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 Doreen Spadorcia - Vice Chairman of Claim Services, Personal Insurance, Operations & Systems and Member of Operating Committee

Randy Binner - FBR Capital Markets & Co., Research Division Jay Adam Cohen - BofA Merrill Lynch, Research Division Brian Meredith - UBS Investment Bank, Research Division Adam Klauber - William Blair & Company L.L.C., Research Division Meyer Shields - Stifel, Nicolaus & Co., Inc., Research Division Michael Nannizzi - Goldman Sachs Group Inc., Research Division Michael Zaremski - Crédit Suisse AG, Research Division Amit Kumar - Macquarie Research Vinay Misquith - Evercore Partners Inc., Research Division Gregory Locraft - Morgan Stanley, Research Division Jay Gelb - Barclays Capital, Research Division Matthew G. Heimermann - JP Morgan Chase & Co, Research Division

Good morning, ladies and gentlemen, and welcome to the Fourth Quarter and Full Year Results Teleconference for Travelers. [Operator Instructions] As a reminder, this conference is being recorded on Tuesday, January 22, 2013. 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, Andre. Good morning, and welcome to Travelers' discussion of our fourth 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, Vice Chairman and 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 your 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. Jay S. Fishman: Thank you, Gabby. Good morning, everyone, and thank you for joining us today. Given Storm Sandy, we are very pleased to report fourth quarter net and operating income of $304 million and $278 million, respectively. For the full year, we reported $2.4 billion of operating income and operating return on equity of 11% and a 9.8% return on equity. Our financial and operational results for both the full year and the quarter were strong and demonstrate the success of the strategies we have implemented over the last few years. In Business Insurance, excluding National Accounts, we continued to achieve meaningful positive rate gains of nearly 8% in the quarter, while retentions continued to remain stable and new business remained at only marginally lower levels. Our Management Liability businesses have also produced similar pricing successes over the past year, achieving nearly 8% rate gains in the quarter. The operating environment for our businesses remains the same. Interest rates remain at historically low levels and weather patterns continue to be uncertain. Catastrophe losses this year totaled $1.2 billion after tax, making it another year with a very high level of catastrophe losses by historical standards. As such, we remain committed to our strategy of pursuing rate gains across all of our businesses in order to improve returns. We are particularly pleased with the rate gains we've achieved over the past 2 years and the compounding effect we are now recording. In Personal Insurance, both our Home and Auto business were also significantly impacted by Storm Sandy. Notwithstanding that impact, we are especially pleased with our Agency Homeowners business, which produced a 96.4% combined ratio for the full year and a 78.7% underlying combined ratio for the year. In our Agency Auto business, while we believe we are very much on the right path, we are not yet satisfied with our results. Our 99.2% underlying combined ratio for the year is not at a level that produces sufficient returns. As we have discussed previously since the fourth quarter of 2011, we have been experiencing an increase in auto severity, resulting in levels that have continue to exceed our expectations. But with the rate increases we have already achieved, earned rate gains exceeded our updated view of loss trend in the third and fourth quarters. In this regard, we are now achieving widening margins and we intend to continue this strategy. Assuming that we continue to achieve rate gains at this level, and the loss trend does not materially increase from current levels, this will result in improved underwriting profitability in the line. Brian will discuss this in more detail later. Turning to capital management, we remain committed to our strategy of returning excess capital to shareholders. In that regard, we repurchased $400 million of stock in the quarter, bringing total capital return to shareholders in 2012 to over $2.1 billion, including $700 million of dividends. Since embarking on our long-term strategy of returning excess capital to shareholders in 2006, we have reduced our shares outstanding by $362 million or 52%. Just another quick word on Sandy. The storm had a significant human impact on Travelers, as well as our customers and employees. We're extremely grateful to all the individuals who made personal sacrifices to help us deliver on the promise we extend with each one of our policies. Storm Sandy produced the largest number of claim notices in our history and even in this event, we were able to handle virtually all non-NFIP claims with our own adjusters. Just to give you a few highlights, in total, nearly 5,000 employees contributed to our response, we followed up with 80% of all claimants within 48 hours after their initial claim report and virtually 100% were recontacted within 72 hours and we paid or resolved approximately 90% of our Personal Insurance, non-NFIP claim estimates within 30 days of the first loss notice. These are really solid results that I was able to witness firsthand. Our claims group really makes us proud. In closing, we remain very pleased with the ability to deliver rate improvements across all our segments and generate superior returns over time. We entered 2013 with a real sense of optimism based on the strength of our competitive position. And with that, let me turn it over to Jay. Jay S. Benet: Thanks, Jay. Let me begin by stating that, as always, we've maintained our strong cash position, ending the year with holding company liquidity of just over $2 billion. Operating cash flows were over $450 million for the quarter, despite the high claim payments related to Storm Sandy, as well as a discretionary $150 million contribution that we made to our qualified pension plan to maintain its high funding level even in this extremely low interest rate environment. Note, too, that for the full year, we generated over $3.2 billion of operating cash flows and, as Jay said, we returned over $2.1 billion of excess capital to our shareholders, consistent with our ongoing capital management strategy. Turning to reserve development. We once again experienced net favorable prior year reserve development on a consolidated basis, $222 million pretax for the quarter, up from $126 million in the prior-year quarter. Also, as has been the case throughout the year, each of our business segments experienced favorable development. Business Insurance accounted for a little over half of this quarter's total, driven by better-than-expected loss experience related to last year's catastrophes, lower-than-expected claim department expenses or ULAE, and a modest improvement in Workers' Comp reserves. Most of the rest of this quarter's favorable development came from the Bond & Financial Products businesses within FP&II, and surety and management liability. For the full year, net favorable prior year reserve development, on a consolidated basis, was $940 million pretax, up from $715 million in the prior year. I'd also like to share with you a preliminary view of what our 2012 Schedule P will look like when it's filed. All accident years, other than 2011, have developed favorably, including accident years 2002 and prior, notwithstanding the A&E charges we recorded earlier in the year. Unfavorable development related to the 2011 accident year was small, only approximately $155 million pretax that was mostly due to higher-than-expected severity in Commercial and Personal Auto. Similarly, looking at our preliminary Schedule P data on a product line rather than on an accident year basis, shows that all of our product lines, with the exception of Commercial and Personal Auto, experienced net favorable prior year reserve development in 2012. Commercial and Personal Auto developed unfavorably on a pretax basis by only approximately $110 million and $25 million respectively, again, driven by the 2011 accident year. Finally, I would note that all of our capital ratios remained at or better than our target levels at the end of the year. Net unrealized investment gains, which were up for the full year, decreased slightly during the quarter to almost $4.8 billion pretax or $3.1 billion after-tax, while book value per share was $67.31 or 8% higher than at the beginning of the year. So with that, let me turn things over to Brian. Brian W. MacLean: Thanks, Jay. I'm going to give an overview of the segment results, beginning with Business Insurance. Although operating income was down quarter-over-quarter in Business Insurance due to the losses from Storm Sandy, the fundamentals of the business continued to strengthen. As Jay just mentioned, prior year reserve development was favorable and higher than last year, and the underlying combined ratio of 92.8% was a 4-point improvement from the fourth quarter of 2011. On a full year basis, the underlying combined ratio improved more than 3 points year-over-year, with about 2 points of that improvement due to earned rate increases that exceeded loss cost trend. Looking at the production statistics, starting on Page 9, retention continued to be strong at 80%, while new business was up slightly versus both the previous quarter and the fourth quarter of 2011. Net written premiums increased 6% in the quarter, with the largest increase in Workers' Comp, driven by higher pricing on our guarantee cost business and growth in residual market pools. Drilling into the pricing results, renewal premium change was 10%, up about 1 point from the third quarter, driven by pure rate increases of 8% and exposure of 2%. The 8% rate gains we achieved this quarter were up slightly from the third quarter and up 2 points from the 6% that we saw in the fourth quarter of 2011. The rate increases ranged from 6% to 10% across all lines and were once again led by Workers' Compensation and Commercial Auto. I would note that within the quarter, rate change was highest in the month of December, with total Business Insurance rate change of 8.3% and Commercial Accounts rate change of 9.8%. Loss trend, excluding catastrophes, has remained at approximately 4% overall. Specifically, the loss trend concerns in Workers' Comp and Commercial Auto that we spoke about several quarters ago, appear to have mitigated and are trending toward long-term historical trends. So overall, excluding weather, a very stable picture. Now I'd like to take a moment to discuss our results in Select, our small commercial business. In this business, we are pleased with the performance of our platform and operational dynamics, and we've seen great success in the marketplace. On the pricing side, as seen on the data on Slide 10, we've also made significant progress. Renewal premium change this quarter was nearly 12%, up consistently over the past 2 years. Along with these pricing gains, retention was up 1 point in each of the last 3 quarters, coming in at 78% this quarter. New business was down somewhat from recent quarters due to an increase in new business pricing. These pricing and underwriting actions have had a meaningful impact on our product, profit margins in this business, and we are very pleased with the results to date. So summarizing the entire Business Insurance segment, we were able to sustain our level of rate gains this quarter on top of the 6% gains from the prior-year quarter. This compounding of pure price improvement, combined with the 4% underlying loss trend, has had a meaningful impact on returns. However, given the ongoing weather volatility and challenging investment return environment, we will continue to -- we continue to see the need to execute on our targeted pricing strategy. In the Financial, Professional & International segment, our operating income decreased by 14% due entirely to the impact of Sandy, largely in our Lloyd's business. The underlying combined ratio improved about 3 points, for both the quarter and full year, driven primarily by increased rate in portfolio management in our Management Liability business, as well as risk selection in International. In Bond & Financial products, net written premiums were flat quarter-over-quarter, with higher production in Management Liability, offset by continued decline in surety volume. In Management Liability, retention remained strong at 85% in the quarter, while pure rate gains increased to 8%, making this the sixth consecutive quarter of sequentially increasing rate. Renewal premium change was 6%, which included negative exposure of 2% in the quarter, driven by a reduction in multiyear policies. In International, net written premiums were up about 4% or $12 million on a constant currency basis. The increase was driven by results in the U.K. and somewhat lower levels of seeded premium, partially offset by lower volume in the Canadian surety market. The International underlying loss ratio continues to trend favorably on a quarter-over-quarter basis. I'd also like to note that in December, we exercised our option to increase our ownership interest in the Brazilian joint venture from, round numbers, 43% to 49%. The results in Brazil have been very good, and we are pleased with our progress there. In Personal Insurance, fourth quarter results were significantly impacted by Sandy with after-tax cat losses of $370 million for the segment. Excluding the impact of cats and favorable prior year development, the underlying combined ratio improved over 9 points for the quarter and nearly 5 points on a full year basis. However, to really understand the dynamics in the segment results, we need to look at Homeowners and Auto separately, because they are somewhat different stories. In Auto, As Jay mentioned in his opening comments, we are not yet satisfied with our performance but believe we're on the path to improving our results. We continue to be very pleased with pricing gains, with renewal premium change of 9%, up about 1 point from the third quarter and up 5 points from the fourth quarter of 2011. Retention remained solid at 81% but as anticipated, our new business volumes have been impacted by our pricing actions. These pricing actions are in response to the severity challenges across all coverages that we began speaking to you about in the fourth quarter of 2011. The physical damage severity pressures we saw in 2011 persisted through the first half of 2012 but have since returned to near-normal levels. The bodily injury severity trends, however, have continued at elevated levels throughout the year. In addition to expected general inflation, we believe this increase in bodily injury trend is a result of more severe accidents. There are a number of environmental factors that can contribute to this trend, for example, data recently released by the National Highway Traffic Safety Administration shows road deaths for the first 9 months of 2012 increased 7% year-over-year, the largest such increase since 1975. So over the last 2 quarters, severity for all coverages is running just over 5%. With mix adjusted frequency about flat, overall loss trend is in the 5% range. Given the renewal premium change we've been able to achieve over the last 4 quarters, earned rate is now more than offsetting the aggregate loss trend. This margin expansion is somewhat difficult to see in our reported results due to prior period adjustments. On the Agency Auto combined ratio exhibit on Page 16, you can see these adjustments and the resulting adjusted underlying combined -- GAAP combined ratio, which improved about 1 point for both the quarter and full year. So on a go-forward basis, as you heard Jay say, at our current level of rate gains, and assuming loss trend does not increase materially from current levels, this will result in improved underwriting profitability in the line. Turning to Home, pricing was also very strong. Renewal premium change coming in at 13% was up 1 point from the third quarter and up 5 points from the fourth quarter of last year. Retention continued to be strong at 84%, while new business volume was lower than recent quarters due to the execution of our pricing strategy, higher deductibles and other profitability initiatives. Given these initiatives, our core underwriting margins are seeing significant improvement. However, it is difficult to evaluate this line's performance, excluding catastrophes, especially given the last 2 years experience. Accordingly, with the ongoing volatility of weather patterns, we will continue to seek improved underwriting and policy pricing terms and conditions. We feel good about the progress we've made in the Personal Insurance segment, but challenges remain. We will continue to locally execute rate increases, underwriting changes and modifications to terms and conditions to further improve the risk profile and profitability of this business. So in summary, looking across all the business segments, let me reiterate what you've heard from us for the last several quarters. We are encouraged that we've been able to successfully execute a strategy that has resulted in tangible improvements in our underlying underwriting margins. We believe that these pricing and underwriting actions are appropriate, given the current level of investment returns, weather volatility and general loss trend, and we are well positioned to continue to thoughtfully execute on our strategy in 2013. With that, let me turn it over to Gabby.

Great. We will now begin the question-and-answer period. [Operator Instructions] Andre, we're ready to begin.

[Operator Instructions] Our first question comes from the line of Randy Binner with FBR. Randy Binner - FBR Capital Markets & Co., Research Division: I have a question on the Sandy loss. I'm just trying to get a sense of where you might be in the kind of the finality of your loss estimates. I'm particularly interested in kind of cost inflation on reconstruction, if there's a tail from business interruption? And if you can quantify closed versus open claims at this point. Brian W. MacLean: So one is -- this is Brian. Obviously, with this magnitude of event, there are always claims and issues that are still outstanding, but broadly speaking, there's no significant issue that we're looking at from the business interruption or liability or whatever side that leaves us feeling that there's a big tail. I'd turn to Doreen on the specifics with the closed claims. If you've got some of that data, Doreen, and want to share it.

I do, Brian. This is Doreen Spadorcia. Jay quoted you some statistics as to where we were 30 days from thinking -- reported from first notice of loss. At this point, since we're further on than 30 days, probably on the property side, we're at about 95% of our losses in that position. In terms of tail issues, on Auto, we still have salvage losses to collect and we're in the middle of doing that, but those we've factored in. And on Boat & Yacht that those losses tend to take a little bit longer as well, but again, all of them have been factored into our analysis. So we're pretty well through most of those claims that were reported. Jay S. Fishman: And Doreen, my comments were to Personal Insurance; your comments were to all lines.

Yes. Absolutely. Thank you, Jay. Randy Binner - FBR Capital Markets & Co., Research Division: So just to clarify, you're saying you're 95% closed on those property claims?

Yes. Randy Binner - FBR Capital Markets & Co., Research Division: Wow. And then just on, would -- just one on demand surge, a lot of reconstruction going on in New Jersey. Is that becoming an issue, like it has with past events, or is that under control?

I would say, at this point, our estimates of what we saw when we initially evaluated the property, our holdings, we always have situations, Randy, when someone has additional damage and we look at those. And that's kind of a normal course. But at this point, nothing out of the ordinary.

Our next question comes from the line of Jay Cohen with Bank of America Merrill Lynch. Jay Adam Cohen - BofA Merrill Lynch, Research Division: Yes, just 2 questions. The first is in the Personal Auto side, the PIF count shrank, a bit more than it had been. It looks like renewal is relatively stable, new business is down. I'm wondering if you could just give us some color around that PIF growth because that was down a bit more than it had been, and more than we had expected. Jay S. Fishman: Jay, this is Jay Fishman. I think you're analyzing it correctly. The PIF is being impacted by our pricing increases. We recognize that we're -- in some areas that there are less expensive alternatives that are available to agents and in some cases, they're embracing it. Our pricing strategy is based on what our loss trends look like and the severity dynamic is real, and it is substantive and doesn't appear -- we may be wrong about this, but it doesn't appear to be a Travelers-only phenomena. This seems to be broad-based. So we're taking the actions that we know we have to, to produce appropriate returns over time. And if the policies in force shrink somewhat, that's fine, and we will just continue to price it as we know have to, to achieve the returns that we speak about. Jay Adam Cohen - BofA Merrill Lynch, Research Division: Makes sense. Second question, in the past, you have given some discussion or guidance around capital management kind of forward-looking. As you look at 2013, can you talk about your buyback expectations relative to your earnings? Jay S. Benet: Jay, this is Jay Benet. I'd go back to the wording that we had in the Ks and the Qs. Nothing's really changed. We have earnings that we're going to generating. From those earnings, we need to have capital to support whatever business growth we have. There may be some needs like pension contributions into our qualified plan, whatever. We lay those out. But ultimately, the idea behind our capital management strategy is to return all the excess capital that we have to the shareholders, and that'll go back to shareholders in the form of dividends, as well as common stock repurchases. And it's really going to be driven by the level of earnings, ultimately. Jay S. Fishman: Let me -- it's Jay Fishman. Let me just come back with one sentence on the PIF count. You did see it, but I just want to make sure it's clear. Retention remains stable in Auto. We are not losing existing accounts. What we have done is experience a lower level of new business than historical trends, but retention importantly, is remaining stable.

Our next question comes from the line of Brian Meredith with UBS. Brian Meredith - UBS Investment Bank, Research Division: I have 2 questions here for you. The first one, I was hoping, Jay and Brian, maybe you could comment about where we are with respect to rate adequacy in the Business Insurance area. If I look at it on an at or near basis, it looks like you're probably kind of low to mid-teens ROEs on that business right now, is that correct? Jay S. Fishman: Well, we -- Brian, it's Jay Fishman. We haven't disclosed what our product line returns are. I think you can look at the overall business and really make 2 adjustments that are available to you, and you'll come up with what, I think, a reasonable way to think about return on equity is. One is, you can easily eliminate the favorable development. And two, I think you can adjust the catastrophe costs back to what a more normal run rate would look like. Those 2 are not identical in the absolute level, but they're actually pretty close. They just about cancel each other out. And if you do that, you'll get to a business ROE overall that's round numbers. And I'm not being precise here because I don't have it in my hand, but round numbers, it's about 10%. Now in terms of -- 10%, to 11%, let's say, and people are looking at me. And the real question that you're posing that's so difficult to answer is rate adequacy in the environment where the risk-free rate is 1.5%. We've talked about -- in previous quarters, we've talked about maintaining our aspirational goal of mid-teens return on equity over time. But we've said that under the current environment, it's simply not achievable, and it remains not achievable. And I think most of us who are responsible for managing complex financial businesses are wrestling with if your cost of equity at the moment is sitting in the 7% to 8% range, what is rate adequacy, what is return adequacy in this environment? This is unprecedented, and I spend -- we spend a fair amount of our time thinking about what level of returns is really appropriate. Because I think one of the risks in managing any business, and I think we saw it in 2008, is when management decides that it's going to try and get blood from a stone, it just doesn't happen. But if you look at it overall, we're sort of in that 10% to 11% range at the moment, and a bunch of factors coming at us. But I know you have a longer question, but that's sort of where we sit at the moment. Brian Meredith - UBS Investment Bank, Research Division: Okay, great. And then, the second one is a little more detail on the Commercial Auto loss trend situation there. Is it physical damage, is it liability, what's going on with that line? Brian W. MacLean: Within Commercial Auto, it's not dissimilar to the conversation we just had on Personal Auto. So it's fundamentally severity-driven. And at this point, the bigger severity issue is on the bodily injury side.

Our next question comes from the line of Adam Klauber with William Blair. Adam Klauber - William Blair & Company L.L.C., Research Division: The accident year combined in Business Insurance improved by -- improved materially 400 basis points. Was that all quarter-to-quarter, or was there some catch-up from other quarters this year? Jay S. Benet: Yes, this is Jay Benet. There really aren't any catch-up adjustments relating to earlier quarters in the year. So that's -- with minor variation on that, that's a pure number. Adam Klauber - William Blair & Company L.L.C., Research Division: Okay. That's -- it was a very good number. Could you ballpark that -- and not looking for an exact number, but I mean how much of that 400 basis point improvement is from nonrenewing poor business versus getting rate versus loss trend and existing business? Brian W. MacLean: So this is Brian. Hard to precisely segment that, but I did say in my comments, we're getting about 2 points from pure rate over the loss trend now. And then, there's always other stuff within a quarter that's moving within the loss ratio, so the non-cat weather, other large losses, fire losses, et cetera will be volatile. And some of that was moving to the good. So the biggest driver is the rate in excess of loss trend, which is 3 points? Okay, 3 points in the quarter -- in the quarter, yes, right. So that's the biggest piece of it. Obviously, there's something with the re-underwriting issues in a lasered way that you're talking about, but that is not the broad-based issue moving through the results. Jay S. Benet: And we don't track, because I wouldn't know how, candidly, to track the improvement in profitability that resulted from nonrenewing accounts. That's -- specifically to your question, that's just not something that I -- that we measure or keep track of.

Our next question comes from the line of Meyer Shields with Stifel, Nicolaus. Meyer Shields - Stifel, Nicolaus & Co., Inc., Research Division: If I can go back to Agency Auto really quickly, I think, when we saw the decelerating new business despite, maybe, incremental improvements in pricing, does not mean that the environment is becoming more competitive? Jay S. Fishman: Well, I don't think it's an incremental improvement in pricing. I'd use a different word, and the kind of pricing change that's occurred here is substantial. And at the Agency Auto level, with increasing use of comparative raters at the agency level, pricing is certainly more important, we suspect, than it was, let's say, 5 or 6 years ago. A greater ability, technological change at the agency level for agents to generate multiple quotes relatively quickly and efficiently so that it's -- the comparison process is just easier. So I think that, that's really what's driving it. This is a meaningful change in pricing, driven by a meaningful change in loss trend and weather and interest rates and all that goes with it, and the price is the price, and it'll show up on comparative raters. I don't believe, I may be wrong about this, but I don't believe that the Agency Auto business has become a business where the only thing that matters is price. We do believe that agents have views about companies and their ability to serve individual customers in certain ways, and that's what we've traded on all these years and continue to now. So whatever the new business dynamics are, they are. We have a really strong philosophy here, which is that volume is not a goal, it's a result. If we do all the things that are right, and we're moving in the right direction, the revenues, the premiums will be what they'll be, and then we manage the business accordingly. And that's really what you're seeing in Personal Auto. Meyer Shields - Stifel, Nicolaus & Co., Inc., Research Division: Okay, that's very thorough. The tax rate on the non-net investment income component was a lot lower than, I think, it's been in prior quarters. I was hoping someone can explain that. Jay S. Benet: I'm not sure that's the case. So if you take a look at the tax accruals, if you were to just separate investment income from the rest of earnings, effectively, what you do is, you take round numbers, 35% of the non-investment income, that's the tax provision or tax benefit, if there's an underwriting loss. And then, as it relates to the investment income, we give you the effective tax rates. So I think if you do the arithmetic, you really come very close to what the tax provisions are.

Our next question comes from the line of Michael Nannizzi with Goldman Sachs. Michael Nannizzi - Goldman Sachs Group Inc., Research Division: So I had one question, just kind of looking over the last couple years, looks like you've got 8 points of rate in Business Insurance over the last 12 months or so and then, another almost 7 points the year before that. So 15 points in the aggregate of a couple years, looks like combined ratio is a bit better. Just trying to understand, does that mean that more rate -- we should expect more rate to earn through over the next 12 months? How much of that dynamic is due to mix changes, the Commercial Auto stuff that Jay mentioned upfront, or is there something else? And I just have one follow-up. Jay S. Fishman: I think we'll answer this in a couple of pieces. First, obviously, written rate becomes earned rate, so yes, what we've written in these last several quarters will convert to earned over the next several quarters, and that's just arithmetic. The -- our intent here is to continue. This was always a granular, active pricing strategy. We've commented before we're not big believers here in sort of mythical cycles that carry the industry up and then carry it down. We think that it is one agent, one underwriter and one account at a time. So this strategy here has been extremely active and very granular and applied to individual accounts, putting rate at accounts that need it, taking all of the factors into account: taking interest rates, taking the perspective of changing weather and taking some of the loss trend that affect some of the lines into account. So that's going to be the -- that's going to be our prospective policy, and we're going to continue to do that. I don't know if you have any... Brian W. MacLean: Well, specifically on the mix issue, it's not fundamentally a mix issue. We've talked for a number of quarters now that, although Commercial Auto and comp are at the higher end of the rate change grid for us, it's really pretty broad based, and we're getting between 6 and 10 points in all lines. And we haven't really shifted our line mix significantly at all. So that's not driving it. Jay S. Fishman: We shared with you in the last quarter, in the previous quarters, the breakdown of the accounts, the segments. The one we show you are the ones that track profitability, a long-term profitability, combined ratio by account -- loss ratio by account, actually. And we said in the last quarter and before that, that our greatest opportunity was in those segments of the accounts that were at the low end of the profitability spectrum. We still believe that to be true, that the greatest opportunity for us in terms of improving profitability is at that less-profitable end of the segment, which again speaks to the granular nature of the strategy that we're applying. Michael Nannizzi - Goldman Sachs Group Inc., Research Division: Great. And I just wanted to revisit maybe direct. I mean, we've talked about it maybe a little while back. How are you thinking about that? It looks like the first time that PIF started to fall a bit, maybe ratio. I don't know if that's advertising came down a bit in the quarter on a sequential basis. How are you thinking about that? And what is your outlook for when that business needs to rightsize in order for that to be an area of continued investment? Jay S. Fishman: No, it's a pleasure. In the direct business, we've been involved for the last, I don't know, several quarters, 2, maybe 3, something like that, in a very granular test of elasticity pricing strategy applied in, I -- 2 or 3 states, I -- where we really didn't have a robust agency business. And we were able to change the pricing in those states, both agency and direct, to begin to test the elasticity of demand in the direct arena. That's one of the reasons why the advertising spend is down, because it's being applied in a very narrow way in these last several months, just trying to drive activity in these 2 states, I'm told. And the results have been extremely informative and encouraging in that we're getting increasingly close or closer to having a pricing schedule, a pricing strategy, that we think can work in the direct arena and can be applied both at reasonable marketing success, we believe, as well as an underwriting success. But we remain some distance away yet from what I would call launching in a broad way, but we continue to make real progress and are encouraged by the results. There's nothing that causes us to think that we should discontinue the spend on it. Quite the contrary, we feel really encouraged about what we're learning and discovering. And as I've always said, the thing that we could do wrong here is have a marketing success and a financial failure. And that would not be acceptable here, so we're going to continue to take our time and make sure that when we aggressively approach the business, that we do so with all the requisite knowledge and learning and skills and with a high degree of confidence in the success of the program.

And this line is -- Mr. Zaremski, your line is open. Michael Zaremski - Crédit Suisse AG, Research Division: Okay, great. I was hoping to better understand the underlying margin improvement drivers within Home. So other than pricing, is there quantifiable elements coming from geographic culling and/or terms of condition changes that you guys have been talking about for a while? Brian W. MacLean: Well, first of all, the biggest driver is pricing. There's no question about that. And... Jay S. Benet: Well, and weather. Yes, and there's been a number of things. In the fourth quarter of last year, we had an unusually high concentration of fire-related losses. This year, we had a low level of fire-related losses, relative to some norms. In addition to that, there was a little more weather in the fourth quarter of last year than this year. So Brian spoke before about just things that happened on a quarter-to-quarter basis that caused some variations but those along with, the major item being loss cost trend x weather, which is a hard thing to really talk about as it relates to Homeowners being lower than earned rate. And if you look at those 3 components, the lower fire losses, the lower weather-related losses, as well as earned rate over loss trend, then that's really what's making up the difference. Jay S. Fishman: And to answer your question directly: We really don't have a metric to convey to you the impact of, let's say, increasing deductibles on the profitability of the business. We would have to go back and rerun the losses using the deductibles that existed previously, and that's just not possible for us to do. Michael Zaremski - Crédit Suisse AG, Research Division: Okay, that's helpful. And then lastly, as a follow-up to the questions about the capital management for 2013. So I see you've contributed, I think, over a couple hundred million dollars to the pension this year and you have $500 million of debt coming due in a couple months. In regards to the pension, if interest rates stay low, should we expect a similar contribution? And are you guys considering refi-ing with that? Jay S. Benet: Let's talk about the pension first. If you strip away all that takes place in valuing the -- a PBO versus the assets, there's a normal level of earned credit, if you will, that runs through the numbers. And it's about $100 million, round numbers, a year. So if everything worked in accordance with the assumptions, in a steady state, the pension contribution would be about that. The reason the pension contribution is higher this year is things don't work as a steady state. We had a good return on the assets. On the other hand, when you look at the valuation of the pension benefit obligation as of 12/31, which is a spot rate interest concept that you use, you had very, very low interest rates at that point in time. So we looked at that and said, "Gee, on balance, it feels like $200 million is the right number to put in this year." It got us to a funding level that was high, approximately 90%, which is where we started the year. If rates spike up, we're not going to be terribly overfunded. We'll just adjust future funding accordingly. And that would be a nice position to be in, frankly, if rates went up a bit. But $100 million is what we think about as more of a normal level on a year-by-year basis. As it relates to the debt, as we generally do, we try to maintain full flexibility in dealing with our capital position. So we ended the year, as I said earlier, with over $2 billion of holding company liquidity, which allows us to be in a position of paying the debt when the debt's due to avoid any kind of liquidity concerns. And then we'll just continue to evaluate what the proper structure of our balance sheet is going forward based on interest rates and based on how we feel. But certainly, in terms of capacity, we have the capacity to keep the debt ratio just where it is today if we so choose.

Our next question comes from the line of Amit Kumar with Macquarie. Amit Kumar - Macquarie Research: My first question is on your catastrophe reinsurance coverage, which I know expires on June 30. And maybe this is a bit too early: The -- when you new look towards 2013, 2014, do you think your cat cover will look similar? Or has the thinking changed post Sandy? Jay S. Fishman: The thinking actually changed a little bit a couple of years ago when we reintroduced -- I'm sorry, when we introduced an aggregate -- cumulative aggregate element to our reinsurance program. And we started that, I think, last year at the July coverage and actually expanded it some effective January 1. We actually increased this cumulative aggregate program, which has the effect of giving us a benefit against the retention in the cat program should the individual events exceed retentions and then cumulatively apply against the overall cat program. With respect to the cat program itself, I wouldn't anticipate any significant changes to that. Our property coverage exposure really has not changed. We were comfortable with where we were last year. It may depend upon pricing move a little bit up or a little down, but the important change, the more important change, is the introduction of this aggregate program to our overall cat merits. Jay S. Benet: Yes, the one thing I would add is, when we did Long Point Re II, we had pegged it to RMS. And when RMS 11 came out, it raised the trigger points associated with that. And in this [ph] past year, we issued a $250 million cat bond with Long Point Re III. And as Long Point Re II turns itself over, we will probably bring those limits -- bring those trigger points down into something that looks more like Long Point Re III. Amit Kumar - Macquarie Research: Got it. That's very helpful. The only other question I have is on Business Insurance pricing and maybe using Slide 9 as a backdrop, and perhaps this question is for Jay. The -- when you talk to your largest agent or business, post Sandy, is there a greater acceptance on pricing? Or do you think there was an uptick in December, and just based on the economic conditions, it's already beginning to sort of normalize as you look towards Q1? Jay S. Fishman: I think it's -- well, I think the environment, particularly for the January 1 renewals -- and this is anecdotal, what I'm giving you, because I don't -- we don't have the data yet, but the anecdotal observations for January in renewals and the related rate was very similar in the aggregate to what December looked and felt like. So I don't -- I wouldn't anticipate any immediate change of consequence in January. I -- Sandy is a tough one to really talk about pricing. I think that it continues to reinforce the notion that weather patterns may be changing, and we've been saying that now for a couple of years and embracing that in our pricing strategy. And I suspect -- again, no evidence on this, but I suspect that the storm will simply make that argument easier to make, both at the agent level as well as the insured level. I would tell you this, notwithstanding how -- sort of the perceptions of Sandy, it was in the playbook. The storm itself, the track, the path, it wasn't something that -- as we do our analysis and modeling, that we didn't contemplate. It was certainly there. I think the surprise, to the extent there was one, in Sandy was, to some extent, the timing of it. Meaning, it was full moon, it was 2 tides. There were a lot of ancillary factors that made it more significant than it might otherwise have been. But importantly, the flooding was more severe. And this is really -- it's predominantly a -- I'd say it's largely a Homeowners kind of question with respect to the NFIP program, but it does have its impact on commercial insurance as well. There were buildings that we insured in Lower Manhattan that we would not had contemplated would have been part of a flooding event. The flood coverage that we extended was really episodic for them, meaning we contemplated broken pipes or backup or things of that nature, not a flood coming out of the East River or the Hudson River. And importantly, the dynamic there is the sea level. And this is just from things I read, but sea level is, depending upon who you ask, anywhere from 6 to 12 inches higher in Manhattan than it was about a century ago, driven largely by the melting of the polar ice caps and what that's doing to the water level. So I think what Sandy will do is cause people to think somewhat differently about concentrated flood risk than they had before. It'll take some time, it's a complicated issue. But all of us, I suspect, we certainly are, looking at flood concentrations in the commercial segment and analyzing it with a little different perspective than we did before. Those were -- that was, I think, the big takeaway from Sandy that was different from what we might otherwise have contemplated.

Our next question comes from the line of Vinay Misquith with Evercore. Vinay Misquith - Evercore Partners Inc., Research Division: This is a follow-up to Brian Meredith's question before on sort of rate increases in Business Insurance and the ROE, because the ROE right now seems to certainly be pretty healthy [ph], in fact close to about 11% x the OCI. So just first, curious on the Business Insurance front, how do you see rate increases this year versus last year? Do you see them tapering off a little bit? Or are you pushing for the same magnitude of rate increases? Jay S. Fishman: Well, it's -- as I've said before, it's a very granular approach. We have accounts that are at pricing levels that we're satisfied with, that we feel good about, and we have accounts where that's not the case. And that may be driven by the loss experience. It may be driven by the mix and long tail versus short tail, and the impact of interest rates. And it may be impacted also by geographic concentrations relative to catastrophe management. We look at this very granularly. In some businesses, as we've told you before, we break accounts down into quintiles. In some business, we actually break them so far as deciles. And our pricing strategy is driven by the account and its sports class and its performance and where it fits. So we'll let the -- we will continue to do what we do by looking at individual accounts. We have no artificial limits. We look at an account and we try and understand its profitability and how it contributes. And if it's not contributing in what we think is a substantive way, we have a discussion with the agent. So that strategy is no different than it was a year ago. And we'll continue to do it, and the number will be what the number is. And I do -- I would acknowledge to you that a very important question in all this, and I don't have an answer for you but it's one that I think about a great deal is, again using a 1.5% risk-free rate, what level of overall returns are achievable and sustainable without jeopardizing the risk profile of the organization. That is the question. And we're not there yet, but we're -- so we're going to continue to push. Vinay Misquith - Evercore Partners Inc., Research Division: So fair enough. And you mentioned that the rate increase is, I think, 8.3% in December, so tipped towards the higher end. Could you give us some color on what the rate increases were in January of this year? Brian W. MacLean: Yes. So when the -- so this is Brian. In the middle-market businesses where we have a little bit better transparency into that, the January ones, the early January that we've seen the middle market look a lot like December, so that's encouraging. Small commercial, it's harder to state, and so we'll get a better look at that at the end of the quarter. But the little peek that we've got is encouraging, but that's -- it's a little window. Vinay Misquith - Evercore Partners Inc., Research Division: Okay, that helpful. And just one follow-up. From the Commercial Auto, there were some adverse physical [ph] development. If you could please provide some color on sort of the subcomponents of that -- those small fleet versus big fleet, along with trucking, and whether that was more of a severity issue or a frequency issue. Brian W. MacLean: So it's a -- just like we said in personal auto, it's a bodily injury severity issue. And it's really across the book of business. So it -- we don't think -- the Commercial Auto book of business, right. We don't think of it as a really lasered underwriting as in a problem in one business, but we're seeing it kind of systemically in the marketplace.

Our next question comes from the line of Greg Locraft with Morgan Stanley. Gregory Locraft - Morgan Stanley, Research Division: Just wanted to -- in your prepared remarks, you mentioned Schedule P and sort of gave us a sneak peek. I just wanted to confirm the -- so the Workers' Comp line has inflected. So I seem to recall that the 2010 policy year developed adversely when we look at the '11 data. And it sounds like, from your remarks, that '11 policy year is going to have developed favorably when we look at the '12 Schedule P. Is that true for -- and this is for Workers' Comp, specifically? Jay S. Benet: Yes, well, for Workers' Comp overall last year, we did see, as you talked about for the last -- yes, I should get into the years because we're in 2013. Well, if you look -- if you go back a year and you're in the end of -- the beginning of 2012 looking backwards, the 2010 accident year did develop unfavorably at that time for Workers' Comp. If we now roll the clock forward, there hasn't been any further developments associated with the 2010 accident year in 2012. And then if we look at the 2011 accident year for Workers' Comp, a year later in 2012, we've seen some slightly favorable development for that. And overall, in 2012, as I said, our Workers' Comp has developed favorably. So that's the Workers' Comp story. And I think it speaks to what we were communicating a year ago. Given the nature of our processes, the granularity of the data that we have and the quarterly diligence we have, we like to believe that when we -- we see trends early, we react to those trends and we book them and move on. Gregory Locraft - Morgan Stanley, Research Division: Okay, good. And I mean, that's a great inflection. Do you -- this may not be fair, but do you think the industry is going to see the same or, as, I guess, your remarks just suggested, are you way ahead of it? Jay S. Benet: I hope that -- well, we certainly can't speak for the industry. I mean, everybody has got their data, their views, their reserving practices. And I think you just have to look at what companies will be doing. But in terms of our data and our analysis and being able to stay on top of things, we think the numbers speak for themselves. Gregory Locraft - Morgan Stanley, Research Division: Okay. And last on Comp is just I haven't heard a good explanation: Why did frequency spike up? I think it was a frequency issue, at least at the industry level. With the benefit, in some ways, of hindsight, what is the reasoning behind this, the higher loss trend that came through that was somewhat unexpected that, obviously, now you guys have gotten in front of? Brian W. MacLean: Yes, so we've talked about it a good bit, I don't know, a year-plus ago. But as you hit the economic downturn, clearly, losses mitigated -- frequency mitigated as it usually does. People are concerned with maybe losing their jobs so they don't go out on a Comp claim. As employers lay off workers, they lay off the least experienced first, et cetera, et cetera. And then you would typically see a rebound from that as there's a -- there's kind of a pent-up delay in loss reporting. And so what we saw clearly in our results, we can't speak to the rest of the industry, was that, that spiked up a little harder than we thought it was going to. And so we had a backlog, and so delayed reporting, from the previous years, and that kind of came through. So our frequency picked up. Again... Jay S. Fishman: It was 2010. It was in 2010 that we saw a spike in what we have come to call late-reported claims, accidents occurring prior but reported in 2010. Brian W. MacLean: Right, right. So these are people who maybe an incident in '09 and tried to work through it and by the time 2010 comes along -- and again, we expected some of that. It was just greater magnitude than what we had originally expected. And we've seen it mitigate. Jay S. Benet: And this is Jay Benet again. I just want to -- I'm being reminded by one of my colleagues, the word "preliminary" is a very critical element of my discussion of schedule fee. It is early. We still are putting it together, but directionally, we feel that this is -- what I said before, is where we're going to end up.

Our next question comes from the line of Jay Gelb with Barclays. Jay Gelb - Barclays Capital, Research Division: I had just a couple of follow-ups. The first is on investment income. Can you talk about the seasonality of that? It seems to be stronger in 4Q than in 1Q, and I'm just thinking if that trend will repeat in 2013, 4Q '12 versus 1Q '13. Brian W. MacLean: Well, in the fourth quarter of '12, a lot of private equity funds were trying to sell companies because of an uncertain tax filing in 2013. That may be a one-year occurrence... Jay Gelb - Barclays Capital, Research Division: I was referring just to the fixed income aspect. Brian W. MacLean: I'm not sure there really is seasonality there. I mean, the -- there's $1 million difference where someone seized on. There are 3 elements to that. There's volume of the portfolio, there's rate and then there's calendar days in the quarter. And the net of those accounts for -- generally accounts for any small difference quarter-to-quarter. But net investment income from fixed income comes in pretty much like the tide. Jay Gelb - Barclays Capital, Research Division: Well, I was looking at fourth quarter '11 stronger than first quarter of '12. And I'm just thinking, well, should we not use that $500 million of after-tax fixed income from 4Q '12 as the run rate for the beginning of '13? Jay S. Benet: Well, what's taking place in fixed income is there's a constant maturity of certain securities in the portfolio, and they're being reinvested at today's rates. And there's been several analyses we've put out into the marketplace, which I'd refer you to, that talk about some estimates of the maturities in the portfolio, how much the yield is on that block that's maturing. And then you can pick your number whether they're going to be reinvested at 200 basis points; 225; 175, less. What you're generally seeing is a deterioration in the overall yield in the portfolio. It's not dramatic, but it's real. And the numbers that you see in the webcast, in the fixed income slide, are reflective of that in terms of, as you know, quarter-to-quarter comparisons. Jay S. Fishman: But there's no seasonality in the fixed income result at all. Brian W. MacLean: We track book yields quarter to quarter and year to year. And over the 12 months of 2012, book yield of the portfolio declined by 33 basis points. And that's not unusual. But the decline is inexorable until we reach the point at which reinvestment rates -- in which all the securities with high book yields have run off and they're replaced, and that's 2 or 3 years off. Jay Gelb - Barclays Capital, Research Division: Okay. And then my separate question is on Personal Auto. To get you to your targeted returns, what type of combined ratio do you feel you need to be able to generate? Brian W. MacLean: I would say, about a 97, between 96, 97 combined.

Our final question comes from the line of Matthew Heimermann with JPMorgan. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: Two questions. First was just -- is -- you've noted non-cat weather this quarter, and that's been a theme all year. I guess I was just trying to get a handle, is there a number you could give us to kind of quantify how much of a tailwind that was this year? And I recognize that we're comparing 2012 to what was a slightly worse-than-normal year in 2011. Jay S. Benet: I don't know that we can give you a number. I think we try to do it the other way around. Brian's talked about what the impact of earned rate versus loss trend is. And these other things, whether it's non-cat weather or, as I said earlier, we had fire losses in one quarter higher than another quarter, or there's large loss activity. But we try to do it the other way and talk about what's underlying it in terms of rate versus "loss trend." Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: Is there a way to give us a sense of whether or not it's -- I would assume -- tell me if this is right, I would assume it's just more a personal lines benefit that it is a commercial lines benefit. Is that perception correct? Jay S. Benet: No, it's meaningful in both, depending upon the quarter. Jay S. Fishman: It's not unusual. And I -- we're not, not answering because we don't answer. We're not answering because you're asking a question that's exceptionally difficult for us to actually quantify. It's how much of a -- you're sort of asking, within our normal loss estimates, how much are we relying on of weather, and that's a tough question for us to answer. I -- we do note, though, that it is not unusual in a high-cat year to have lower, what we call, non-cat weather. And I -- and we suspect, can't prove, but we suspect what happens is that some of the claims that would ordinarily show up as non-cat weather get included into the catastrophe estimates, and those customers who had some loss previously now get rolled into here. I -- and it's an anecdotal observation, but it does appear to have some actual fact in the numbers. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: No, that's fair. I guess the fact that you know it's happening, though, makes it a little surprising to me that we can't get a ballpark number but -- because -- you know what I mean, because that implies there's some recognition of the trend somewhere but... Brian W. MacLean: But for example, in BI, as I said before, about a 4-point improvement in the combined, roughly 3 points of that is the core price in loss trend. So it's about 1 point of other stuff, a decent chunk of which would be what we're talking about, so -- but it's not... Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: Yes. I was -- yes, can you give -- but do you have that handy for the full year? I mean...

It's not out. We'll try and take some of this off-line because I know we've talked about particularly what affected 2011 for the full year, and you can see some of that in the run rate. So I'll get you off-line. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: Yes, no, that's fair. I just don't want to over- or underestimate the core. I guess the other thing was, should we think about the normalizing loss trend we're seeing in -- or what you characterize, I think, as a normalization of the trends in Work Comp as knocking on the -- to kind of the implied rate need in 2012? I know that's been an area where you've been getting some of the bigger headline gains, so I just wonder -- as we're all really focused on what the headline rate number is, but I'm just wondering if that's kind of a very -- a transparent example of where a headline rate might come in, but when you think about what you're getting relative to loss trend, might not affect that spread whatsoever. Jay S. Fishman: Well, when -- we're looking at each other. First, I'd make -- we -- in Workers' Comp, most certainly, the rate gains are exceeding loss trend in Workers' Comp, resulting in improved profitability and improved returns. So while the loss -- or the projected losses in Comp are substantive because we -- it's long tail and we're dealing with medical and wage inflation, so there's a real number that we use. We -- it's not a number that we disclose. I think it is proprietary and a pricing strategy element. But we do assume a real loss level in Workers' Comp that contemplates changes in those -- in both wage and medical. But at this point, the rate gains are exceeding it. Now I would remind you that Workers' Comp as a return -- thinking about Workers' Comp in a return manner, that it's real long tail, and interest rates really matter there. So it's one of those lines where the interest rate, long-term interest rate, impact is substantive relative to the returns. So that may be as much of the driver as the loss trend dynamic in that line.

Yes. And Matt, just to be precise, the only thing we talked about moderating was the frequency dynamic that we've seen in 2011. We didn't comment on the other. Jay S. Fishman: We continue to assume a real number, both in severity, both related to wage and indemnity. It is just a frequency, to Gabby's point, that is -- that we've seen return back to the normal trend line.

This does conclude the Q&A session for the call. I would like to turn the conference back over to Ms. Nawi.

Thank you very much. Thank you for joining us today. And as always, if you have any follow-up questions, you can call myself or Andrew Hersom in the investor relations department. Thank you very much, 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.