ratio, ex-cats and prior-year development. Turning to reserves, third quarter's total net favorable prior-year reserve development was $124 million after tax, down a bit from $147 million after tax in the prior-year quarter. Once again, all 3 of our business segments experienced net favorable prior-year reserve development as a result of better than expected loss experience. Within Business Insurance, reserve development included a $114 million after-tax increase to our asbestos reserves. While the underlying asbestos environment is essentially unchanged, recent payment trends have been moderately higher than we had previously anticipated. I'd also point out that this year's asbestos reserve increase was actually less than last year's increase on a gross basis. You will recall that last year, the asbestos reserve increase was net of a reduction in the uncollectible reinsurance reserve of $70 million or $45 million after tax, that resulted from a favorable ruling related to a reinsurance dispute. Page 7 of the webcast summarizes our holdings of sovereign debt securities issued by Eurozone countries that are considered troubled: Spain, Ireland, Italy, Greece and Portugal, amounting to only $8 million as of the end of the quarter. The schedule also shows our holdings of all securities issued by banks, insurance companies, finance and leasing companies and brokers and asset managers in these troubled countries, consisting of debt and nonredeemable preferred stock. These holdings amounted to only $16 million as of the end of the quarter. Lastly, the schedule shows our holdings of securities issued by other Eurozone financial institutions where the securities are not guaranteed by the sovereign country. In total, all of these Eurozone holdings represent only a tiny fraction of our $73.7 billion investment portfolio. While our analysis of the RMS version 11 hurricane model update remains a work in process, we do have some preliminary observations we would like to share with you this morning, all of course, subject to the completion of our works. Based upon our evaluation of RMS 11 and other hurricane models that we use in combination with our own expensive data, we currently believe that we will be increasing our expected hurricane average annual loss, or AAL, as well as our expected hurricane probable maximum loss, or PML. This is one of the reasons that our strategy is focused on pricing and terms and conditions. Notwithstanding these expected increases on our hurricane AAL and PML, we currently expect our total catastrophe exposure adjusted for RMS 11 will remain within our targeted equity impact range as contemplated by our existing capital structure and that, as a result, we are not expecting to increase our target capital levels, significantly reduce our exposures, or purchase additional reinsurance as a result of RMS 11. I'll remind you, this analysis is still a work in progress so these observations are preliminary. Lastly, let me mention a few key numbers. Operating cash flow was very strong this quarter at $917 million, despite the large spike in claim costs due to weather. We ended the quarter with holding company liquidity in U.S. dollars that are in the United States and therefore, not subject to tax on repatriation of $2.4 billion. Net unrealized investment gains increased to $4 billion pre-tax or $2.6 billion after tax, and book value per share rose to $60.98 at the end of the quarter, a 4% increase since the beginning of the year. And consistent with our ongoing capital management strategy, we continue to return excess capital to our shareholders. During the quarter, we repurchased $375 million of common stock and paid $173 million in dividends, bringing year-to-date common stock repurchases to $1.7 billion and year-to-date dividends of $503 million. Based upon our current liquidity position and our expectations for fourth quarter operating income, which as always, does not include any provision for reserve development and subject to other needs that may arise, such as contribution to our qualified pension plan, which we currently do not expect to have to make. We are targeting fourth quarter common stock repurchases at approximately $1 billion. And with that, let me turn the mic over to Brian.

Thanks, Jay. As we just mentioned, our third quarter earnings were meaningfully impacted by the catastrophic weather events listed on Slide 4 of the webcast, as well as an above-average amount of non-catastrophic weather. Excluding the impact of weather and prior year development, we continue to see margins that were essentially in line with our expectations. The other important part of our story this quarter was the strong increases in pricing that we continued to achieve broadly across our businesses. This is especially true in Business Insurance, where net written premiums were up 7% year-over-year, driven by double-digit increases in our Commercial Accounts and industry-focused businesses. The overall increase in net written premiums included renewal premium change of 5%, of which pure rate accounted for over 3%. This marks the third consecutive quarter of positive rate increases in Business Insurance and in fact, these are the largest rate gain since the Travelers-St. Paul merger in 2004. In addition, we are achieving the rate gains broadly with increases in all product lines for the second consecutive quarter. The workers compensation line had the highest level of rate increase, and the property lines showed the most significant improvement over last quarter. As you can see on Slide 14, the pace of rate increases, both in Commercial Accounts and in the segment as a whole, continued to accelerate as we ended the third quarter, with segment rate at 4% and Commercial Accounts at 6% for the month of September. Importantly, we believe that these written rate levels exceed our current view of lost trend. I know this is only one month, but after speaking to you about margin compression for the last several years, we believe this is significant. And based on our preliminary view, we believe we are making further progress in October. In addition to the rate gains, our top line benefited from our customers purchasing more insurance. Exposure was up about 2% this quarter, and audit premiums were up meaningfully, both year-over-year and in sequential quarters. Retention remains solid, while new business is down slightly from the same period last year. Along with our renewal pricing strategy, we are also achieving better pricing on our new business. And although this may have an impact on new business volume going forward, we are comfortable with that trade-off. Turning to third quarter operating earnings. In addition to the significantly higher catastrophe losses and less favorable prior-year development versus the third quarter of 2010, results have been negatively impacted by some non-recurring items, primarily the non-catastrophe weather largely in the property lines. Last quarter, we provided you with an update on our workers compensation loss trends, specifically noting a modest uptick in frequency. Through this quarter, overall loss trend has remained stable, and frequency is performing as we expected. As always, we continue to manage this line closely on a state-by-state basis. So we continue to feel good about the underlying performance of this business and are extremely pleased with the rate gains that we've achieved and in particular, the pricing momentum that occurred in the latter part of the quarter. In the Financial, Professional & International Insurance segment, operating income of $211 million was strong and consistent with strong operating income in the prior-year quarter. The current quarter's underwriting gain was up modestly due to favorable prior-year development. The 3.2 points of deterioration in the adjusted combined ratio is due to the impact of lower levels of earned premium and unusually low level of what we define as large losses in the prior-year quarter and an increase in infrastructure investment to support growth in the International business. For the first time since the first quarter of 2009, our management liability business and Bond & Financial Products experienced growth on a production basis. And excluding the impact of a personal lines distribution arrangement in Ireland that we terminated last year, the International business also grew on a production basis for the first time since the first quarter of 2010. In both cases, this is the continuation of a positive trend going back several quarters, and we believe is the result of IT investments and other strategic efforts. Surety volume continues to be down, reflecting sluggishness in construction spending and underwriting selection. Moving to Personal Insurance, extraordinary weather events, both catastrophic and non-catastrophic, continue to adversely impact our financials and drove an operating loss for the quarter. Net favorable prior-year development for the segment was somewhat less than the third quarter of 2010. Excluding cats and prior-year development, the Personal Insurance combined ratio increased approximately 3 points quarter-over-quarter in both Homeowners, as well as Auto. The majority of this increase was driven by adverse non-cat weather. Certainly, we're not going to try and predict the weather, but by incorporating the last 2 years of actual experience into our long-term averages, our expectation of weather losses in this business has increased. In response, we will be driving for additional rate increases, as well as adjusting terms and conditions and looking for higher deductibles. Turning to the production results, both lines of business posted year-over-year growth in policies in force this quarter, while net written premiums were up 3% in Agency Homeowner and down slightly for Agency Auto. Renewal premium change and retention were strong in both Auto and Home, while new business volume was down from the same period last year. We expect new business volumes may decline modestly going forward, as we continue a disciplined execution on our pricing strategy. As in Business Insurance, this is a trade-off that we are comfortable making. Now before we go to Q&A, I'll turn it back to Jay Fishman for a comment.

Thank you, Brian. I just want to take 15 seconds to acknowledge an important anniversary, our much appreciated and much loved partner, Bill Heyman, is celebrating his 20th anniversary with the company. And consistent with Travelers' expense-minded culture, I'm handing him his inexpensive and cheap Lucite plaque. No watches here, but I can't think of a better place to acknowledge 20 years, Bill, than in front of the people for whom you've been working all this time and doing such a terrific job. Congratulations.

I have no idea. Thank you.

With that, operator, we're going to open it up for questions.

[Operator Instructions] Our first question, coming from the line of Mike Nannizzi with Goldman Sachs.

Just a question on workers comp. Obviously, you saw some -- a lot of premium growth in the quarter that's been a trend this year. Can you talk a little bit about pricing versus loss trend there? And kind of what is the geographic breakdown of that business and any problem states, or states where you're seeing loss trend increase in what you're doing there?

Mike, this is Brian. I'll take a shot at some high level comments, and then throw it to Bill Cunningham and I'll probably preface all of this with were not going to get to granularity into talking about a specific state or market. As we said in the comments, we feel comp pricing has been moving. It started moving before the other lines of business. And we continue to see some decent increases there. So we feel good about where we are pricing versus loss trend broadly in that business. We spent a good bit of time on last quarter's call talking about what was in the end of the day, a pretty little movement in frequency, but we wanted to break it out. So the loss trend continues to move as we discussed then.

Which importantly, just to interrupt, if you remember the conversation last quarter, it was that we were moving back to the previously anticipated trend line just a little bit faster than we had expected. So that trend line remains intact, So there's no subsequent issue or problem or change with respect to the workers comp loss relative to our conversation last quarter.

So important, when you look at just premium change, though, and this is true of comp and all of the commercial businesses, what's really driving it is 3 things: The rate component, and really, in this order: Rate, audit premium and exposure growth. And so it's not like we're adding a ton of market share here. We're looking at moving pricing and getting better pricing for the risks we're taking. The other thing I'll say and then throw it to Bill is not only is it a state-by-state, but it's business-by-business. So we have a kind of a different view of small comp versus middle market comp versus large comp. We're always trying to balance all of those.

And just a few things to add. First, and I don't know if mentioned payrolls, payrolls are up.

Yes, which is what's driving the exposure in the Auto premium.

Right, and I guess, again, not getting into specific geographies, for competitive reasons, what I will tell you is that there are wide ranges of rate change that we're gaining, and it depends on our view of a given state. So the overall rate gain for the line continues to move in more dramatically in the states where we need it. And we see a loss trend dynamic. As we said last quarter, we reflected in our pricing, in our reserving that uptick. And as we look at loss trend and rate gain, we talk about being in front of loss trend projections comp based on our pricing today. And I think we were there sooner on that line and the gap between rate and lost from that line is actually wider than the all lines average.

And just to clarify my comments to make sure we don't confuse anybody. In the second quarter, we increased -- lost pick for workers comp overall to reflect that acceleration into the trend line. That loss estimate is the same one we used for the third quarter, mix adjusted and everything else. But we didn't go back. It currently is at the level that we reestablished it in the second quarter.

Got it. And then just on the Travelers Select, I mean, obviously, commercial line is up a lot. I mean, is that a reflection of pushing your systems into larger policy sizes, whether it's Select or Select Express? I was just curious to see a big lift in commercial line, and I imagine a lot of that is comp. But the Select was flat. Is that a result of maybe to Brian's comment that pricing is up higher maybe than others?

I apologize, Mike. I'm not getting the question of that was that I know of.

Well, it's just that Select as a platform and new business was flat, or premiums were flat whereas commercial lines were higher.

Yes, okay. The basic dynamic that's going on there as we've talked about in Select, Select Express is the smaller end. What we call Plus is the larger end. The Plus market, the larger account market is in aggregate down for us and one that we're pushing harder in from a pricing perspective. So that's what would be fundamentally driving that.

The returns, if one breaks Select in 2, it's 2 pieces, Select Express, the lower end Select Plus, the upper end. The returns in the Select Plus business have not been at levels that we feel comfortable just sitting on. And that's partially because we're competing in that segment with smaller, more regional companies who view the upper end of Select as their middle market. So it tends to be a segment of the business that attracts a fair amount of competition. The pricing dynamics there had been a little more difficult, relative to a lower end, more difficult. And so we set our standards and whatever happens to the volume happens. And if we shrink, we shrink; and if we're flat, we're flat, it's okay, as long as the business that we're pricing is being priced at a level that we're prepared to take on and live with. Is that fair?

Absolutely.

Our next question, coming from the line of Jay Gelb with Barclays Capital.

My first question is more broad. It has to do with the P&C cycle. Jay Fishman, are you calling for a broad cycle turn here? I mean, you're talking about price increases in Business Insurance in particular, now in excess of loss cost and positive territory for 3 quarters in a row. Is that what you're calling for here?

No, sir. In fact, if you go through the comments here today, you won't actually hear the words market or market cycle in anybody's commentary. We decided a year ago -- as I've said, and you've heard us speak about this in other meetings and in our investor days. We're much more skeptical that the amplitude of the cyclicality that many of us grew up in, in this business remains. For a whole host of substances systemic reasons, we just don't think it exists in the same way anymore. And we made the decision a year ago, if you recall, that regardless of whether a red light went on somewhere that was going to declare a cycle turn that we were going to attempt to take our destiny in our own hands and move our own pricing. And whatever happened would happen. We were very encouraged as following the second quarter last year, the results were moving in very much the right direction. We were never nearly as negative as the surveys would've suggested. We shared that data with you. But we made progress. And went from, broadly speaking, negative to positive. Following the weather this year, we decided after the second quarter to further accelerate our own efforts. And so I'm not viewing what anybody else is doing. I'm not declaring any change or turn, absolutely not. I'm just telling you what we're doing. This is active, not passive. We're not reacting to the marketplace. We're just deciding for ourselves what we're going to do in the way we're going to price the product. And we'll see what happens.

That makes sense. And then for Jay Benet, 2 other ones. First, looking in the 10-Q in terms of the pace of the buyback for 2012, it says Travelers does not expect that we'll continue to repurchase shares in excess of its operating income in 2012. So does that mean it could potentially be less than the operating income that's generated next year? And second, why the shift, especially given $1 billion share buyback pace potentially in 4Q?

As you've heard us say before, we've been generating excess capital, we've been returning the capital to our shareholders. We had built up excess capital. We've been operating, as I mentioned during the Barclays conference, in ways to free up capital on the balance sheet, either through reinsurance collections or changes in reserves that have taken place over time or looking at other things. So I mean, eventually, you get to the point where you return the excess capital to your shareholders, and then what you're doing through your operating earnings is creating more capital, evaluating how much of that capital that you're creating through earnings you need to run the business and then from that returning the excess, if you will, to your shareholders, either through dividends or through buyback. So we think we're getting close to that inflection point where the excess is actually being or will have been returned, and we'll be dealing with the creation of new excess through the earnings.

Okay. And then, Jay, last one. The impact of DAC accounting changes in 2012, any potential write-down for Travelers?

Actually, I think if you look at the new rules closely, you'll see that the new rules really paralleled what we've been doing, so there's not going to be any impact from those new rules for us.

So big impact for the life companies but not for PNC or Travelers.

Certainly not for us. I mean, our DAC policy has been to DAC commissions and premium taxes essentially and nothing else. So that coincides with the rules.

Our next question, coming from the line of Josh Shanker with Deutsche Bank.

In responding to Jay's question, you said that by the year ago, you decided to take destiny into your own hands and you've seeing favorable results. You're talking about 3 consecutive quarters here of improving renewal rate trend. But also over that same period of time, I'm seeing on running margins deteriorate and it's accelerated. Excluding cats and reserve development, this insurance ratios were up 4 quarters go 100 basis points, 200 basis points, 300 basis points, and then this quarter, it's up year-over-year 700 basis points. I'm seeing a disconnect there. I realize one is earned premium and one is written premium. But it's been going on for a little while here. Can you sort of walk me through what's going on here?

Sure. Well, first, your first point is exactly correct. There's an enormous difference between earned and written. And the rate gains reflect themselves in written premiums and obviously will be earned, respectively. The comment in the press release that we issued this morning was important relative to in Brian's comments as well. The fact that we were now at a point where if, in fact, these written rate gains continue, that they exceed our current view of loss trend. So that's the first, written versus earned.

Well, let me just say, a year ago, you said rates were flat and now we have 700 basis points of deterioration from stuff you wrote a year ago. Is that a wrong way to...

The numbers are public. My recollection is that overall rate in the second quarter of last year was still actually modestly negative. Loss trend obviously was positive, and we shared with all of you, you included, that we were anticipating margin compression. We were very clear, very transparent at the end of last year and into this year that we were anticipating, projecting and shared with you, in fact, even we were giving guidance the magnitude of the margin compression that we were expecting. And that was driven by the fact that underlying loss trend was running at a higher rate than the -- and at that point, the negative rate that we were achieving. It wasn't substantially negative, but negative is nonetheless negative. And we made progress in moving that price, moving the rate up, but it wasn't -- it isn't until just recently, it wasn't until just recently that the numbers are such that the written rate gains now are projected to exceed current view of loss trend. We've never said that before. You can go back to all the transcripts and you can check. When asked, the closest I think we ever got was that we were saying it was getting close to a push. That was about as close as we got to talking about changing margins. And the last piece, and particularly in this quarter, the most important is non-cat weather. It plays a substantive role in our quarter-to-quarter margins and in our estimates of combined ratios. What I can share with you is this. When you look at the underlying combined ratio now for every business segment that we are, and I'm saying this exclusive of whatever catastrophe load each of us would think is appropriate, but every single one of those segments for the 9 months, for the year-to-date, has a run rate combined ratio of under 100%, some a little more, then some others a little less. But I think that what we've been sharing with everyone about what we were expecting in rate gains and loss trend is the most transparent in the business. And I don't quite see the disconnect. I think we've been very clear about where rate gains and loss trends have been moving.

Can you give a little detail on you said non-cat weather. How much were the percent of the combined ratio do you think non-cat weather has been running over the past few quarters compared to where it used to be?

Yes, Josh, so this is Brian. I'd stepped from -- because you can pick a part that combined ratios by quarter and look at a whole bunch of different things. But if you look at our year-to-date in Business Insurance combined ratio and the ex-cat, ex-PYD number of 98.2, I would say knock about a point off of that and you've got what we view as our run rate combined. Round number, which is roughly a 2-point deterioration, as you can see on the page. It's 2.9, so knock a point off of that, it's about a 2-point deterioration in the margin. That's -- anytime you look at the quarter, you've got weather spikes, you've got weather compared to what weather was last year, if a quarter was made -- an adjustment was made in this quarter or an adjustment was made in prior-year quarter and then we put so 3 quarters worth of the impact in one spot, the numbers get wacky.

And in a given quarter, there'll be an episodic large loss that's not really run rate driven. Our best analysis, as Brian says, is looking at the reported ex-cat of 98.2 and we see about a point in there that's driven by what we would certainly consider non-run rate items.

Right. And back to about 2 points of margin compression for the year.

Our next question, coming from the line of Keith Walsh with Citi.

First question, you mentioned on previous calls that the delta between new business pricing and renewals are shrinking. I just wanted to know, is that still the case? If can update us on that. And the reason I really asked this is broker Brown & Brown yesterday made some comments that would maybe contrast with that statement a little bit, and then I've got a follow-up.

No, our experience is that it is continuing to shrink a little bit.

And we would, as a matter of -- I don't have a number to give you on this. But as we've been speaking with our field people, a very relevant question is okay. We get what you're doing in the renewal rate arena, but what are you doing in the new business arena? What we're instructing our field people to do is to bring that higher level of expected rate to their new business negotiations. That's very, very important. It's the instructions that they're getting or to be as return focused in new. It doesn't mean the returns are equivalent but to be as a return focused in new as they are in renewal. And to the extent that new business falls off, that's just fine. No issue there. Do you want to add any? No? And anything in Personal or FPII that you want to add or...

I don't think it's much of the dynamic in [indiscernible]

Yes, because you're all the same. You have the same pricing. It's a Business Insurance phenomenon.

What we know factually, Keith, is our experience, the target that we're putting out and what we believe the prices that we're writing in at is the GAAP is narrowing some for us. Whether it's unique, we don't know.

Okay, great. And then for Jay, you guys have done a nice job achieving and articulating the progress on rate the last several quarters. But just more big picture. How do you link a needed rate increases with the current yield realities? And then more specifically, is the rate you're getting today commensurate with the drop in net investment income you're experiencing now and going forward?

Those are great questions. They really are. And something that we spend a tremendous amount of time here analyzing and looking at. So first is, as I've shared with all of you before, we do have a quarterly analysis that looks at renewal, return on allocated capital, as well as new business return on allocated capital for every product that we sell in every business. And it's a policy view. What I mean by it, it's literally the term we use internally. It's an analysis which attempts to say given everything we know today, rate today, loss trend today, what is the return on that product. About -- I got to check specifically, but maybe 9 months ago, we determined to update that analysis for the new yield curve in such a way that the analysis now calculates these returns as if all the new money is going in against the yield curve as it's currently structured. Now the last time we -- meaning, all the premium money. The capital bears a percentage of the return embedded in the portfolio but the new money goes in against the old curve, and we just count it back. We have not yet updated it for what I would say are the events over the last 6 weeks because the yield curve is most certainly flattening again. And I don't know how long it will stay but once we become convinced that that's longer rather than shorter, we will update it for that. The returns are not -- I'm not going to share with you what they are, but when we know them. We know them and we understand the impact both of the declining investment environment, as well as the impact in our property lines of weather. And obviously, the long tail lines are more impacted by the investment environment. The shorter tailed lines are more impacted by the weather. But no matter how you look at it, they were all impacted. That gets us to driving for rate. Because while I believe, and I do believe this, I believe that our accident year return on equity overall in a normal weather year, whatever that means, is still in excess of our cost of equity. And I believe that it is. We still need to begin to drive rate because the portfolio will continue to mature and roll over into the lower yields. And we've got to make sure that we always have in mind delivering superior returns relative to our cost of equity. So we're on it. And that's why the urgency here for, as a matter of tactics, I would say, it is now all about rate. It is all about rate.

Our next question, coming from the line of Gregory Locraft with Morgan Stanley.

Wanted to just follow up on a couple of items that obviously we've been discussing. One is just on the potential for margin inflection. I just want to be clear. Has your margin already inflected? I mean, effectively, what you're saying is in September and October, pricing is outstripping loss? And so what I'm trying to assess is margins on an accident year ex-cat, ex-reserve basis, core margins have been deteriorating for several years now. Should we begin to inflect them into next year in our models?

So the first important thing and we try to say it, but let's emphasize it. We're talking about written rate. So the 4.3 points of rate that we're seeing in September in Business Insurance, we believe is ahead of our current view of loss trend. That needs to earn into the income statement over time. And so to the extent that we can string a reasonable number of months together like that, it will begin to earn in. But that earns in the way premium earns in. So it's 1/12 at a time. And so we are not saying that in September, the month in our income statement, we saw margin expansion. Just to be crystal clear.

Yes, and I'm more thinking about next year. So are there any businesses right now where margins are -- where pricing is below loss trend?

Right now, in all of our segments...

On an earned basis, let's make sure we're answering the right question.

Okay, yes. On a written basis, we feel that in all of our segments, we are ahead of loss trend. Maybe modestly. Yes, assuming a normalized weather, if any of us know what normal weather is anymore. But...

But the realization of that, 2 things have to happen. We have to continue to deliver at least that level of rate and also, we have to keep monitoring loss trend to make sure that doesn't change because of inflation or any other element. But that -- not even back of the envelop which is said to me, when will this begin to evident itself, and it doesn't mean that in the aggregate, margins will at your face expand but, Brian, given the timing, by the time we get to the second quarter of next year, if this pattern continues at that point, and we should test that because I'm doing this really on the fly. But by the time we get to the second quarter, I would think that, in fact, the combined ratio would begin to shrink all other things being the same.

Okay, that's very -- Brian, were you going to comment? Or that's very helpful. Okay, Great. And then actually just a question, again, you guys are very thoughtful on ROE and capital allocation. It's tied into your management compensation. So how are you -- it's very, very difficult unless we have a substantial margin improvement next year to get to an ROE into the double digits, well into the double digits. So how do you set goals for next year, given where yields are at, sort of building off to Keith's question earlier? But the combined ratio is only one lever. The other lever in investments is sort of is what it is, and it's out there.

No, that's -- it's, again, another thoughtful question, Greg. Two things. One is that from a GAAP perspective, meaning the earnings per share that we will report, the investment yield continues to take time. It rolls -- the portfolio rolls obviously, but it doesn't happen all at once. So from a pure GAAP reported perspective, it's not as bad as if one did the analysis, assuming the entire portfolio were currently invested at current yields. So we have 2 separate analysis to do. What drives our pricing is pure economics. What current rates are available to invest new premium dollars that we bring in. We don't kid ourselves one bit. We understand and fundamentally discount back at today's yields in the aggregate. And we haven't done our budget yet for next year, and it's not as if management -- we don't think that we have levers to drive things differently. And we will do it, but we will be benefited in that regard by the fact that the portfolio doesn't all turnover at one instant. And so it's always important to keep those 2 different tracks in mind.

Our next question, coming from the line of Jay Cohen with Bank of America Merrill Lynch.

Just a couple of questions. The first is, there was another company that in their pre-announcement suggested some loss pressure in the surety business because of economic conditions. And I was wondering if you're seeing any signs of that? And then, secondly, the favorable development, particularly in the Financial, Professional & International segment, was really quite strong. And I'm wondering if you can give us more color what lines of business, what accidents this year that's coming from -- what's driving that very strong favorable development?

Sure. It's Alan Schnitzer. On the surety side, we've certainly seen that the same commentary you've seen on others. We're just not seeing that in our portfolio, and I would attribute that to careful selection, and we've been talking about that for a number of quarters and that would explain in part why our surety premiums have been under some pressure. But that business is really all about discipline underwriting and risk selection, and we really focused intensely on credit quality. And we measure the credit quality of our book, and we've got our own proprietary credit scoring. And while I wouldn't say it's an all-time high, when you look at it on a historical basis, it's still very, very favorable and it's where we would've expected to be, thinking back to what our expectations were several years ago. So there's been some movement in that credit quality, but entirely within expectations, and we're very comfortable with the risks in our surety portfolio.

Just to come back as you went through it quickly, you said premiums were under pressure, that's from a perspective of underwriting thoughtfulness, a voluntary.

Risk selection and essentially seeding to the market risks that we weren't prepared to write. Your second question related to prior development and it's hard to point to one thing. It expands a number of action at years, expands really all the geographies and there isn't one particular item or trend I'd point out as necessarily driving that. It really is broadly base.

Our next question, coming from the line of Brian Meredith with UBS.

A couple here. First, Jay, historically, you've given us what the impact of the lower investment yield environment would potentially be on investment income going forward. Could you give us any update on where we stand right now given the current investment yield environment?

Well, yes, I mean, in terms of the basic analysis, it really hasn't changed fundamentally. What we do -- what we've done historically is, we've provided an estimate of how much of the fixed income portfolio would "mature" -- and I'll say why, quote, unquote in a second -- over the next 3 years. We don't just look at the maturities as a function of what the documents say. We take a look at, given the interest-rate environment, what's likely to prepay as well. So I think in the past, looking at 2012, we had said something to the effect of about $6 billion or so would roll over in the portfolio, given where rates are, that number will be a little bit higher, possibly $6.5 billion, $7 billion. Although I will say that my own, and this is my view, my question, not even a view, I mean, given what drives that meaning mortgage prepayments, in this environment, it's really not really clear how they're actually going to perform. So it could actually be closer to the $6 billion. But in any event, it's that kind of a proportion of the portfolio, and then the rest is we've given the rates as to what is embedded in the portfolio. I mean, generally speaking, on a tax equivalent basis, it's been somewhere in the 4%, 4.5% range of what's coming due. So it's anybody's guess what rate to use for a reinvestment assumption, I think when we started doing this, it was 75 basis points, then it was 150 basis points. I think if you look at it today, it would be closer to the 150, maybe a little more, so I mean, pick that. But that's basically the nature of the analysis.

Also, a lot of fix income securities have periods of call protection. Traditionally, about 10 years in much the muni portfolio. So we have to incorporate that. And that's not really apparent to someone looking at it from the outside unless you wade through it security by security.

Got you. And are you doing anything with your investment portfolio, try to limit the degradation you're going to see in the book yields?

We have always operated on the principle that we'll try to get the best return the market gives us, but not more than the best return the market gives us. Actually, right now, we're emphasizing credit quality. Because we think, no matter which way things break, it's going to be even more important than it was 2 or 3 years ago to have the highest credit quality you can. So we're not reaching for yield. We're taking much what the market will give us.

Brian, just to build his comment further in the portfolio, I don't recall it being structured any shorter from the duration standpoint than it is now.

It has gotten shorter simply by mathematical operation. Obviously, as you move along the price yield curve, the slope of the tangent changes, and the duration changes even if the portfolio is unchanged. But our duration is about as short as we would want it now.

And that actually, given the flat yield curve, is really a place you sort of want to be at the moment. And to the extent that we're wrong, and we always consider the possibility that we're wrong and rates rise, we're going to be happy we positioned the portfolio that way.

And if they don't rise, we don't think we're much prejudiced.

Our next question, coming from the line of Matthew Heimermann with JPMorgan. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: Just a quick technical follow-up on the investment portfolio. It's just -- the private equity securities, those get marked on a quarterly basis, correct?

They are -- we mark some according to the audited statements from the funds themselves, although obviously, much of the NII from private equity is the result the realizations, when private equity funds sell companies in their portfolio. And there's a slight lag, but it's not great. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: Okay. All right. Just most of those private equities -- the audited statements, though, are about quarterly lag, right?

Yes. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: Okay. The other that which is -- that's fine. I just wanted to make sure I had that technique correct. And then I guess either for Brian or Jay. You've talked about what you're doing on pricing as being very much a Traveler specific effort. But are you seeing any signs in the field that your efforts are maybe emboldening some of your competition to potentially take the same tack?

We've been so focused on our own business and our own strategy and our own steps. I mean, I've been out meeting with agents. Brian has been out meeting with agents. We're just -- this is -- it's absorbing all of our attention. So the answer to that is I don't know. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: And then just on the agency side, on the distribution side, with respect to new business, are you seeing any change kind of in kind of where new business is coming from at this point? And the only reason I asked is just that there has been continued talk by some of the brokers and some of the private distributors that there's increasing churns so just was curious whether or not you're seeing any difference, not really different sources bringing you new business or any differences between kind of distribution coming on renewal, who's bring you renewal versus new?

Yes, this is Bill. The short answer is no. We're not seeing a change.

It will be interesting to watch. Obviously, churn is the inverse of retention ratios. And our retention ratios remain pretty high. I would guess listening to a few the other companies over the next few days into what their retention ratios have been will give you real insight into whether that's true or not. From our experience, the answer would be no. But that doesn't mean it's universal.

Our next question, coming from the line of Larry Greenberg with Langen McAlenney.

Just on going back to non-cat weather. So I guess I presume that year-to-date, the entire 4.3 point underwriting deterioration accident year ex-cat for personal lines, that would all be attributed or primarily attributed to non-cat weather?

Hey, Larry, this is Greg Toczydlowski. Predominantly, that's absolutely true. Close to 80% of it is non-cat weather, and we saw that on both the automobile lines and the property lines.

Okay. And then I think, when you previously had been given -- giving annual earnings guidance, your suggested cat load at that time was I think around $600 million. And from Jay's comments, anybody's guess on normal weather these days, I know it's probably a hard thing to peg down. But is there any guidance you can give us on what a normal cat load is today?

Well, obviously, when we finally get to our budget for next year, we'll have a number such as it is. The only hesitation is that we're still sort of I'd say in the midst of RMS 11 and what that does to our cat loads. Jay said that it's going to increase some and it will as a result of that program, not so much just to put us out of, in effect, out of policy or out of line with our capital committed to it, but it is going to change. So I just think it's early, Larry. And we'll give some thought as to whether after that work whether we feel comfortable providing some more insight. I understand it's an important factor for everyone to understand, and we obviously have made a disclosure of it in our -- not of the average annual loss, but of our tail exposure in our 10-K. So when we get there, maybe we'll have a basis for saying something about what our average annual loss estimates are.

Our next question, coming from the line of Cliff Gallant with KBY. Cliff Gallant - Keefe, Bruyette, & Woods, Inc., Research Division: Most of my questions are answered. I guess, I could ask on the property lines just to clarify the point. You talk about rate increases being pretty strong in the high single-digit range. But the RMS model changed again, changing your weather expectations. So how should we think about the expected return on the business? And in net, is expected return still rising?

Well, first, let me -- because I'm not sure I'm connecting a part of your question. Let's make it clear. RMS 11 is, in effect, a hurricane model. What it's doing is it's taking our estimates of windstorm, hurricane windstorm and changing it, which is important in lots of property risks, but not every property risk. We have lots of property risks that are not hurricane exposed. So RMS will have an impact, obviously, on our -- I think in the end what it will have an impact on is our pricing view of hurricane exposed risks. Not everywhere because we're going to get down to the point of identifying where the changes are into very specific geographies. This is not where we get a one-number output, and we spread it for everybody. We're going to use that model and help us drive thoughtful pricing decisions in specific geographic areas. But it's a windstorm focused model. In terms of pricing on property, generally, I don't know what we said. We didn't say a number. We didn't say high single-digits. We said that...

Our largest quarter-over-quarter improvement but we didn't give you the numbers.

And what's really driving that -- and it's important because what's driving that is not yet, I would say, RMS 11. What's really driving that is the experience, predominantly windstorm and hail, tornado and hail, that we've had the last several years. If you go back to -- I don't remember if it was last quarter or the quarter before, we can either make the decision that we're really smart and we've been unlucky or we can make the decision that something different is happening, whether we fully understand it or not, and we should react to it. We've chosen the latter. And as a consequence, if you look in the areas that are meaningfully tornado exposed, we used to talk about tornado alley, the data suggests that's not so relevant anymore. That the people remind me all the time, we have one in Springfield, Massachusetts this year sucking the water right out of the Connecticut River. So as you sit back, as we sit back and just think about weather in fact we're in some cycle, I'm not making any global warming statement here, but maybe we're in a cycle of more active storms, we've got to react, and so our reacting here. And all of -- many analysts tend to think all the time of Homeowners. They immediately go there. We have a substantial property exposure in our commercial book, in our small commercial business and in our middle market businesses. So the dynamic of property extends way beyond Homeowners, which actually, that's the reason I gave the data. We've been achieving 8 to 9 points of what we call their renewal price change for the last several periods.

Right. So there, we have been getting high single-digits for a couple of years now.

That's right. And that maybe where you're getting it from. That was a comment about renewal price, renewal rate change, or price change in Homeowners. But there is a full dynamic of property exposure in our commercial business where, obviously, for competitive reasons, we're not being quite as transparent about what our particular strategies are other than to say it was the largest increase, right, it was the largest improvement.

Our last question, coming from the line of Randy Binner with FBR. Randy Binner - FBR Capital Markets & Co., Research Division: Just trying to get a sense of how systematic the asbestos reserves charges might be. And in the press release, you mentioned the current litigation environment was driving higher settlement costs. Is it plaintive tactics, is it jury wars, is it the bad economy in certain states? Just trying to get a feel for what's continuing to drive the adverse development there and what you might be able to do to try and mitigate that?

I'll try to answer the question and maybe Dorian [ph] or Alan will try them as well. But we make estimates all the time as to what our future costs regarding today. And in this particular case, we had made an estimate that would suggest that the costs would go down slightly. And instead of going down slightly, they went up slightly. We are in a situation where -- and we think is a good thing that the miso cases are the ones that are getting tried. It’s not like many years ago where they were whole package pre-bankruptcies, and whole different set of circumstances. But this is slugging it out on court on a case-by-case basis. And depending upon what the situation is, your defense costs are going to be either smaller or larger than what your expectations are. And they just have been running a little bit larger than what we had previously expected, when we set the reserve. Having said that, we don't view it as any fundamental change in the way the asbestos environment is operating.

Right. And the only thing I'd add to that is there are several jurisdictions that you see an increase of the attack on peripheral defendants, which really don't have a lot of liability. But those meso cases are fairly serious. And so we're assisting in short defend against them. And so that's what's really driving up the defense cost.

For the record [indiscernible] Randy Binner - FBR Capital Markets & Co., Research Division: Now that's helpful. Just to clarify then, so if it's your point, now it is a most mesothelioma cases and that kind of the broad asbestos kind of thing then, I mean, how do we think about this annually? It should be a Decane [ph] liability, yet you're continuing to have issues. So is this kind of the last gasp of this as these mesos move through the system? Or should we think about this being something we have to deal with every third quarter?

Well, we do want to make a point that if you look at the gross reserve increase last year versus the gross reserve increase this year, it was less this year than last year. And as it relates to the population, one would have to believe that there's some point in time where this drops off, now just given the passage of time and the exposure periods, as well as when the asbestos exclusions kicked in, in terms of the way the policies were written. Having said that, whatever this decay rate is, it will be what it will be, and we're trying to estimate it each year. And that's what these adjustments relate to, what's the actual decay rate versus what's been anticipate in the reserve setting.

Ms. Nawi, I will now turn the call back to you. Please continue with the presentation or closing remarks.

Very good. Thank you for joining us today. As always, both myself and Andrew Hersom are available for any follow-up questions. Thanks, and have a great day.

Ladies and gentlemen, that does conclude the conference call for today. We thank you for your participation and ask that you please disconnect your lines. Have a great day.