Good morning, ladies and gentlemen, and welcome to the Second Quarter Results Teleconference for Travelers. [Operator Instructions] As a reminder, this conference is being recorded on Thursday, July 19, 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, Andre. Good morning, and welcome to Travelers' discussion of our second 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. Given the active weather pattern we saw in the second quarter, we're pleased to report solid performance driven by continued strong underwriting results and investment income. In the quarter, we produced net income of $499 million or $1.26 per diluted share, including $357 million of after-tax catastrophe losses or $0.90 per diluted share. Year-to-date return on equity is now at 10.5%, and operating return on equity is nearly 12%. The weather losses in the quarter, though much improved from last year's extraordinary $1.1 billion level, are still much higher than historical experience would suggest. It was not too many years ago that $357 million in after-tax cat losses is what we would have expected for a full year. Today, weather events and their impact on local communities and our policy holders seem to be a constant news headline. In this quarter alone, there were tornadoes across the Midwest, wildfires in Colorado, there were hailstorms in Texas and most recently, violent thunder and windstorms across the Midwest and Mid-Atlantic. Brian will discuss our cat losses with you in more detail. But once again, I want to extend my personal and professional thanks to our 13,000-person-strong claim organization who continue to be called upon to do extraordinary work in difficult circumstances and they continue to make all of us very proud. I just want to say thanks. The continued high level of weather losses, along with the continuing decline in interest rates, validates our strategy to drive improved rate, terms and conditions and therefore lift underwriting results. That strategy is on course and is now very much visible in our earnings. Net income, excluding catastrophe losses, net favorable prior-year development, improved an impressive 16% from the prior-year quarter, driven…

Thanks, Jay. Looking beyond cat losses, I would characterize our second quarter results as strong. As Jay already indicated, underlying underwriting results showed margin improvement in each segment, including expanding margins in Business Insurance, and we continue to see our reserves develop favorably. As for investment results, non-fixed income investment returns were strong this quarter, driven by private equities in real estate, while fixed income investment returns were in line with our expectations, given the low interest rate environment. Other items to note: First, we reduced our debt-to-capital ratio in the quarter as we used holding company cash to repay $250 million of maturing debt. The debt-to-total capital ratio now stands at 22.3%, down from 23.1% at the beginning of the quarter. Second, we modified our cat coverage by modestly reducing our General Cat Treaty and increasing our Northeast coverage. Effective July 1, maximum recovery under our Gen Cat Treaty was reduced to $400 million from $525 million and the attachment point was increased to $1.5 billion. Also effective July 1, we renewed our $600 million Northeast Gen Cat Treaty with the same $2.25 billion attachment point as in the prior-year. And finally, effective June 6, we entered into a 3-year reinsurance agreement with Long Point Re III, a newly formed entity that issued $250 million of cat bonds, providing us with Northeast hurricane coverage on specified lines of business, which is subject to a $2 billion retention. After which, we can recover 50% of covered losses up to $250 million. All of this was accomplished at a cost that was effectively the same as in the prior year. For your convenience, we've shown the structure of our cat coverage on Page 21 of the webcast and a complete description is included on our second quarter 10-Q, which we filed earlier today. And finally, all of our financial strength indicators remain in excellent shape. Given seasonal cash flow patterns and recent cat losses, second quarter operating cash flows remain strong at $451 million, a holding company liquidity of $1.98 billion and all of our capital ratios were at or better than their targeted levels at the end of the quarter. Net unrealized investment gains, which ended the quarter at $4.57 billion before tax, increased by over $200 million in the quarter, and book value per share rose to $64.90, 4% higher than the beginning of the year and 9% higher than a year ago. And as always, we remain fully committed to identifying and returning excess capital to our shareholders. During the quarter, we returned a little over $530 million to our shareholders, with common stock repurchases of $350 million and dividends of $181 million. Brian is now going to talk about our underwriting results.

Thanks, Jay. Before I get into the segment results, I'd like to take a minute to discuss the impact of catastrophes in the quarter. In total, there were 13 industry cats, 7 of which were significant for us. While our cat losses this quarter were much lower than the unprecedented levels of 2011, they were considerably higher than what we would've expected based on long-term historical experience. Although thankfully, we didn't have the headline-grabbing events of last year, to put all this into perspective the application of state market shares to our catastrophe losses and estimates reported by competitors would imply an industry loss over $10 billion. At this level, but for last year's results, this would have been the worst second quarter for weather losses in history. Particularly in light of the significant and unpredictable impact of weather, we remain very pleased with the performance of the franchise, particularly the underlying margin improvement that we are seeing across our businesses. Beginning with Business Insurance. Operating income excluding cats in prior year development is $488 million, up more than 9% from the prior-year quarter. On the same basis, the combined ratio improved nearly 3 points year-over-year, driven largely by earned rate increases exceeding loss cost trends, along with fewer large losses. Net written premiums were up 5% year-over-year, driven by strong pricing gains, as well as by growth in exposures and audit premiums. Retention remained solid and was in line with the first quarter at 79%, while new business of $451 million was down from the prior-year quarter but up from the recent quarters. Renewal premium change for the quarter was 9 points, which included pure rate of 7%. You can see on Slide 10 that although this looks like a drop of a full point in rate from last…

Thank you. Andre, we are ready to begin the question-and-answer period, please.

[Operator Instructions] Our first question comes from the line of Mike Zaremski with Credit Suisse. Michael Zaremski - Crédit Suisse AG, Research Division: I've been hearing you and others talk about terms and conditions changes within Homeowners. Can you update us on that initiative? Specifically, how far along in the process are we? And can you detail what the main initiatives are? I believe they're on the deductible side.

Yes. Mike, this is Brian. Let me start. Yes, clearly, the main initiative is the deductibles. We've been pushing this -- I would say, we started about 12 months ago. We've been pushing more dramatically in the last 6 months. We've got just about 40 states now where we are pushing higher deductibles than we have historically had. And that is the main initiative. In some places -- in most places, it is heavily on the new business side of the equation. So it's not as much broadly across the renewal book although, in the more troubled areas, it's even there. We're pretty encouraged by fairly recent activity it's gotten. And there was actually an article in The Wall Street Journal online the other day about insurance deductibles soaring and talked about kind of a widespread movement to get deductibles up. So that is clearly the biggest area. Also from the underwriting and terms perspective, I think we and some other companies are looking hard about how we cover roof hail losses. And so, there's an underwriting dynamic about being more -- doing a better job of selection on the age of roof, the quality, the type of roof, et cetera. And also looking at how we could potentially be changing the coverage around covering roof losses. Those are probably the biggest areas.

The only thing I'd add which Brian made brief reference to but I think over time will matter is a change in our underwriting process to actually exclude homes that have roofs that are of a particular age compared to -- and it varies based upon the type of roof it is and its structure, but setting up a hard-line standard to the extent that the age of the roof exceeds x number of years or x percentage of its useful life, we won't underwrite the home at all. And I don't know how many states that's in, Brian. I don't know if you know either.

Yes. I don't know. Michael Zaremski - Crédit Suisse AG, Research Division: So I mean, do you -- is this -- 40 states seems like a lot, so should we -- I mean, should we expect -- I mean, in your increasing rates in Homeowners in excess of I think 8%, 9%, 10% plus you're making deductible changes, should we expect a -- is the expectation meaningful improvement down the road?

I'm trying to think of what -- what exactly do you mean by meaningful? Michael Zaremski - Crédit Suisse AG, Research Division: In terms of the margins. It just seems like there's a lot going on there. Just more than just rate.

It's Jay Fishman. I think, 2 things. First, the rate side is not only new, but also renewal. And so the impact of the rate is taking effect in the book right now. What is completely unpredictable is the weather. And the weather losses are such a meaningful part of Homeowners losses that for us to speak about anticipating improvement implicitly implies an assumption about weather. And I would tell you that importantly here, the actions that we're taking on rate, we're doing fine, pleased with the retention in the Homeowners book. To the extent the change in deductibles are predominantly in new business, we anticipated and are seeing a drop-off in our new business as other competitors have not yet become -- have not changed as much as we have. But the article that Brian referenced a moment ago about soaring deductibles was interesting to us. So I'd say that yes, we feel very good about the rate actions and the renewal book, and that's looking pretty solid to us. And we will ultimately see what happens with our new business on the -- driven by the change in deductibles in those states.

But you're right Mike, together, they should have a meaningful impact on margins. The big variable is what the weather's going to be. Michael Zaremski - Crédit Suisse AG, Research Division: Okay. Good, that's very helpful. And then lastly, if I look at Slide 10 on Business Insurance and the renewal rate change on 1Q, 7.5%, 2Q, 7.4%. I believe you guys have a pretty decent line, it's 30, 60 days out on pricing. Do you think you've reached an absolute [ph] Level of rate increase that we should maybe think about rates moving downward?

Yes, so let me -- it's Brian again, Mike. Let me take a shot. And again, I'd start by reemphasizing that 7.5% rate, 9 points of price and 4 points of trend, we're getting some pretty good margin expansion at these levels. But I get your question, which is, where is it going? And I'll start with the caveats we always give. We don't have a magic crystal ball, so we're not going to make any long-term predictions. With that said, a couple of comments. We've talked for a long time that the 2 main things that are driving pricing improvement at this time are the lowering interest rates and the increased weather volatility. Certainly, nothing happened this past quarter that mitigated those 2 factors, so there's a lot of reason why there should be something sustaining it. But the thing that I'd reinforce, is to us it isn't about a magic number. Believe it or not, we do not have an aggregate target for a pure rate number that we are shooting for. We've got about a million accounts every year, so 1/12 of those coming through every quarter. And it really is a very granular -- how we approach it. No broad brush answer. And so what we really, really focus on is the execution underneath, and are we getting the right rate increases on the right accounts. Our goal is not to get higher rate. Our goal is to improve profitability. Now, we understand that in the aggregate, more rate is better than less in that venture. So it's hard for us to pinpoint exactly where. I mentioned in the distribution slide the fact that we had 1 point higher rate in our worst-performing business and 7 points higher retention suggests that there's more opportunity there. If that is in fact the case, we'll be encouraged. But we're going to execute on a very granular level.

I'll add just a couple of comments because I think it is so important. It really speaks to the fundamental strategy. We've got underwriters out there who are looking at all these individual accounts every day. Brian is right, almost a million business accounts over the course of the year. We've got 1 underwriter doing an evaluation of an individual account or an individual class of business and suggesting pricing. We have accounts that produce very attractive returns. We have accounts that produce much less attractive returns. How the mix actually shows up this month -- which accounts are coming due, are they weather exposed, are they less weather exposed, have they been consistent with our loss expectations, have they not been -- is really going to determine what the individual underwriter decides as it relates to that account. Though it's -- our look at 7.5% versus 7.4% is that you can't conclude much of anything from that. And I'm not even sure in the aggregate it's all that relevant. The breadth, and we've shown this before, the breadth of account of pricing changes in our account base is really quite remarkable. It goes from accounts that are getting 10% or more of rate reductions, all the way up to accounts that are getting 10% or more of rate gains. And so, it's really left to the underwriters. What we will say is that we're not hearing anything anecdotally from our field organization that would cause us to believe that continuing to improve profitability is becoming meaningfully more challenging. We ask that all the time. We're just not hearing a message back that says, "We're beginning to find this competitively difficult." The feedback from the field organization is actually pretty encouraging, and it's one of the important reasons that we have confidence that we can continue to move forward.

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

First question for Jay or Jay just on the non-fixed income portfolio. I read in the 10-Q, you expect income in the second half of this year to be more consistent with the second half of last year, which would have a lower after-tax yield than what you guys put up in the first half of this year. Is that driven more by how returns are reported on that asset class? Or more a statement on where those returns should be relative to current interest rates? And I've got a follow-up.

Yes, this is Jay Benet. We don't have a crystal ball to see how the alternative investment portfolio is going to perform, but in looking at what took place during the quarter, we did have a very strong quarter with regard to private equities and real estate in particular. But I'd say in the case of private equities, it was geared more towards the first half of the quarter than the second half of the quarter. I mean, you're seeing, as well as we are, what's taking place in the economy and the -- in the markets. So in looking out at the second half of the year, we were just more or less looking at the performance of the portfolio in the second half of the second quarter and saying, things seem to be moderating a little bit from where they were in the first half of the quarter and where they were in the first quarter. So it's hard to predict, as we said. And we'll see what happens. But we're not forecasting a precipitous drop. We're just saying, take a look at the first -- the second half of last year as some indicator as to where the second half of this year might be.

Okay. And then second question for Brian, and I know you've already alluded to this on Slide 14. It appears you may have some runway here on the above 90% combined ratio business to really drive rate. Maybe another way of asking this question, is there a more optimal spread in retention between your worst-performing business and your middle- or best-performing accounts? Because it seems like it's a pretty tight spread, and you can really drive a lot more rate on that worst-performing, if you could talk to that a little?

Yes. So the thing I would emphasize, like I said in the comments, is -- so the picture we're showing there is a real summarized version of how we do this. And so as we look at it, I would say directionally, what we've seen in the last 90 days is, to exactly that point, more opportunity on the worst-performing business, whether that's combined ratio over 90 or however we measure it in different businesses. And it gets back to what Jay said is driving that. Potentially, there is a more optimal balance in the retentions there. We've got to see how account by account how that all plays out. But we're very focused on improving the profitability, specifically on those accounts, and retaining our best business at the best possible terms, conditions and price.

This is Jay Fishman. In some of our business, we're looking at these numbers on a decile basis, that granular. In some, we're looking at it on a quintile basis, that granular. But when you look at the page that you're looking at, this is only 1 quarter to 1 quarter and sequentially. And I'd hesitate to overreact or over-predict because you just don't know what happened in the field on so many individual accounts. But to see a lift in retention in that 90 and above on a summarized basis with only 1 point of lift in rate suggests that there's -- it'd be a useful bucket for us to focus on over the next quarter and see if we can make better -- effect more better profit improvement actions in that bucket. But it is only 1 sequential quarter to another, and I wouldn't overreact to it or overstate its significance. It's an indicator, an important indicator, but only 1.

And just quick follow-up, that 90% or higher combined ratio, what percent of your business is that currently?

I don't have that exact number, so...

I don't know that we've ever disclosed it either. And it varies, by the way, by month. What comes due? That was the other part of this that is so important. That competitive perspective isn't something that we want to share. But the mix isn't the same quarter-to-quarter, which obviously affects not only the retention but also the rate gains. It keeps moving. But from a competitive standpoint, I just don't think that's something that we'd like to share.

And I would just say because I think you said combined ratio, first of all, it's loss ratio. And you can look at our aggregate loss ratio, obviously, it's a relatively small percentage. Or else our aggregate loss ratio wouldn't be where it is.

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

I guess just staying on that discussion on Slide 14, which is obviously very helpful. If you sort of look forward, and if I understand this correctly, you're saying that the slope of the rate change line will change but the retention percentages, even if they grow, they would not continue to grow at this one-off this quarter what we saw? Is that a fair assumption? I mean that number will not continue to climb meaningfully on a retention basis.

First -- again, Jay Fishman. We're reluctant and not unwilling to, but reluctant to make predictions because it is so granular and it's so account by account. But broad base, you'd look at the loss ratio over 90%, and what we saw was a 1 point increase in rate from the first quarter and a meaningful increase in retention. It may be that we should have a somewhat higher rate and a even lower retention, so I don't think that we're -- in fact I'll say it definitively, we're not predicting anything. We're not predicting a change in the slope nor are we predicting a change in the retention. This is again, only 3 months on a million business accounts a year. What we're trying to do from all the data that we see is interpret behavior and then we sit down with the management with Bill Cunningham and the rest of the Business Insurance team. And we have a serious discussion about the way the strategy is being executed. And what Brian was suggesting -- what Brian said was that we look at that data and you see, we see an opportunity to do it better. What form that takes, what the relative mix of retention versus rate is not -- I don't know, I don't know. We're going to see what accounts come up next quarter. It's not the same accounts. It's a new group of accounts. With a new set of demands and a new set of attributes. But you look at what in that small section in the quarter, we say, "Alright, there's an opportunity there to do it better and therefore improve profitability at a different pace." At a 90% and above loss ratio, it's hard to argue that we're making money, any serious money. I mean, this is loss ratio only. So it may be that allowing more of it to non-renew is a step in the right direction. Remember, even though it's above 90%, it goes all the way out. We have accounts that have loss ratios that are worse than 90%. So you're getting into very granular decisions about what you retain and what you don't and what rate demands you put on the book.

One quick thing, if I got your question, too. Just to make sure that you don't overblow the arithmetic here by thinking that, "Gee, the retention ratio went up a lot in that bucket and it might come back down." That is by far the smallest of the 3 buckets. We are also at the same time focused on the left-hand bucket of hoping to keep a little bit more. We could have retention come down significantly on the 90% and above, go up slightly on the under 60%, and the net of that might be flat retention or even increased retention. So be careful with the overanalyzing the arithmetic.

In its simplest form, not every account in the book returns the same. And ultimately, the analysis is, individual account or class performance, what we've already accomplished from an earned rate -- from a written rate perspective what will become earned prospectively and then what steps do we take on what comes up for renewal next month, next month, next month. We're always looking 30, 60, 90 days out. So it's not the same accounts. It's going to be a new group of accounts. And we will figure out what the best course of action is.

Yes, that's very helpful. The only other question I had is on, and maybe I missed this, on the reserve addition in -- on the environmental side. I noticed in the Q, you talk about new claim levels increasing for I guess the mid-1980s policies. Maybe just expand on that a bit. And is that just a one-off? Or is that something which is beginning to tick up?

Yes, they're actually -- this is Jay Benet. They're actually not increasing. What we see is each year a flow of new claims coming in. And the flow has been decreasing. When we do reserving, we make estimates as to the level of decrease that we're going to see in future claim activity. And in relation to the actual flow, the actual flow has been a little higher than our assumptions have been but they have been decreasing. And given the nature of the coverage that these things relate to, a lot of these things relate to policies that were written decades ago. So that's what we're trying to communicate in the Q.

Our next question comes from the line of Josh Stirling with Sanford Bernstein. Josh Stirling - Sanford C. Bernstein & Co., LLC., Research Division: So I'd love to ask a follow-up question around severity trends. You guys mentioned helpfully, I think Brian, the 4% rate you're seeing overall. And I'm wondering if you could give us some color on your major drivelines of business? What you're seeing in terms of, if you want to sort of model rate need versus pricing you're taking? What's your best assumptions we should be using for loss trends? And then also, there's been a lot of conversation broadly within the industry around severity acceleration, lately around BI. But we'd love to see because you guys have exposure to a lot of different lines sort of what you're seeing? And if we should be thinking about the emerging issues there starting to work their way into either casualty lines or workers compensation or even some of these sort of higher layer and sort of less slip-and-fall type stuff?

Let me start and take a shot and you threw a lot of stuff in there. So again, just to make sure that we're grounded, the 4% trend that we're talking about is in BI. I would say...

Business Insurance.

Business Insurance, I'm sorry. Across the entire place, we didn't see anything dramatically changed this quarter. So whatever I'm going to say about severity trends is the same thing we've been saying for the last several quarters. In Business Insurance, we've talked to you starting, I don't know, probably a year ago now on the workers comp trends. Those have kind of leveled off but they've leveled off at a higher than traditional rate. So that's one of the places that we're continuing to watch it. Again, no deterioration this quarter, but... Auto, across both personal and commercial, are places that we are looking at. The bodily injury severity is something that is driving through the Auto results. Again, this is both personal and commercial lines. And it's really on the medical side running clearly higher than medical inflation, the medical CPI, more hospitalizations. And on the personal lines side, a significant uptick in pedestrian losses, et cetera. I don't think we're the only company talking about that. On the -- also on the Auto side, the physical damage side from a personal line, the increased value of used cars has had a significant ripple through the physical damage coverages and the cost -- repair costs and replacement and total loss valuations are up fairly dramatically. Again, those didn't tick up higher in the quarter, but they continue to run higher than normal rates that we would see. And then, we've got all the dynamics running through property, which is primarily a weather frequency issue. But those are the big drivers that we're looking at. Josh Stirling - Sanford C. Bernstein & Co., LLC., Research Division: That's really helpful, Brian. Within the medical -- the various medical related lines, is this sort of a economically driven sort of customer utilization of P&C insurance sort of growing? Or is it -- do you think it's something due to hospitals and medical providers maybe in a sort of a post-Obamacare in world starting to figure out how better to cost shift to the insurance industry?

Yes. So hard to know exactly. We don't believe it is fundamentally people trying to find health coverage through our products in a fundamentally different way. On the Auto side, it appears to be driven by more serious accidents and more serious injuries. Now we could debate the causes of that and distracted driver, et cetera, distracted pedestrians across the board. And we and the industry are trying to figure those things out, but we don't believe it's fundamentally people trying to access our products for coverage in a way that they haven't historically done.

We do believe that this issue of distracted drivers, importantly, and distracted pedestrians crossing the street while reading a device and getting hit, having accidents. That's a real factor going on here. And how to sort it all out and figure out the implications of it, quite difficult. But we're spending a lot of time with it. Those are 2 societal trends that we believe are clearly showing up in the numbers. Josh Stirling - Sanford C. Bernstein & Co., LLC., Research Division: All right. That's really helpful. Jay, if I could just ask one question. You made your point -- you made a point early on about sort of a challenge of getting back to a mid-single -- mid-double-digit ROE. When you talk to your largest agents and you sort of think about sort of positioning them, is there a consensus among customers, buyers and the people who'll ultimately pay for this that, that's a realistic goal and something that the industry should be able to shoot for? Or is your comment that actually -- should we take that to be that's just probably just an unrealistic thing in this environment?

That's really a great and interesting question. I do speak to agents a lot. One of the things when I talk about anecdotal observations, it's not just from our own folks but it's talking to agents about their interactions with customers. Remember, we're quoting an account. It's the agent or broker who's actually having the conversation with the agent. And I -- we never really talk about returns -- agents. It's just not something that they really think about or contemplate. When I ask the really good ones what the nature of their conversation with their customers are, what I often hear is a conversation around -- that the agent tells the customer -- "I tell the customer notwithstanding the increase, they are still paying less than they paid for coverage in 200X, whatever X is." So the cumulative effect of the rate declines over those last few years, notwithstanding the increases -- I hear things like, "I tell them that they're still paying less than they paid in 2009 or in 2007." And that the conversation between the agent and customer is okay. I will tell you that we have not had much angst or pushback from agents or brokers. And believe me, if we were being disruptive to their business, we would hear it. There is no doubt in my mind that we would hear it. You can go back a couple of years ago when we started this and we identified as an important factor not being disruptive to agents, brokers or customers. That was step #1 for us. And so far, so good. I don't think agents -- we don't talk about returns on account. Agents know when an account has had attractive loss performance and when it's had poor loss performance. They're on top of their accounts, they know. If we execute our strategy thoughtfully and intelligently here, we're having conversations about accounts that have had worse than anticipated performance, and that's where the rubber really meets the road. And there's -- they understand and they talk to their customers and they understand. I think that's one of the reasons why we've been able to execute this as successfully as we are. It has not been one-size-fits-all approach to the pricing. It's been very account- or class-driven.

The one thing I would add -- this is Brian -- is that, and completely agreeing with Jay, our agents and customers don't think about our ROEs. They do, the agents clearly and the middle market and above customers do understand the impact of investment yields on our business. And so, we can have conversations around, given the decline in the investment yields, logic would say the insurance product, the underwriting should be more profitable. They understand that dynamic.

Our next question comes from the line of Matthew Heimermann with JPMorgan. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: I guess the first question would be, I think in aggregate, most people's view of reserves right now is that they're still [ph] generally redundant, but I think -- I also think that some of the recent accident years is where there are some current concern to. So given we're still seeing some pretty significant favorable numbers coming out of your book, just be curious if just how the 2010, 2011 accident years are performing within the context of that?

Yes. This is Jay Benet. In the quarter, we didn't see any movement of any magnitude associated with the 2010, 2011 accident years other than what we talked about in the releases. Some of the cat losses that we had put up last year developed a little favorably. But at this point in time, as we've said before, whatever we are seeing in terms of loss trends and claim activity, we're building into our loss specs, we're building into our reserves, we stay very current in that in a quarter-by-quarter basis, and nothing new has come up in the quarter. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: Okay. Any -- and how about -- would 1Q be the same comment?

I think in the first quarter, we talked about some development that had taken place in the 2011 accident year. Part of it is -- as we refine our calculations about particular years, there are some -- I don't know I'd call it almost arithmetic or administrative adjustments that we make. But there was nothing alarming, nothing causing us to shift the view as to profitability associated with a particular type of product.

We did last year strengthen 2010.

Last year, yes.

Just to reaffirm that. So in 2011, we did take actions to strengthen the 2010 year. And I think in the aggregate that 2010 last year in the aggregate did develop negatively. We don't know that there was any other year that in the aggregate has developed that way prior to '10?

No.

No.

And what Jay is really saying...

And year-to-date 2010, we haven't made any...

Any adjustments. So the changes that we made last year for 2010 continue to look okay. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: That's fair. That's right. That's in part why I was asking the question. The other -- I guess and Jay, Brian, I don't know which of you wants to go first on this, but can you just maybe talk about kind of the nature of the competition you're seeing right now? And I guess, there have been some broker comments about regionals generally being more price-friendly than maybe some of the national carries. I don't know if that necessarily is really all that different from what we normally hear. But I guess part of the reason I'm asking that question is when we go back to the cohorts you show, the fact that you've got maybe more rate opportunity vis-a-vis retention in that last cohort, is that all that unusual relative to what you'd expect to see in an environment where prices are probably moving up? Or is that -- or do you think in your minds that's actually somewhat unique?

No. So part of our hesitancy here is well, we obviously don't have perfect transparency into what our competitors are doing. We -- I would say, to your point, not fundamentally different. Your comments about how regional companies are acting relative to national companies, I would say is by and large the normal environment and that seems to continue. I think...

By that, I mean, because we're -- I would say that we don't feel out of pattern. I'm putting -- I'm seeing if those are words that fit your instincts, right? Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: Yes. And I think that's a great way to phrase it.

We don't feel out of pattern, broadly speaking. Every account always stands up on its own. And there is individual discussions about an individual account. Most of them are okay, sometimes they're not, sometimes there are issues or problems or a history. But conversations are unique to the individual account. But we don't feel out of pattern. We don't hear anything anecdotally that suggests that we're out of whack here. Matthew G. Heimermann - JP Morgan Chase & Co, Research Division: Okay, that's helpful. And then the cohort, too, just kind of maybe having more opportunity to push rate. And that last cohort, is that kind of within the pattern as well?

Yes, yes. We'd say it's not unusual.

Underwriters don't like to lose accounts. They are sales people. They are thoughtful, technical, smart sales people, but they're sales people. And they don't particularly like to lose accounts. So part of the process here is to cause them to feel secure that our goal is to improve profitability. Volume will be what it will be as long as we are in the broad framework, moving down the road to improved profitability. And importantly, culturally, here, you won't hear criticism from senior folks about an account that was lost. If an account wasn't renewed, it's just fine as long as the process behind it was thoughtful and engaged the agent and was reasonably done. And so it's not unusual in that -- in those more difficult cohorts as you describe it. It's just a little more challenging to get underwriters to feel that confident about that group as they do about any other group of accounts that they're working on.

And this is Jay Benet again. I just wanted to just add something to what I said about reserves. Gabby reminds me that we did say in the first quarter, as it related to the 2011 accident year, we did strengthen commercial Auto a little bit. That would be the one place that's...

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

The first question is again once again on Slide #14. Just wondering how those bars have shifted in the sense that, have you achieved greater profitability on the 60% to 90% and more than 90% loss ratios? And is that the reason why pricing is flattening because less of the accounts are performing badly?

Well, that's actually -- it's a good question. We -- because we talk about that actually right here, is there -- does there come a time where the returns in the business are such that we're simply not seeking that kind of rate increase that we might have before. On a theoretical level, I'm sure that's -- the answer to that is yes, but we've not hit that yet. We're not -- we are not at the point where there's a fundamental shift in strategy -- pricing strategy going on because we've hit some artificial barrier. There are no such artificial barriers. And I would resist the temptation, as you've described it, to say that rate is flattening out again. Even we can't make that statement, knowing everything that we know. It's which accounts came due, what were the returns on the individual accounts, what were the rate needs. It would be inappropriate to draw a conclusion -- good or bad -- from one quarter to the next. We'll have a better sense as we get into the third and fourth quarter as to what's actually happening. That I think is the meaning in the context of how we're continuing. Is the improvement in profitability going to continue to [indiscernible] itself in rate or in some other way? And the goal here, as Brian said before, is not to increase rate. It's to improve profitability. That's the goal. And it's -- so it would be a mistake, I think, to draw a flat conclusion from 7.5% versus 7.4%. What we do know is that if we look historically at the slope of that line, we feel very good about it. And that's the magic. That's the -- it's so easy in our business to not know enough that an increase in rate yet a modest decrease in retention not have it convert to improved profitability. It's not just getting rate and maintaining retention, it's getting rate for the right accounts, retaining the right ones, not retaining the wrong ones. And so it is indeed that granular to come to a conclusion of whether we are making progress on improving profitability, and we have a high degree of confidence that we are.

That's very helpful. You've been good on raising rates. I'm just curious, you've a softening on the economy. How do you think that's going to impact your ability to raise rates?

Again, that's a sort of an outlook kind of question. When we started this -- this is interesting, and it's a legitimate question -- one of the things we were concerned about was the ability to raise rate in a difficult economic environment. I'm not sure that the environment got that much better or so far, it's getting that much worse. I think when you look broadly at it, it is largely kind of about the same. We do see pickup in exposure, that's been a plus. We do see a turn in Auto adjustments that were negative and have now become quite positive, that's a plus. It will -- we will certainly be challenged if, in fact, exposure begins to drop. Real exposure, not disclosed at the front but disclosed and then subsequently audited. It will be certainly more challenging to continue to improve profitability in that environment. And exposure has been modestly positive. Maybe even a little more than modestly. But it's been nicely positive. And so yes, I think if exposure really changes, not sort of the talking heads discussion of the economy on cable news, but if our accounts really begin to have less demands for insurance, it will be more challenging I'm sure to continue on in the process. Not impossible, but more challenging.

Okay. That's helpful. And then one last thing, if I may. [indiscernible] On the personal/Auto severity side. There was some discussion on high severity. Just a question on whether it seemed that ticked up in the last couple of quarters? Or is it similar to what you were seeing maybe over the last 1 year?

We talked about in the fourth quarter and probably going back to the third quarter and fourth quarter of last year and very slightly into the first quarter of this year, what we've seen the last quarter was pretty flat with what we've been seeing before. But it's running at an elevated level, at a higher than historical norm level.

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

I just want to circle back on a couple points. First on the return on equity goal, if we look at the return on equity contribution from investment income, it's running in that 9% to 10% range. In this sustained low interest rate environment, how much downside do you see in that aspect of contribution? Then I have a follow-up.

First, I'd make an observation that portfolio continues to mature at about $6 billion to $7 billion a year, talking about the fixed income portfolio. So there is more repricing that will occur. It is not -- the current returns and, in fact, the decline over the last 12 months is not insignificant going from a tenure of 3 to 9, more precisely to 1.5. That will continue to have an impact. So it's -- the fixed income portfolio, all other things being the same, yield will continue to decline. It will decline modestly period-to-period because only $6 billion to $7 billion mature. But it will continue to decline, and that will indeed put -- make it more difficult. Jay, you're exactly right, make it more difficult to achieve that mid-teens ROE. Certainly, in the near-term, which is why I said in the near-term, one of the reasons important is the investment environment. You can't do it today. It's just not achievable. Now the question is, do we return back at some point to what we all would consider a more normal environment whatever that is, and how quickly does that occur? We've been at this level for actually a long time already. It was September -- the crisis was in '07, Lehman Brothers' issues were September of '08. We're up to 4 years now since that happened. This has been a long haul. And I have no crystal ball other than to think that rates are going to stay at this level for at least the intermediate term. Certainly, everything we're hearing from the Feds suggest that. But yes, it will just make it more challenging. No question.

Understood. Okay. And then maybe for Brian. Around that Business Insurance rate increase level, 7.4%, can you give some discussion in terms of the range you're seeing around that midpoint by major line of business whether that's property, workers comp, GL, Auto?

Yes. I mean, it's actually pretty tight. The...

Tight by business, quite broad by account.

Right. So tight by line when you talk about -- and as I said in my comments, Comp and Auto are still running at the highest comp price change levels, but that's at about 9 points and -- but I think the lowest is about 6.5. So every product line is running at a decent rate increase. To Jay's point, the distribution by accounts is dramatic. So one point I'll make to just contrast how things move. Rough numbers, but if I look at our Commercial Accounts business a year ago, about 35% of our accounts were at 0 or a price decrease, and about 15% were getting 10% or more increase. That is actually flip-flopped. So right now about 35% of our Commercial Accounts renewals get a 10-or-more percent increase. But still 15% are at 0 or a negative. So it's a very granular execution. So there, there's a really broad range. Does that help?

Very much.

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

Very good. Thank you very much. We wish you a good day. And as always, Andrew Hersom and I are available for follow-up questions. Thank you.

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