Gabriella Nawi - Senior Vice President of Investor Relations and Member of Operating Committee Jay S. Fishman - Chairman, Chief Executive Officer, Chairman of Executive Committee and Member of Operating Committee Jay S. Benet - Vice Chairman, Chief Financial Officer and Member of Operating Committee Brian W. MacLean - President, Chief Operating Officer and Member of Operating Committee Alan D. Schnitzer - Vice Chairman, Chief Legal Officer, Executive Vice President of Financial, Professional & Inte and Member of Operating Committee

Randy Binner - FBR Capital Markets & Co., Research Division Michael Zaremski - Crédit Suisse AG, Research Division Vinay Misquith - Evercore Partners Inc., Research Division Josh Stirling - Sanford C. Bernstein & Co., LLC., Research Division Michael Nannizzi - Goldman Sachs Group Inc., Research Division Jay Gelb - Barclays Capital, Research Division Matthew G. Heimermann - JP Morgan Chase & Co, Research Division Gregory Locraft - Morgan Stanley, Research Division

Good morning, ladies and gentlemen. Welcome to the Third Quarter Results Teleconference for Travelers. [Operator Instructions] As a reminder, this conference is being recorded on Thursday, October 18, 2012. At this time, I would like to turn the conference over to Ms. Gabriella Nawi, Senior Vice President of Investor Relations. Ms. Nawi, you may begin.

Thank you. Good morning, and welcome to Travelers' discussion of our third quarter 2012 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 Investors 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 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. And now, Jay Fishman.

Thank you, Gabby. Good morning, everyone, and thank you for joining us today as we report a very strong third quarter. Net income was $864 million or $2.21 per diluted share, producing a 13.6% return on equity. Operating income was $867 million or $2.22 per diluted share, a record for the quarter for operating income per share. Operating return on equity for the quarter was 15.5%. Our earnings were driven by superior underwriting results, as well as solid investment income. From an underwriting perspective, we benefited from lower weather-related losses, net favorable prior year reserve development, as well as the cumulative effect of rate gains we have achieved over nearly 2 years. In Business Insurance, excluding National Accounts, written rate increased to nearly 8% with retention increasing as well to 81%. In Personal Insurance, renewal prices increased in both Auto and Homeowners to 9% and 12%, respectively, while retentions remained stable. Also in Personal Insurance, we continue to execute on improving terms and conditions, along with pricing, and while we are experiencing lower new business volumes, we are taking the right actions in order to improve our profitability and returns. In FP&II, rate increases continued to improve in our Management Liability businesses, reaching 6% in the quarter while maintaining stable retentions. We continue to be pleased with the execution in each of our segments. While we are encouraged by the success we are having executing our granular pricing strategy, I'll remind you that it is not just about the headline numbers, that is what is happening underneath the numbers is the real story. Our success is premised on really understanding which accounts or lines are in need of improvement and which ones meet or exceed our expectations. Brian will take you through Slide 13 in more detail, but the conclusions…

Let me begin by echoing Jay's comment, the third quarter operating income of $867 million and operating ROE of 15.5% were very strong, and along with the timing of certain payables, drove extremely strong operating cash flows of $1.5 billion for the quarter. As Jay indicated, third quarter operating results benefited from a relatively low level of both cat and non-cat weather-related losses when compared to both the prior year quarter and to what we would consider normal for the third quarter, as well as a significant amount of net favorable prior year reserve development. Operating results also continued to benefit from earned rate increases that exceeded loss cost trends in 2 of our segments, Business Insurance and Financial, Professional & International insurance, as well as a relatively low level of large loss activity. In addition, our investment portfolio had another quarter of solid performance, driven by real estate partnership and hedge fund returns, and the non-fixed income portion and fixed income investment returns that are in line with our expectations. Another item I would like to mention is that our net favorable prior year reserve development included an addition of $114 million after tax on a net basis to our asbestos reserves. Coincidentally, the same amount added to our asbestos reserves on a net basis in the prior year quarter. On a gross basis, the current quarter's asbestos reserve addition was actually $24 million less than in the prior year quarter. Although the underlying asbestos environment remains essentially unchanged from recent periods, payment trends have been modestly higher than we had previously anticipated, resulting in a need to modestly add to these reserves. Turning to financial strength. All of our indicators remain in excellent shape. Holding company liquidity of $2.04 billion and all of our capital ratios were at…

Thanks, Jay. So very strong results. And from an insurance operations perspective, I believe there are 3 key takeaways that apply across all 3 business segments: First, we're really pleased with the progress we've made on our strategy to improve margin, both in the quarter and year-to-date; second, given the interest rate environment, the weather volatility and loss trend, we need to continue to execute on this strategy; and finally, we see opportunity for better results in those areas of our commercial business that are at the lower end of the return spectrum and in the case of Personal Insurance, where weather losses have been most challenging. With that, let me go to the specific segment results. Beginning with Business Insurance. Operating income was up 85% year-over-year with catastrophe losses lower and favorable prior year development higher. Excluding these impacts, operating income was $549 million, up nearly 36% from the prior year quarter and the combined ratio improved more than 6 points. The strong results this quarter were driven primarily by earned rate increases in excess of loss cost trends and lower non-cat weather-related losses. Looking at the production statistics starting on Page 9, we saw renewal premium change, pure rate and retention all increased in the quarter. Renewal premium change was 9.3%; rate, 7.8% and retention was above 81%. The rate progress that we continued to achieve was broad-based with increases between 6% and 10% across all lines led by Workers' Compensation and Commercial Auto. These increases in pricing, along with the growth in exposures and audit premiums helped to drive net written premiums up 5% year-over-year. New business was somewhat lower in the quarter as it has been for the last several quarters. Over time, pricing on the new business we've written has improved. But given our strategy…

Thank you. Andre, we are ready to start the Q&A session. And if I could please ask participants to please limit yourself to one question and one follow-up. Thank you.

[Operator Instructions] Our first question comes from the line of Randy Binner with FBR. Randy Binner - FBR Capital Markets & Co., Research Division: I was wondering if you could provide a breakout of the reserve release by business line within the commercial line. So the $208 million outside of the A&E, just wondering where that broke out and specially with Worker's Comp, wanting to see if that was net redundant and especially in recent accident years.

Yes, this is Jay Benet. So we look at what's taking price in terms of reserve development in Business Insurance. As we said, it's net about $41 million. And if we look at the components of that property as a big component, if we look at favorable loss development, where did it come from, the recent accident years, 2011, was a big driver of that. And we also had some better salvage and subrogation recoveries in some of the older years than we had expected. Some other areas, CMP, same sort of thing in terms of the property-related items. Worker's comp actually developed a little favorably mostly due to some of the tabular claims that we had for accident years 2010 and prior. We had some strengthening of Commercial Auto given some of the claim severity that we've been seeing in recent accident years that we've been talking about. But net-net, as you saw, everything was on -- was favorable. Randy Binner - FBR Capital Markets & Co., Research Division: And I guess, I mean, did that -- I appreciate the color. I mean, so the Worker's Comp was stable, which is, I think, unique to Travelers. But kind of stepping back, I mean, how should we think about where the reserve trends going overall. For the industry overall, I think, the expectation is that redundancies would diminish. Travelers has had an excellent reserve result. But as this goes forward, should we think of this redundancy as winding down?

We don't make any predictions whatsoever about where our reserve development is going. We make our best estimate every quarter and we stand by that. And then, as the data develops, as more information comes in, we adjust the reserves accordingly. Randy Binner - FBR Capital Markets & Co., Research Division: And just one more on comp and then the accident year '11, was there -- was that also favorable or was that deficient?

In the quarter, it was essentially breakeven. It was a tiny bit -- we had a tiny bit of strengthening in the 2011 accident year in the third quarter. It was under $10 million. So this is kind of the quarterly analysis that we do to get things right and effectively given the size of the reserves, we're talking about rounding them.

Our next question comes from the line of Mike Zaremski with Crédit Suisse. Michael Zaremski - Crédit Suisse AG, Research Division: The underlying combined ratio improvement within the BI segment clearly very healthy. Can you provide any color on which lines of business within the segment are exhibiting the most and/or least margin improvement?

This is Brian, Mike. I would say it's pretty broad based. As we've said, comp and auto have been getting the most rate increases, but the range across the lines is 6% to 10%. So everybody is getting some significant rate. As Jay Benet just mentioned, Auto is one of the lines we're seeing a little more heat, so we need the rate there. But I think that by and large, it's been really across the board with our products. Michael Zaremski - Crédit Suisse AG, Research Division: Okay. In terms of -- if you're getting -- what kind of rate increase are we getting in commercial auto then? Because it seems like we've been -- it feels like you guys have been getting healthy rate increases for a while now and you're still talking about margins not being that great in that line.

Yes. I think about 9. Yes, 9 points in auto and comp was around 10, so those were at the top end of the spectrum. Which I think it's consistent with what we've shown in the past. We've talked about and others have talked about the bodily injury trends in the auto business, both commercial And Personal. And so that's been one of the reasons why we're getting those increases. Michael Zaremski - Crédit Suisse AG, Research Division: Okay. And then as a follow-up in regards to rate increases across all of the segments. Is there a dynamic of stronger momentum in some of the longer tail lines that are more impacted by the low interest rate environment? And are you at all surprised at how well retention is holding up in the face of the rate increases?

I don't think there's any -- yes, obviously, as anybody looks at the economics of the product, the longer tail lines which would be more significantly impacted by the investment yields, that factors into how we think about it. I'm not sure it factors into how the customers think about it, so we work at it. And on the retention thing, I would go back to our comments on Slide 13. So we're obviously pleased that aggregate retention is going up, but it's really the texture underneath that we obsess about and are we keeping the right accounts. So honestly, I'm not really surprised that we've been able to execute in a way where we're keeping the best business. And...

And certainly -- it's Jay Fishman. And while the interest rate environment certainly plays a meaningful role in the evaluation, it is just one factor. And I'd make an observation that our Worker's Comp portfolio, broadly, and of course you can get different conclusions in different businesses, but broadly speaking, has actually been doing better. I'm not sure, Brian, how you might characterize it, better than what we perceive the rest of the industry to be or it just hasn't been -- it hasn't been in a situation where the rate need there has been enormous. I'm not sure how to exactly characterize it. But we're starting from -- we're starting from a position in Worker's Comp that was not as problematic as, I think, sort of conventional wisdom might have thought.

Well, you can look at our line results for a good number of years, and our performance in Worker's Comp has been significantly better than the industry. We think that's due to a lot of things. Risk selection. We think our claim model around medical management is the best in the business and all that stuff really matters.

Well, Gabby hosted an Analyst Day where there was a considerable amount of time spent in the nuts and bolts of Worker's Compensation management. And really, I think demonstrated clearly, or at least attempted to demonstrate clearly, the real competitive advantages that we bring to the Worker's Compensation line. So again, it's not that we're -- we are just not all in the same situation or the same environment. And so I'd resist the temptation to just leap to the conclusion that long really needs help and short doesn't. It's just more -- unfortunately, more complicated than that. Michael Zaremski - Crédit Suisse AG, Research Division: Okay, that's helpful. And I agree that you guys did demonstrate very well that you have guys have a best-in-class Worker's Comp business at Travelers at the Investor Day.

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

You guys have done a great job of taking rates pretty significantly. Just curious what you're seeing so far in the fourth quarter? And the reason I asked is because, you've taken rate starting fourth quarter of last year. So I just wanted to see if you're successful in getting significant rate on rate.

Yes, Vinay, this is Brian. Let me take a shot and I'll start with, we say it all the time, but we're not prognosticators here. We're not going to guess at what's going to happen in the future. I'd point to 2 things. Number one is what we saw in the third quarter, we were encouraged by. So the results that we were able to post, the texture underneath the results, and the dynamic and the feedback that we've gotten from our agents and our customers. Anecdotally, as we look at what we're working on in the fourth quarter, because we've obviously got some view of that but it's anecdotal, is similarly encouraging. So we continue to see a little bit of progress here. And at the end of the day, whether it will be up a little or down a little, we don't know. But the anecdotes right now feel pretty good. But we're also real cautious. The world can change in 30 days, and we'll see where we are at the end of the next quarter. But that's our mood right now.

Okay, fair enough. The second question is, how much of a benefit that you guys get from low non-cat weather? I think you mentioned that it was slightly below normal this quarter.

This is Jay Benet. I think, when you take a look at what's taking place in Business Insurance as an example, you see a very strong increase -- I'm sorry, increase in profitability, decrease in the combined ratio. And it's difficult to split out the components on an exact basis, but I would say this, at least, a couple of points in there, where favorable non-cat weather has influenced us. And I think, if you go over to PI, you've got the same kind of result. I don't know if it's couple of points or not, but certainly working its way through and seeing some heads nodding around the table. So I think, a couple of points is probably right.

But it's a very -- it's truly a guesstimate because you're asking us to compare a level with whatever we thought normal might have been. And I don't know that we know normal compared to anybody else. So it's a challenging question to answer. And again, because you're asking for what was your experience relative to what you thought the experience would be, I don't know if what we thought was the right number to begin with. So that's the complexity.

Vinay, let me -- it's Brian, let me give you a perspective at what you might be looking for, which is the margin expansion in BI, we think, was about 2.5 points of the combined ratio. So the price over the earned rate over the loss trend dynamics. So -- then the rest has got non-cat weather, it's got other stuff that always -- large loss movement that goes through the quarter, et cetera.

Our next question comes from the line of Josh Stirling with Sanford & Bernstein. Josh Stirling - Sanford C. Bernstein & Co., LLC., Research Division: So in the Q, you seem to be guiding to -- for a continued pricing in the fourth quarter in 2013, and I'm trying not to be redundant in the last question. But we're all trying to figure out how pricing -- how long pricing lasts. And I'm wondering, 2 questions -- well, I'll ask them first and you guys can respond as 1Q. First, when you're actuaries do indication analysis, how much more -- what's the output of that? How much more rate do you think you actually need to get to target accident year, sort of ROEs and target accident year returns? And then separately, Brian, I'd love to just follow up, you made the point about sort of visibility to a pipeline, obviously, you have the sense of what's going on in sort of actual production that you haven't yet reported. But you also obviously have to file rate increases and -- across probably what, 30% or 40% of your business, at least, where that's primarily sort of driving the rates you end up charging. And I'm wondering if you can give us a sense of actually what's in the pipeline and what you think you'll be effective in written rate levels in the fourth quarter or '13 as we look forward.

Josh, it's Jay Fishman, I'll take the first part of your question. We have not set a goal, a threshold number at which we stop. And it's driven by the fact that the environment remains really unpredictable and somewhat volatile. Not only do you have the weather dynamic, and obviously the interest rates which are -- at least, the interest rate portion is quantifiable. But you've also got the uncertainty of the fiscal cliff and what happens to the economy and all the factors around it. So for the moment, what we've decided to do is to continue to business it -- continue on business as usual, continue to pursue what we're doing because we know that to produce the kind of returns I think investors will come to expect, we need to do more. So that's honestly, as far as we've gone, which is we're going to continue to do this and we'll see how we do. And I have no idea what the competitive environment will be or what will happen to us. It is nonetheless a very competitive environment and remains fragmented. All we can do is tell you what we're intending to do for the next -- for the near-term outlook. I don't even want to go beyond that. All I want to do is to say in the near term, this is what we're going to do and we'll evaluate how successful we are and do the data and analytics continue to show us that we're doing the right things and on the right track. And we just don't have that goal line to say, we're X months away from hitting it. It's just -- we just don't have it, and that's really all we can say on it is business as usual, and we're going to keep going.

On the filings perspective, you could -- someone could go and pull them. I would say on the personal line side, we don't have any real big filings in the works that are going to change where we are today. So we took a pretty good bump up in Auto and that was because some things kicked in, in different states and so a little bit of steady as she goes there. In Select, we've recently put in some filings, which are just becoming effective. And so we've got some momentum maybe there, and those would be the 2 big places where we've got filed rates.

Where filings matter. In Commercial Accounts, really it's not a file environment, we're really doing -- wrestling with that. And I'll just kind of revert back to Brian's comment about the anecdotal observations. It's still pretty early in the quarter. And so what we're getting is feedback from our folks in the field who are in agents' offices every day, renewing accounts and having conversations and getting a sense of what the market is. And the anecdotal feedback -- and I wouldn't overemphasize it because it's -- it can change. First of all, it might not be right, it might not be accurate of the real environment and it might change. But the anecdotal feedback that we're getting is -- continues to be so far, so good. And so it's encouraging and gives us a basis for saying, we're going to continue. That's -- it's about as good as we can do right now for you. Josh Stirling - Sanford C. Bernstein & Co., LLC., Research Division: So Jay, not -- just to paraphrase, make sure I get it. So basically looking out over your guys' planning horizon, you're not going to give a target, of course. But ultimately, you're going to march to your own drummer, raise pricing, sort of get to continued levels and the market's going to what the market's going to do, but you're going to drive to sort of continued pricing momentum along the lines you've been doing thus far in 2012.

Well, what I'm -- what we don't want to do is to tell you that with certainty that we're going to succeed because I don't know. It's just -- it's been environment where every company decides its own strategies and has its own conversations with agents and brokers about individual accounts. So all I can tell you with some degree of certainty is in the near-term how we're going to act. That I can tell you because we drive that. How the market responds and how competitors perceive the marketplace and what they do is well beyond our control. The data that we have from the third quarter is encouraging that we're not hitting some wall. The important thing -- the question that I asked definitely here is, are we achieving what we're trying to achieve in the market? A lot of your -- not your questions, particularly, but many of the questions that we get seem to be, are we asking for 20 and getting 8? Or are we asking for 8 and getting 8? And that actually is a relevant question and it really -- we're not trying to get 20. We're not attempting. We've always said from the beginning of this that it was going to be disciplined, thoughtful, account by account, loss driven, evaluated in the context of interest rate changes, thinking about weather, and most importantly, nondisruptive, nondisruptive to our agents and our brokers and our insureds. We're trying to manage a business and relationships. This is not a trading account. It's real people with real businesses, paying real bills and real agents that stand between us, and we're just trying to manage this thoughtfully, gently, doing what we need to do to make sure our business remains viable and has substance to it, but not doing it in the way that causes disruption. So again, I look back at the third quarter, the data all says we're making that happen and gives us a basis for saying, let's keep going.

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

I guess, one question I have is when you look at your profitability and kind of what you've talked about from here, what do you think is more important or how would you kind of prioritize the rate actions you're taking versus the actions you seem to be taking on the loss management side? And where do you think the industry is on those measures, whether it's terms and conditions or infrastructure to help kind of push down claim cost? And just one follow-up.

Brian and I are going to split this. I think if you looked at Business Insurance, our strategy is predominantly rate driven. We've always been a company that's been actively engaged, helping our insureds manage and lower their losses. We have a substantial risk management group. They are involved and always have been with our insureds to try and help them manage their losses and do better business for them. So that's -- our dynamic in Business Insurance really price driven, rate driven. In Personal Insurance, underwriting terms and conditions play a much more significant role, and I'm going to let Brian speak to that.

Yes, I mean, and I'd say clearly, personal line heavily on the Homeowners side. And to some degree, more broadly property all around. But the place where terms and conditions are really important is our Homeowners' book. And just to give you a perspective. Industry deductibles in the '70s were typical average $250, maybe a $500 deductible. Now the average is probably a $500 deductible. If we just kind of loss inflate that over the 40 years, we think the average deductible should be somewhere around $2,000. So the product has changed. So we're driving deductibles. We've actually been pretty successful historically in getting deductibles on the coastal hurricane exposed, so that's not the issue. The issue has been the inland tornado, hail exposed and we're making real progress. And so there, that's important. The other dynamic is underwriting, and a lot of that is around roofs. So age and quality of rood is an important underwriting decision now on time reporting, on claims and is different in the policy in a lot of places than it was before. So all of those pieces really matter on the property side.

And our hope in deductibles, by the way, is not unique to loss mitigation, that's an important part of it. But we also believe that customers shouldn't be buying more insurance than they need, right? Insurance should be to protect for catastrophic events. And we would encourage all insureds to not pay for more insurance than their own capacity to deal with losses would suggest. And so part of this is not, again, not just loss mitigation important, but it's also to make sure that we're doing the right things for our customers even if they don't always see it exactly that way. It's a $500 deductible today becomes a policy that responds to a lost iPad, and it just isn't the way a homeowners policy was really designed to be.

Got it. And just one question on the directors. I know we haven't talked about it a lot recently. But I mean, it seems they kind of be floating along in terms of premiums are growing a little bit, the drag on, the combined ratio is about flat, looks like 5% to 7% of earnings right now. And what is the strategy there? I mean, it's -- you've kind of been in this state for a little while. At some point, would you look at trying to get -- purchase something to help kind of build that platform? Or is it -- or is building it still kind of the expectation?

Well, we have been and continue to be in the test-and-learn phase. It's -- and I don't mean test. What I'm talking about is testing in the very specific dynamics of building the business, whether it's how you get customers to respond or how you encourage a high conversion rate from response to quote. And then most importantly, how do you generate sales from quotes, and that gets to not only the technology or interactive behavior at a call center, but also to the pricing structure. And so we're still in the phase of doing what we need to do to gain the learnings necessary to launch, in effect, sometime down the road a successful business. This is not yet launched. We are still in the test-and-learn arena and we have more work to do in that regard. We're going to resist the temptation. We've done it -- I've certainly in the old, old Travelers days, have been involved into one where we launched a program way too early and had a marketing success and a financial disappointment. And that's a real problem. And so we're going to resist the temptation to do this early. There are always young customers coming along. There's -- it's not as if there's a finite group of customers and they're there. And once they have insurance, they're gone. We all get a day older every day, and there's always another group of customers that are emerging, that are getting drivers licenses that become young adults that need to buy insurance. And so as we think about it that way, we're more than willing to be patient and do it right. More important to do it right than to do it quick and fix as we go. So that's where we are.

And to the timeline from here is just kind of as it develops. And do you feel like you're getting the learnings from it to kind of offset the cost that you're investing in it? And I'll stop.

Well, we're certainly learning whether it offsets the cost or not ultimately depends upon whether we conclude that we have the ability to launch a successful business. I mean if you're asking, do we think that, that's still -- are we still on track to do that? The answer is, yes, but nothing is ever certain. And the real question that we're down to now, as we've said this repeatedly, we are charging the same price in our direct channel that we charge in our agent channel because it is the only price that we know. And the real question is going to be, in that kind of framework, can we actually write a policy, underwrite a risk and produce a reasonable return based on that, whereas, there's some other pricing structure that's going to be necessary? And it's going to take us some time to figure that one out. If we end up concluding that we have everything that's necessary to build a business, yes, we will recover these costs. If it turns out that we ultimately conclude we can't, then unfortunately, we won't. So the substance of your question is, is a challenge to answer, but I -- but we still feel pretty good here. The people working on it are first-rate people. They -- one of the things, and I've said this before, I used to think this was an insurance business. It actually is a lot more about human dynamics and technology management than it is about insurance. And it took a little time for us to even understand that. I think we've got people from outside the insurance industry with real expertise in this arena, and we are learning a lot, really a lot, and we're hopeful is sort of where it stands.

Our next question comes from the line of Jay Gelb with Barclays.

I wanted to touch base on 2 items. The first is the exposure growth in Business Insurance. Based on what you sort of see as the forward-looking indicators, to what extent might that pick up over time?

We're looking at each other here trying to figure out what do we know in the data and what do we see. It's -- first, one of the things we have learned is that notwithstanding the fact that intuitively, we keep thinking that the exposure data is going to turn out to be predictive of the overall economy, and we've spent some time trying to actually link those up and in effect, build an index around it. It's just not nearly as predictive as we might hope. If you look at the data broadly, I think what you would characterize it as is sluggish growth. Now those are words against some numbers, but that's kind of what we see. Now it's -- we should be careful not to confuse the dynamics that we measure in exposure to GDP. We're measuring payroll, we're measuring projected sales, we're measuring trucks on the road, we're measuring real estate square feet. And as we all know, you can have recoveries that have -- jobless recoveries, not my words, words other people have used. And so our data suggests sluggish is what you'd see, but I'd be careful to link that up to a conclusion about GDP. It's not nearly as predictive as we might have hoped.

Okay. Next issue, on the return of capital. I think in the past, Travelers has talked about returning, in terms of the combined share buyback and shareholder dividend, having that equal to around 100% of operating income. And I think, on track year-to-date, it's around 75%. So based on the commentary that, that will probably be higher in 4Q versus 3Q, I'm just trying to get a sense of whether you feel confident that could get closer to 100% as we think about next year.

Jay, this is Jay Benet. As we said, we do these things based upon expectations. And thankfully, we have a good problem that our earnings were stronger than our expectations, so we created excess capital. So it really is a matter of timing as to when we repurchase shares. The expectation, as we said, for the fourth quarter is to do something higher than what we've been doing on a run-rate basis for the first 3 quarters. And as we say in our disclosures, there's lots of things that go into that decision. There's capital that we need to support the growth of the business. There's capital that we perhaps need or cash that we need to deal with what's taking place in the funding of our pension plans. I might add that we don't have a very large outstanding pension plan given the history of the company. But when you have interest rates as low as we have today, we want to make sure that the thing is properly funded against the assets. So we'll look in all of that, but at this point, what we could say is that we have created the excess capital beyond what we need at this stage and we haven't bought it back at the same rate, so we will be increasing it. And over time, which I think is really the critical phrase, over time, it will be the earnings that will drive the amount of capital that we return to shareholders, both in the form of dividends, as well as the share repurchases.

All right. And I just want to clarify a prior response. I think there was a question about pace of rate increases in BI in 4Q. Now you were saying roughly similar to 3Q, and I just want to clarify that, that using the 7.8% rate increase that was in 3Q is sort of the baseline.

No, no. You've misheard that, I'm afraid. We were asked the question and we said, we don't make predictions and we're certainly not prognosticating on rate gains. What I think Brian said was that the anecdotal feedback and that's all it is, it's not data, it's just anecdotal feedback just suggests so far that the fourth quarter is, so far, is sort of looking the same way the third quarter did. Don't over read that. There's no precision in that comment other than conversation. That's all it is. But if we had heard from people out in the field that something substantive had changed, that there'd been some meaningful difference in the environment, that I think is important. We've not heard that. We just heard, yes, it's about the same. Now it's going to be a matter of what accounts actually come up for renewal, what's the mix, where on that grid do they fall, what lines of business. So no predictions whatsoever. And all we were doing there was providing the anecdotal feedback that fourth quarter is sort of so far, so good. Bill is -- I'm looking at Bill and asking him to nod, yes, and he did. So that's where we are.

Our next question comes from the line of Matthew Heimermann with JPMorgan. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: One clarifying question was just, I think, it was Brian who said about 250 basis points of kind of core margin improvement year-on-year. And obviously, the last quarter had a bunch of non-cat weather and large losses in it. But I guess, what I was trying to get a sense of was, how was non-cat weather, large loss activity in the current quarter relative to normal? Was that a benefit or not? And if it was, any numbers you could put to it?

I'm trying to -- so 6% -- if I look at the quarter-to-quarter combined ratio in Business Insurance was improved 6% CapEx, ex-cat, ex-PYD, about 2.5% of that was the earned rate over the loss trend. The next largest piece would have been the non-cat weather which was clearly better, and then we'd had large losses, I think that's the right sequence. I don't have the exact numbers on each of those pieces, but... Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: I'm sorry, I think I confused the issue. I understand on a year-over-year basis why the non-cat weather and the large property is a such a big contributor. I guess, I was -- if you look at 3Q of this year, individually, would you characterize non-cat weather and large loss activity as kind of a nonfactor in kind of the way the underlying margin played out? Or was there any benefit to speak of? Because you've mentioned in the press release, but I wasn't quite sure if that was just with respect to the year-over-year or there was something -- it was just a reversal of that a year ago or if it actually was also good this quarter.

Yes, the bigger impact was the reversal of bad from a year ago with regard to non-cat weather activity. But as it relates to the current quarter, the current quarter also was very light in weather, whether it was cat or non-cat. So we did have a benefit in the current quarter from what we would otherwise consider to be normal. And as Jay Fishman said before, the consideration on normal has lots and lots of judgment in it. So we're not going to sit here and say, well, it was worth 1 point, 0.5 point or 1 point. But what we can tell you is that the weather was benign in the third quarter and our results benefited from that. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: Okay, that's helpful. And then one question just on the construction surety, volumes are in decline. I was curious if you could get a sense -- give us a sense of how much of that is coming from commercial end markets versus residential end markets? And is there anything you see kind of as you look out that potentially could start to dampen that trend?

Yes, this is Brian. I'll flip it to Alan. But we don't do a lot of residential. So...

It's way more driven by municipal finance and municipal construction, so not very much of it would be coming from the residential side. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: Okay, but I would assume that...

Well, we're not active in that market. Our data is largely driven by the municipal arena but it's not because, it's not the -- we don't know what's happening in the residential segment, we're just not a real participant.

But way more of the surety market relates to municipal construction projects where there is a statutory requirement for the surety product. So you don't have the same requirement or the same level of surety premium coming out of residential construction generally. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: Well, I guess, kind of as a pass-through though to the municipal piece, there's always part of that is extending services into new areas, correct? So I guess on -- I'm just -- that was kind of what I was thinking -- asking the question was kind of what I had in mind, completely recognize the comment.

The infrastructure buildings, roads and other utilities to service, development broadly. But I certainly don't have at my fingertips have the breakout between that and say highway construction. I wouldn't have that breakout.

Our final question comes from the line of Greg Locraft with Morgan Stanley.

Just wanted to -- the one disconnect that I'm sort of wrestling with is the top line, in other words, premiums written and earned premiums. The percentage gains year-over-year are just not even close to where rates are at, and retentions look good. So it seems to be coming down to new business. So my first question, I guess, is it new business falloff that's sort of driving the lack of flow through, the actual reported top line? And if it is, I'm sort of wondering why do you think people are choosing not to join Travelers as much as they were before?

So the short answer is, yes, it is primarily new business. As you pointed out, retentions are down slightly in most of our businesses, but very solid and we feel great about those. We're getting price increases.

Yes, down slightly from historical norms.

From historical norms, right. And actually across Business Insurance, for example, the overall retention, on a historical basis, a very strong number. Premium is up because we've got price exposure and audit premiums kicking through there. But across much of our Business Insurance area and clearly in personal lines, as we've executed on our strategy of improving margins and raising prices, the new business marketplace has become more difficult. And...

I think I'd separate the discussion into Commercial and Personal, because I'm not sure they're nearly as similar to that. But first, just an observation business, low exposure is a part of this 2, low exposure growth. That's talking about the overall, general economy. In Business Insurance, there is a phenomena that's been interesting to us. We have historically tracked a relationship between renewal pricing and new pricing. And in this quarter but also in the second, the spread between for us -- I'm not speaking about a broad market comment here, I'm only speaking about us. The spread between renewal and new that we booked actually narrowed, meaning it was better than historical standards would be. So in the context of the new business that we're taking on, we feel pretty good about it. If there's an opportunity here, if there is, it's making sure that our folks who are working so hard and focused on renewal -- renewal retention and gaining rate, don't ignore the new business opportunity. I don't believe, Greg, and I can't prove it to you, but I don't believe that it's people not choosing, as your words, not choosing to be with Travelers. I think if anything here, it's making sure that our folks have the energy as they go through the renewal process to make sure that they don't miss the opportunities in new that at these pricing levels any way are increasingly attractive. And we are focused on that and suspect that -- if you talk to an underwriter, they will tell you that renewing an account flat in a flat environment is pretty quick and pretty easy. But if you're discussing a rate increase with a broker or an agent on an individual account, it just takes more time. A lot…

Yes, that is. And just a quick follow-up then. Turning just to the business side. It sounds like that's a prioritization or a point of emphasis within the organization, deciding in a new business versus retention. I mean, my understanding is a point of retention is -- has a far, far better payback within your P&L than new business. Do the people within your organization get paid sort of the same for retention versus new business? Or in other words, are they agnostic or are they actually pursuing what is -- where the highest payback is in the Travelers' P&L, which is go get rate on the current book?

Well, we can tell you how they're compensated. I hope that they're reacting consistent with it, but of course, you never know. We don't pay for production, we don't pay for new, we don't pay for renewal, we pay for an overall evaluation of performance, not only at an individual office level but actually right down to an individual underwriter level. The data analytics package, I spoke about earlier, really makes it possible for us to have a very granular conversation with each of the frontline people about how they've done. What you can't do is tell people that you really want to improve profitability, and when they take the right steps, even if that means that they lose an account, that you somehow think that's a bad thing. You have to embrace it and understand that it's a good thing. So the numbers, in and of themselves, of production, if you pay based to that, you can get some, not only unintended consequences, but frankly, it will make it difficult to employ the strategies that you know are right to improve profitability. So our payment process is one of -- let's evaluate what each of them has done in the past 12 months as everybody else in this place is evaluated, and we make the payments accordingly. I do think that -- it is hard being a field person right now. They are juggling lots of accounts, negotiating renewals, driving for rate, having to do the analytics behind all that to make sure that the selections they make are thoughtful and smart. And then you tell them, and by the way, don't forget to make sure that you're following up on new business opportunities. I do believe -- yes, and I can't prove it, but I do believe that it is a matter of emphasis and time management far more than it is pricing out of the marketplace. I have no indication that, that's the case or that there's some substantive dynamic of customers choosing away from us. I don't believe that either.

This concludes our Q&A session. I would like to turn the call back over to Ms. Nawi.

Very good. Thank you very much for joining us. And as always, myself and Andrew Hersom are available for follow-up questions. Thank you, and have a good day.

Ladies and gentlemen, this does conclude the conference for today. We thank you for your participation, and ask that you please disconnect your lines.