Good morning, ladies and gentlemen. Welcome to the third quarter earnings review for Travelers. We ask that you hold all questions until the completion of formal remarks, at which time you will be given instructions for the question and answer session. As a reminder, this call is being recorded today, Thursday, October 23rd, 2010. 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 now begin.
Thank you, Frank. Good morning. Welcome to Travelers' discussion of our third quarter 2010 results. Hopefully, all of you have seen our press release, financial supplement, and webcast presentation released earlier this morning. All of these materials can be found on our website at www.travelers.com under the investor section. Speaking today will be Jay Fishman, Chairman and CEO, Jay Benet, Chief Financial Officer, and Brian MacLean, President and Chief Operating Officer. Other members of senior management are also in the room available for the question and answer period. They will discuss the financial results of our business and the current market environment. They will refer to the webcast presentation. Then we will open it for questions. Before I turn it to Jay, I would like to draw your attention to the explanatory note on page one 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 10Q and 10K filed with the SEC. We do not undertake any obligation to update forward-looking statements. 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 investor section on our website, travelers.com. With that out of the way, here is Jay Fishman.
Thank you, Gabby. Good morning, everyone. Thank you for joining us today. We're very pleased with our third quarter results, having posted net income of $2.11 per share, an increase of 28% over last year, and operating income of $1.81 per share, a 12% increase over last year's mark. Operating return on equity was a strong 14.3%, and we experienced very solid underwriting results across each of our business segments. Our consolidated GAAP combined ratio was 90.6%. In an insurance pricing environment that remains largely flat in the commercial businesses, we were pleased to produce an increase of 2% in net written premiums for the quarter. Investment returns remained reduced from historical patterns as remarkably low interest rates, particularly short rates, continue to impact net investment income. Lastly, we've grown our book value from year-end 2009 by 13% to $59.11, all in a very strong quarter.
This morning, we're going to do things a bit differently than we have in previous quarters. All of the data that we've historically provided is still included in our webcast. We're only going to hit the highlights, and you can review the details on your own, and obviously, Gabby is available later to take any questions you may have. We're going to take the opportunity to share a few additional insights into our business that we think you'll find interesting. First, we've spoken at length in previous presentations about our programs designed to grow our business organically. This morning, we're going to share with you summary analytics of our business insurance account growth, and we suspect that a number of you will be surprised at the growth we've achieved. Second, a number of you have asked about the impact of reduced interest rates on our net investment income.
Jay Benet shared an analysis of the impact of reinvestment rates at an investor conference in September, and we've updated it for current conditions. We've analyzed the maturities in our fixed income portfolio for the next three years, and we'll share with you the projected impact on net investment income if reinvestment rates remain at current levels. The municipal bond environment remains a topic of some interest. We've updated an analysis of our municipal bond portfolio that we did for the second quarter webcast based on current ratings and market conditions. As we've said before, it is an actively managed portfolio where we are constantly evaluating risk and return of individual securities. There are never any guarantees, but we couldn't be more pleased with the positioning of the portfolio. Before I turn it over to Jay Benet, I want to comment on a topic that we're thinking about regularly.
As we've discussed before, we believe the most thoughtful way to run a property casualty business for the long term is to produce superior returns on equity. As many of you understand, we can't promise consistent growth in either revenues or earnings and do so maintaining a thoughtful risk profile. A number of years ago, consistent with our aspiration for superior returns on equity and given what we viewed as typical underwriting and investment environments, we determined that we would target a mid-teens return on equity over time and do so by achieving top-tier profitability and returning excess capital to shareholders. Two very important words in that previous sentence are over time.
Since January 2005, the first full year after the merger of St. Paul and Travelers, our average annual operating return on equity is 14.1%, and we've recorded more than $20 billion in operating income and have returned nearly $17 billion in total to shareholders, $13 billion in share repurchases, and nearly $4 billion in dividends. Given the current general economic and investment environments, a few people have asked whether we intend to change our goal now. We're certainly not economists, nor do we know what the future holds. The underlying assumption we are making is that the economy will eventually return to a more typical investment environment, particularly with respect to the fixed income world. We are not changing our goal now.
Having said that, we've been very clear in the recent past that the current environment simply doesn't permit achievement of a consistent mid-teens return right now if one assumes no favorable reserve development and normal catastrophes and weather costs. If, in fact, the market ultimately presents long-term meaningfully reduced investment returns, not only for us, but also for the market at large, which for us are not offset by improved underwriting conditions, we will obviously rethink what that means for our aspirations and what investors can expect of us. For now, our target remains intact, and we continue to execute in the marketplace consistent with that goal. We seek rates selectively and thoughtfully where rate is needed, especially in those cases where our account loss experience has been inconsistent with our underwriting expectations. We are always managing rate and retention with a clear view of maximizing long-term value.
To arbitrarily and aggressively seek rate and be a victim of adverse selection and watch our retention drop precipitously would not be a smart or thoughtful reaction to the current environment. As Brian will explain in greater detail, our field people take their lead from us, and right now their direction is unambiguous. We feel great about how they are executing, and we believe that we will continue to be amongst the best-performing property and casualty companies around. With that, let me turn it over to Jay.
Thanks, Jay. I'd like to start with a few overall comments about the quarter relating to pages four through 10 of the webcast. Operating results, ex cats and favorable reserve development, and including net investment income, were generally in line with our expectations as they have been all year. Our third quarter results benefited from a relatively low level of cats, $117 million pre-tax, which was lower than prior year cat losses of $158 million pre-tax, as well as what we would "normally plan for" in the third quarter. I would remind you, though, that year-to-date, cat losses have been quite high, over $1 billion on a pre-tax basis as compared to $441 million pre-tax in the first nine months of last year, which would approximate a more normal year-to-date amount.
We had another quarter of net favorable prior year reserve development in each of our segments, concentrated this quarter in DI and FPII. The quarter's reserve development included an asbestos reserve increase of $140 million pre-tax. Included in this asbestos reserve increase, and therefore in the current quarter's operating income, was a $70 million pre-tax benefit related to the recent favorable ruling we received against Munich Re and certain members of Echo Re. The total awarded to us by the court in this ruling was $417 million, broken down as follows: $251 million owed to us under the terms of the reinsurance agreement and interest of $166 million. Based upon this ruling, we reduced our uncollectible reinsurance reserve by $70 million, thereby reflecting in our financial statements the full $251 million owed to us under the terms of the reinsurance agreement.
Importantly, the benefit of the $166 million interest award was not reflected in either the current quarter's asbestos charge or in the current quarter's operating income, since GAAP requires this interest to be treated as a contingent gain, generally until all appeals have been exhausted and/or the dollar amount is received. All of our capital leverage and liquidity measures remained at or better than target levels. We had holding company liquidity of $2.8 billion at the end of the quarter, which was more than twice the target level due to the timing of share repurchase activity and the timing of dividends from our operating companies to our holding company.
During the third quarter, we repurchased $600 million of our common shares, a higher amount than we would normally repurchase in the third quarter due to the strength of our balance sheet and the light wind season we have experienced thus far, and paid $169 million in common stock dividends, bringing the total cash we returned to our shareholders in the past nine months to almost $4 billion. Third quarter operating ROE was 14.3%, and book value per share once again increased, up 6% in the quarter and up 15% from a year ago. Page four makes reference to our very high-quality investment portfolio, which includes a $4.6 billion pre-tax net unrealized gain at the end of the third quarter, up significantly from the $2.8 billion pre-tax net unrealized gain we reported at the beginning of the year.
A little over 60% of the net unrealized gain relates to the muni bond portfolio. Page five provides an update of certain data we previously provided to you related to our muni bond portfolio. There have not been any significant changes in the portfolio recently, notwithstanding our active management of the portfolio based upon a risk/reward view for each individual holding due to its very high quality. The portfolio currently includes bonds that have gross unrealized gains of $2.8 billion pre-tax and bonds that have gross unrealized losses of only $8 million. That's right, only $8 million of losses, not billions. I'd also like to make some specific comments related to page nine of the webcast.
During a presentation I made at a recent conference, I referred to today's low interest rate environment and provided an illustration based upon our portfolio schedule bond maturities in 2011, 2012, and 2013 of the impact on net investment income in those years if the current interest rate environment persists through this period of time and all of the variables such as average assets, mix, credit quality, and duration remain constant. The illustration I provided at the conference was based upon reinvesting at 10-year rates or approximately 100 basis points lower than the combined yield on the maturing bonds, rounding the interest rate differential to the nearest 50 basis points. Since then, interest rates have fallen even further, and we have updated the illustration to reflect a 150 basis point interest rate decline.
We've also added another line to the illustration that shows the differential between 2010 NII and the subsequent years. Again, all of the variables being held constant. That includes the partial year impact of reinvesting 2010 maturities at lower interest rates. The full year impact of this activity will not be felt until 2011 and subsequent years, since the 2010 maturities will have taken place throughout 2010 rather than on the first day of the year. As you model these years, I would also remind you that NII will also be impacted by the reduction in average invested assets that results from our share repurchase program. With that, let me give the mic over to Brian.
Thanks, Jay. Webcast slide. Our typical disclosures on the business segment results are on pages 11 through 22. We will of course, take questions on any topic. Instead of going through the slides, I will give a few perspectives on our businesses in the marketplace. In business insurance, we continue to be very pleased with the positioning of our franchise. Both our account retention and flow of new business opportunities remain near historically high levels, which we believe will enable us to grow premiums if and when the economy improves. Core underwriting margins continued to contract modestly in the quarter, consistent with both prior quarters and our expectations. In short, the loss cost trend slightly outpaced earned rate changes. We continue to see the negative impact the economy is having on our insured's exposures.
Audit premiums, which are retroactive premium adjustments charged to reflect changes in insured's payrolls, vehicles, property values, or business receipts, have improved but are still negative. The exposure change on renewals, which is a prospective look, has also improved throughout the year and is now approaching zero. On a combined basis, these impacts are reducing premiums approximately 2% through the third quarter. In other words, our data suggests that the economy is bottoming out. We don't see any evidence of current economic expansion. Given these trends, our underwriting strategy remains consistent. Retain our quality business, optimize the profitability on this retained book by getting rate where it is warranted, write new business for long-term profitable growth.
Directionally, this strategy may be the same as many of our competitors, by aligning it with our competitive advantages, namely our talent, technology, actionable management information, and breadth of distribution, we believe we are very well positioned. Over the last seven quarters, we've seen significant account growth. As a result, we believe that we're growing market share. On slide 16, you can see that since year-end 2008, we have grown accounts in business insurance by an 8.5% compound annual growth rate. On the next slide, we've taken this data and broken it out for two of our major businesses. In our small commercial business or select accounts, you can see that it is growing by a 10.6% CAGR. This increase has been driven by our Select Express product. That's our no-touch insurance solution for small business customers.
Through the combination of our sophisticated rating capabilities and cost-efficient frontline delivery, we are able to provide our agents and customers with an industry-leading product and platform. Within Select, in the plus end or the larger end of this market, we have allowed our accounts to decrease due to extremely competitive market conditions. On the slide, you can also see that we've grown commercial accounts by 4.4% compound annual growth rate. The increase in the number of accounts in this business is driven primarily by the continued introduction of new products and the specialization and expertise that these products and our frontline underwriting and claims staff bring to the marketplace. This growth in commercial accounts is representative of what we are seeing across our middle market businesses.
We're encouraged by this increase in the number of accounts, and again, if and when the economy improves, the resulting impact on exposures should create a compelling written premium trend. Given both this top-line dynamic and the current core underwriting profitability of this book of business, we remain very pleased with business insurance's current performance and its position in the marketplace moving forward. In the Financial, Professional & International segment, I will make just a few overall comments. First, writings in our international businesses are down compared to prior year quarter as we are addressing risk and pricing in light of catastrophe and other severe weather losses over the past year.
Secondly, although we remain very pleased with the market position and quality of our construction surety business, the impact of the economy on construction spending has resulted in fewer new business opportunities, and our writings in this quarter reflect this. Lastly, we continue to monitor the analysis on impacts of the financial marketplace disruptions on our management liability business, and our conclusions remain the same. That is, our losses are developing within or slightly favorable to our expectations. Turning to personal insurance, we're extremely pleased with both the underwriting and production results in the quarter. In both agency auto and property, policies in force continued to grow and core underwriting margins expanded as earned rate increases outpaced loss cost trends.
In our property line of business, weather losses were both less than third quarter expectations and a welcome relief from the pattern we saw in the first half of the year. While the continental U.S. was not meaningfully impacted by any of the storms generated during the peak of the hurricane season, the number of near misses this quarter highlights the importance of underwriting controls and risk management policies in this business. Agency property production results for the quarter continued to be strong in spite of the difficult housing market with quarter-over-quarter PIF growth remaining at historically high levels. In agency auto, our new business continued to improve, and we had our best quarter-over-quarter policies in force increase since the end of 2008. Given the continued expansion in core underwriting margins and strong top-line trends, we remain pleased with both our current and going-forward positions in these businesses.
Across all of our businesses, we feel great about our execution in the marketplace, and one of the primary reasons we believe our organization has executed so well in these times of uncertainty is that we have kept our message to the field clear, unambiguous, and consistent. That is that in these market conditions, just like pretty much any other, the role of our underwriters is to maximize the long-term value of the portfolio, to balance the desire to keep our quality business while at the same time optimizing the returns on that portfolio. Fundamentally, they do that by making sure they understand the account's risk characteristics and can estimate the potential losses from these exposures. They then seek to select risks in the marketplace where the premium level is appropriate for this view of the losses.
We believe our field understands their role, has the best tools in the business to execute, and as long as they keep doing this, we will keep competing in the marketplace successfully. With that, let me turn it back over to Jay.
Thanks, Brian. Page 23 summarizes our updated guidance for full year 2010 and fully diluted operating income per share, which we've increased from the previous range of $5.20-$5.45 to $5.75-$5.95. In round numbers, this should translate into an operating ROE of just under 12%. We're now assuming cat losses of $765 million after tax, or $1.58 per diluted share, which incorporates our actual cat losses for the first nine months of the year and our original estimate for the fourth quarter. No further estimates of prior year reserve development, either favorable or unfavorable. A low single-digit decrease in average invested assets, ex unrealized gains and losses, resulting from a reduction of holding company liquidity due to the share repurchases. Full-year share repurchases of $4.5 billion-$5 billion, and a weighted average diluted share count after share repurchases and employee equity awards of approximately $485 million shares.
Jay would like to say some additional comments before we go to Q&A.
Thanks, Jay. Just before we open it up, we've had a pretty good run here for some time, and the folks around this table, I think, actually get an undue amount of credit for that performance. We've very publicly acknowledged all the folks in the investment department who have done a remarkable job in unprecedented conditions of keeping this company moving ahead. The fact is, and our own folks listen to these calls, I just want to spend 15 seconds letting 30,000 people know that we recognize that business is done in the field a trade at a time, and we couldn't be more appreciative of the underwriting discipline and the thoughtfulness and your attention. I don't care whether they're commercial lines underwriters or personal lines underwriters, whether they're in claim or technology or ops or risk control.
This is a complicated business, and we've got 30,000 folks who understand it and keep the organization moving ahead, and I just wanted to take a minute to say thanks. With that, operator, we're ready to open it up to questions.
Thank you. Ladies and gentlemen, if you would like to register a question, please press the one followed by the four on your telephone. You will hear a three-tone prompt to acknowledge your request. If your question has been answered, and you would like to withdraw your registration, please press the one followed by the three. If you're using a speakerphone, please lift your handset before entering your request. One moment, please, for the first question. Our first question comes from the line of Keith Walsh from Citi. Please proceed.
Hey, good morning, everybody. I guess first for Brian on auto. I guess lots of talk in the industry about positive rate trends. Why the decision here to loosen terms and conditions with the 12-month product? Also, if you can comment on the direct initiative where we stand, I've got a follow-up. Thanks.
Hey, Keith. This is Greg Toczydlowski from personal insurance. On the annual policy, first of all, I'll take that one. There's a number of dynamics that we look at in the annual policy, some of them being agency selling behaviors, retention, and pricing. When we look at all those dynamics together, we thought it was the appropriate time to drive that inside the business, and I think the margins and the growth are showing that. That's clearly not the only feature that we've been throwing out there into the marketplace that has a positive impact, when we look at all of them, we feel good about having that in the product.
Let me make just an observation on that agency costs matter very much in the equation, too. Instead of dealing with obviously 2 renewals a year, you deal with 1, not an insignificant difference. On the direct initiative, we don't actually have a lot to say. We continue on course with the plan that we set. We said we were going to be investing and losing money, and in fact, we are. On the positive side, we're beginning to learn a fair amount about the customers that respond, what they find attractive in the value equation. I'll let you know, we're doing about 5,000 policies a month in our direct initiative now, we're on our way to learning. Make no mistake, we do it as a very long-term investment.
Okay, Jay, just to follow up on pricing. I acknowledge your comment that you said the ROE's been very strong the last several years, the charts show clear price deterioration. Why the continued focus on growing in certain business lines here when clearly the accident year ROE is probably sub-10 at this point? Thanks.
I think the second half of your question or the statement is actually just in error. We look at a lot of data. In fact, I'm not sure anyone analyzes data any better or more robustly than we. I will tell you that even assuming current reinvestment rates now, the portfolio, meaning the combo renewal and new on our business insurance business, is not in single digits. It's actually better than that. The dynamic is we're focused, I thought my comments really were fairly specific about this. We are focused on retention and I would say taking new business selectively. No one should come to the conclusion that we are focused on growth. There isn't anything in the comments or in the data that should point anyone to the conclusion that we're focusing on growth, as the words you used. It is a very selective approach.
It is not only based on price, but it's risk selection. We spend most of the time on these calls speaking about rate. Our folks in the field spend most of their time focused on risk selection. It's, as I said, a complex business where you're balancing elements. I just couldn't be more pleased with how we're managing through this largely unprecedented environment.
Okay, thanks for clarifying.
Our next question comes from the line of Jay Cohen from Bank of America Merrill Lynch. Please proceed.
Thank you. Two questions. I'll just ask them both and let you guys address them. The first is, Brian, I think you had mentioned that your kind of core combined ratio and business insurance had gotten worse, but it looks like it's gotten better, and I'm wondering what's driving that ex development, ex catastrophes. Then secondly, maybe for Jay Fishman, when you think about your ROE, are you making any adjustments for your own cost of equity capital, given the changes in interest rates? Presumably, that has come down as well.
First on the combined ratio, I'll throw it to Jay Benet for some of the specifics. Obviously, every quarter there's stuff running through the combined. We still see when we net out everything that we think is unusual, a slight deterioration in what we call the core combined ratio. Jay, why don't you go through some of the changes?
Jay, in any period-to-period comparison, we try to look at, as Brian said, what are the key core underlying trends of price and loss costs. There are also things that'll impact it. Small weather, non-cat weather in one quarter versus another, or year to date one year versus another. We've talked in the past about large loss activity. Last year, we made a re-estimation of the full year loss pick for business insurance in the fourth quarter. We also did that in the third quarter. When you do that, it impacts prior quarters of the current year. I think when you see the full-year results, you'll see more of what Brian's talking about. If there are impacts associated with large loss activity and small weather, we'll try to point that out as well.
That makes sense. Thanks.
In terms of the return on equity, it's actually an awfully simple story. The return on equity that we publish in our press release and in our financials is straightforward GAAP return on equity, which is the earnings that we reported for the period divided by average equity for the period, Jay and Doug can take you through all of that. We obviously understand that if our entire $70 billion investment portfolio were repriced today at today's reinvestment rates, that we would end up with a different return. We understand that. Internally, the analytics that we have, we actually do look at it that way. We actually do evaluate products and businesses and lines based upon investing cash flows at today's available rates. It's those analytics that form the basis of our pricing strategy and our volume strategy, and they are proprietary. They're important.
They are the subject of substantial discussion here, it's that return that gives us our approach to whether we're aggressively growing, not aggressively growing, or willing to shrink. Brian spoke today about letting the large end of Select shrink because the pricing was such and the risk selection was such that the returns simply aren't adequate to support the business. We're letting that shrink somewhat. Certainly, we understand that one of the reasons that Jay presented the schedule that he did about investment income is that if these investment markets continue, if they are what they are, and the underwriting environment never changes, and there's no more favorable development, yeah, we're not going to be able to achieve the mid-teens return on equity. We certainly understand that, and it's actually not all that difficult to model out the business and see what you think it is.
We certainly do it all the time, and it can be achieved. It's all pretty simple. The numbers that you get in the press release are GAAP, things that we're all accustomed to getting. Internally, we have robust analytics that let us evaluate the current underlying profitability based upon today's reinvestment rate, and that's what establishes our underwriting energy or enthusiasm.
Yeah, I appreciate that. I guess my question really surrounded the hurdle rate, what you need. In other words, if you're comparing your ROE versus your own cost of equity capital, if that cost of capital has come down, shouldn't the hurdle rate come down as well, naturally?
Well, that actually gets to the comments that I made, and it's just absolutely spot on. There has, at least in my career, never been a period where the gap between the cost of debt and the cost of equity has been any wider than it is today. I find it actually remarkable that there are high-quality equities where the dividend yield is actually higher than the debt yield on the 10-year debt, high-quality companies. It tells you that the marketplace is, at the moment, somewhat upside down. The real question that you're asking, and I think it's an extraordinarily relevant one, is that going to continue? Can you possibly be in an environment where the 2-year Treasury is at 2.4% and the cost of equity is at 10%?
Will there not be arbitrage activities that will take place that in one fashion or another will lower that gap, reduce that gap into a more normal historical range? We're extremely attentive to it. One can speculate that one way that the cost of equity will go down is a rising equity market. Some folks have asked me too, why do you think the market generally is rising now? One of the answers to it is that it's reflecting a decline in cost of equity. It's somewhat counterintuitive to think of it that way, but frankly, it's entirely possible. I don't know what the cost of equity will be in a few years.
My guess is that you've got a financial company three years from now in today's environment, and you look back three years from now and the company's achieved a 10% cumulative return on equity, you'll feel great. You won't feel good, you'll feel great. If you look broadly at the financial services arena, and we do, and I do, and you look at the returns that institutions are producing, particularly in the banking arena, where the rules and regulations and leverage dynamics are changing so much, I think that the dynamic of what returns are going to be over the next few years is a fascinating question and what it is that investors are going to see. My comment was, for now, our goal is our goal. We will do the best we can.
We are not going to act arbitrarily and go out to the marketplace and say that the worst thing in the world to do is to go out to the marketplace and say, "We want five points of rate on every account." You'll get five points of rate on accounts that need 10, and you'll lose accounts that don't need any. The easiest way to get adverse selection is to take away the underwriter's authority to evaluate risk and return on an individual trade. No mystery here. We acknowledge the environment is not possible right now, not to achieve mid-teens return. The two words over time are just critical to us, and that's how we think about. This is a long-term business, and we manage it over time.
Great. Thanks, Jay.
Pleasure.
Our next question comes from the line of Brian Meredith from UBS. Please proceed.
Yeah, good morning. Two questions here. First one, I was hoping, Jay, if you could talk about the general administrative expenses in the Business Insurance unit, the big drop this quarter. Then I have one follow-up.
I'm going to sound like I'm repeating myself a little bit from Jay Cohen's question. As it relates to the expenses, again, in any quarter, you can have some things go up and some things go down. I think last year, we had some assessments that came through in the third quarter. This year, we had some credits that came through in the third quarter. I think overall, we've been very thoughtful in trying to manage our expenses and keep the expense ratio in line. Those are the kinds of things that have created the period-to-period variability. There's nothing in particular from an operational standpoint that's driving it. We do have differences in timing of things like advertising costs and travel that go through. I can't think of anything in particular that I'd point to that says this is a fundamental change.
Is there an underlying kind of rate we can think about that, were expenses kind of flat or are they down a little bit? Just trying to kind of think about it going forward. I don't want to necessarily drop expenses by 10% a quarter here going forward.
I would say relatively flat, particularly if you look at it on an expense ratio basis.
Okay. Terrific. Then the second question for Brian. Loss cost inflation in the Commercial Lines area. We've heard a little bit this quarter about some pickup in some loss cost inflation from some other companies. I'm wondering what you're seeing.
Yeah. Splitting it into two pieces. Frequency for us has been very flat, and granted, that is compared to last year and the year before, where we were having some significant declines in frequency in some of our businesses. The declines have leveled off, but we're still pretty pleased with where they're leveling off at. The core loss inflation or severity components of it is pretty benign. It's not a zero. It's a plus number, but it's a mid to low single digits plus number. Overall, loss inflation continues to be in a pretty good place for us.
Okay. Thank you.
Our next question comes from the line of Cliff Gallant from KBW. Please proceed.
Good morning. I had two questions. The first, pretty simple. There was recent news about Chinese drywall and a settlement that looked like State Farm was reaching. I was wondering if you had any comment on or an update on your view on Chinese drywall. Second, I was wondering if you could care to speculate on a sort of a larger basis what you think the industry accident year combined ratios are
Should say the industry accident year ROEs look like in comparison to what you're reporting?
We have enough trouble keeping track of our own. We're going to pass on trying to estimate what the industry is.
Good morning. This is Doreen Spadorcia, I'll take your Chinese drywall question. Bottom line, we haven't seen anything that would cause us to change our view that we talked about previously. I think we showed you a slide in the first quarter. We just don't think it creates a significant exposure to us. Basically, what's changed since the first quarter is we've probably got a few more insureds that have made claims against us. Then on the positive side, we actually had a ruling in the district court in Virginia finding that our homeowners policy does not provide coverage for Chinese drywall. The State Farm settlement, there have been some multi-district issues. There are two cases pending that have a lot of class action plaintiffs as well as a number of defendants.
Some companies that have had large homeowner populations have chosen to participate with some of those settlements, where you've seen some of the Chinese drywall manufacturers and some distributors putting in some dollars. For us, we have not participated in those given where our exposure is and the positive ruling that we received.
Okay. Thank you.
Our next question comes from the line of Matthew Heimermann from J.P. Morgan. Please proceed.
Hi. Good morning, everybody. Two questions. First, could you give us a little bit more color on the BI account growth slide you showed? I guess I'd be curious, one, whether or not there's a dramatic difference in account size for the new accounts versus maybe the existing book, if you go back to your end 2008. Also just given what we've seen in premium, could you maybe just discuss how the exposure trends at those new accounts maybe contrast with exposure change on the existing book? I'll have a follow-up.
For starters, on the mix dynamic within the account growth, we tried to break out the small commercial from the middle market. Obviously in small commercial, we are absolutely growing it from a mix perspective in the smaller end. That's driving some of the dynamic of the total. Within that, the express component of small commercial, which would have a fairly consistent account size, is growing dramatically. We feel good about that.
Importantly, just so that people know the terms, Select Express is the technology platform that we introduced now, I don't know, maybe three years ago, that has really dramatically changed the way the smallest end of our small commercial business is processed. I discussed this before, the predecessor platform was 20% underwritten in the technology, 80% was referred out to underwriters for review. This is a very flow approach to the smallest end of the business. 80% now is done in the technology, only 20% comes out for human intervention. It is dramatically different. That forms the basis of the flow, the significant increase in quote activity that we've seen from agents. We are just being quoted a whole lot more with this platform because it's easier and more efficient, as a result, that business is growing significantly.
It is the smallest end of the small commercial business.
I think, Matt, and you can correct me if I'm wrong, the gist of your question is, are we getting the growth just because we're shifting mix, or do we think we're actually growing like type of accounts? The answer is there's a piece of it that's mix, but the bulk of it is we're growing like type of accounts.
Yeah.
That's fine.
In the middle businesses, the average account size of what we're bringing in new is fairly consistent with our renewable for business.
Right.
The difference between account growth and premium is basically the exposure, the audit premium and the renewal exposure change, not increasing year-over-year.
Yeah.
Okay.
help me a little bit with the second half of what you were asking.
You got to it with the last comment, so that's helpful.
Great.
The other question I had was maybe to follow up on Brian. Actually, maybe I'll ask a different one. On the direct side, I think you used the word experiment, Jay, but it looks like the premium volume is starting to decelerate sequentially kind of low to mid $20 million on a quarterly basis. Would expect that to grow, I would assume. I guess, are you proactively kind of trying to restrain the growth in that channel till you ensure you kind of understand the dynamics? Are you now at a point where growth will be what it'll be, you kind of have a more clear view of the appetite and some of the variables that you want to pay attention to?
I'll give you my perspective, I'll ask Rick to chime in as well. It's sure closer to the former than the latter. We're actually, I wouldn't say constraining, that's not quite the word, but what we are doing is really only doing that amount of business that we need to do to continue to advance the learnings. The learnings are, how do we get a customer to respond to an ad? How do we get that customer either through our call center or our technology platform to actually end up with a quote? How do we convert that quote to a sale? Then most importantly, who are we bringing in and what's the loss experience? We've said the pricing track we're using for direct is the same pricing track we use at our agency plant.
It is our presumption that the loss experience between those two groups will be different, that the act of where one purchases is an identifying characteristic, a projecting characteristic of loss experience. We're just trying to do enough to keep the learnings moving ahead. We are not remotely at the level now where it's go invest in advertising dramatically and whatever we can do. Ed, please.
I'll just echo Jay's comments. We spend a lot of time watching the economics and the operational expectations of the business, they're both right within the targets of where we want to be. As Jay said, the function of the top line is how we advertise, how we entertain ourselves out in the digital space, clearly, that's the amount of investment that we put in this. I think it's a fair assumption that if you look at the last three quarters run rate, that's a fair assumption of what we could expect for this business to continue running at 12 months going forward. As Jay said, we're really trying to really maximize our learnings with minimizing the investments inside this business. As we get more of those learnings, we'll continue to grow the business. Very cautious in how we're doing it.
Wouldn't characterize it, by the way, as an experiment. The implication of that is that we're committed to making this work, it will take years. It will take years for us to get to the point where it gets to break even or frankly, becomes profitable. We are committed to getting there, what we're doing is to Greg's exact point, is balancing the cost of investment with the value of the learnings that come out of it.
Okay, that's fair. That was my poor word choice. I guess, based on your last comment then, isn't fair to think about Quantum as a parallel in terms of, because I think there was about a 24-month period of rolling that out and to the point where you were kind of fully comfortable just letting it run. It sounds like-
No, this is going to take much longer
be much longer. Okay.
Yeah, this could be much longer than that. Yes.
Our next question comes from the line of Jay Gelb from Barclays Capital. Please proceed.
Thank you. I want to ask a broader picture question on the situation with the bank dislocation with regard to mortgage servicing. Could you talk about how or what the implications might be from the insurance side on that, from a directors and officers and errors and omissions perspective, as well as what the implications could be for Travelers' investment portfolio? Thanks.
Jay, it's Alan Schnitzer. Let me take the first piece of that related to the insurance business. It's early days, and when I say early days, I mean in the context that we're still learning about the facts and circumstances. It's hard to really come up with a definitive insurance exposure perspective in that environment. Having said that, we've taken a look at our exposure to the top 15 mortgage servicers, and our exposure is really very limited there. That is in some respects, a fallout of the credit crisis underwriting we've been doing, going back to the first or second quarter of 2007. All the big names that you and we have been reading about in the news, we have very limited exposure there.
We don't view it as an exposure really to the community banks. We think that's because they just haven't been foreclosing in the kinds of volumes that would have involved the processes that are really in question here. We haven't seen any indication that that's the case. At least early days from an insurance perspective, we're not viewing it as an outsized issue for us.
On the investment side, just first, a couple of facts. First, our entire portfolio of subprime and Alt-A ABSs are actually $300 million, and $200 million of that was 2004 and prior, and we've added $100 million selectively since the crisis began. Not a significant investment there. Still all very highly rated. On the residential CMO side, we've got $2.3 billion of that. Again, virtually all 2004 and prior. We turned away from the market as both the mortgage market and the real estate market really heated up. $862 million of it is agency and $1.4 billion is non-agency. The facts and circumstances of what actually is happening and the implications are as confusing as I've ever seen it, and it's not at all clear what the long-term issue really is.
If we're talking about episodic, relatively short-term delays in foreclosure activity, our assessment is no problem with respect to the investments. We've already seen some of that episodic short-term delay. At the other extreme is the notion of a long-term, industry-wide, broad-based foreclosure moratorium. We don't know how to assess that. I don't know how to assess something that actually has never happened before. Frankly, I think of all of the issues that will occur, the impact on asset-backed securities may actually be the smallest in that kind of an environment. I don't know how to size up that kind of long-term, broad-based moratorium. Again, the episodic short-term stuff, we could be wrong on this, but our assessment is no underlying problem.
With respect to the flip side of the issue, which is the ability to put mortgages back to the originating institutions, of course, the GSEs have always had that. There's nothing particularly new on that front. Again, we could be wrong on this, but our assessment has been that it's hard for us to figure out a scenario where mortgages get put back to their originating institution and the bondholders lose. The originating institution, seems to me, could be in a position to lose, but we have some difficulty figuring out the scenario where the bondholder loses. Early days on this would even be an understatement. It's right at the beginning. Bill is-
I wouldn't add much. You obviously read the story about PIMCO and the Federal Reserve Bank of New York sending a letter to Bank of America with a list of 158 pools where they would like to see mortgages put back. We looked at that list, by the way, and found we own three of them. I think there are three cross currents. One, the validity of foreclosures which have taken place. Two, the ability on a prospective basis to foreclose. Courts are beginning to get more particular in terms of what they demand, and that's probably good. Three, the whole put back phenomenon. They would cut differently across the portfolio. My guess is the probability of any of the three having a significant impact to us or anyone else, that might be more remote than the newspapers are suggesting.
Whatever happens is going to happen over time. To the extent that there's a put back phenomenon, we'd benefit. It's easier for GSEs to put back mortgages than non-agency pools because all the GSEs have to do is prove they weren't conforming. We'd be beneficiaries, but we're not counting on much.
Thanks. That was very comprehensive. Just to circle back to the property casualty exposure. Is there a standalone mortgage servicing E&O type of coverage that would be meant to respond to that, or is that all sort of lumped into the broader bank D&O and E&O programs if there's availability for that these days?
When we think about our exposure, it's way heavily weighted towards the D&O and E&O side, and again, underweight on the larger institutions. There is coverage that I think is available in that market. We just don't write a lot of it.
All right. Thank you.
Our next question comes from the line of Vinay Misquith from Credit Suisse. Please proceed.
Hi, good morning. On the personal auto side, could you provide some color on your PIF growth in the agency channel? Some of us find that the direct way is the way to go, and you seem to be growing in that channel pretty well.
Yeah, I'll take that. It's Greg Toczydlowski again. We've been focused very much on a couple areas. One that we talked about earlier is some of the features in the automobile, and two is some of our geographic expansion. We've seen an under-penetration in the Midwest and the West, and we've been appointing agents out there over the past few years. Based on those two really has been driving some of the sequential PIF growth that we have inside the book of business.
Okay, that's great. The second question is on pricing in the business insurance. It appears that you're happier to take maybe prices down just a tad just to keep your business and to grow a little. Just wanted your perspective on the risk to that strategy. We are at all-time historical lows in terms of frequencies. Do you think that this is the right time maybe to grow your business?
Well, I'll let Brian step in, but I want to start off with the premise that tactical and strategic imperative number 1, 2, and 3 is retention. Again, somehow the conversation always seems to kind of drive to growth. We start out with a focus that the book of business that we have is the book of business that we understand, that's priced appropriately, that's the highest return. We approach our retention book in a very thoughtful kind of way. The growth dynamic that you're seeing here in the charts, in account growth, I've got to go back and talk to the fact that one's being driven by Select Express, which is a technology platform that lowers costs, lowers agents' costs, lowers our costs significantly. It's been a platform-based dynamic of growth, not price-based, a platform-based dynamic of growth.
In William Cunningham's, in his middle market business, it's really been about new product development and rollout. We've provided lots of information about that previously, not a price-driven strategy. Again, the data is so clear that it supports this. You're looking at retention that is at historical highs and renewal pricing that's more or less at flat, a little below or flat. Our focus is not, and it does seem to get confused to some folks, is not to cut price to aggressively grow our book. It's to use our competitive advantages where they exist, like Express, like the new product development in middle market that allows us to or the geography that Greg spoke about, where there are ways to grow your business without being pricing competitive to do it.
Yeah. The other dynamic within pricing, and you used the right words. If you look at any of our business insurance statistics that are in the package, the negative price we're talking about is somewhere around a negative 1. We've talked about this a lot in the past, and Jay touched on it. It's the spread across the portfolio that really matters there. If we were getting minus 1 on each and every account, that's 1 strategy. We've got accounts in here where we're getting plus 10s, and we've got accounts that are certainly getting minus 10s. It's blending to something pretty close to neutral. We feel good about the profitability of our book, and we wake up every morning wanting to retain most of those accounts.
I think to characterize our strategy as taking down prices to grow, in our minds, is not consistent with how we think about it.
I want to take the opportunity to clarify something because I'm quite certain Gabby will get 10 calls before the end of the day on this, it gets back to the analytics that I was speaking about before that evaluates returns on an investment return basis today. Let me be a little more specific
We're talking about allocated capital. Each of our products, each of our businesses, based upon the duration and the volatility, our actuaries assign an amount of capital to. The embedded return that we calculate for the portfolio assumes that the premium dollars come in at new investment rates, but that the capital, because the capital sits and just continues to roll, has a much longer duration. The capital that supports it is somewhat more reflective of the historical capital embedded in the portfolio. Premiums come in entirely at new money rates. Again, this is the analytics that we use. The capital, based upon allocated capital to that product, reflecting more of the historical duration kind of, because again, it sits and supports the business. As one piece of business rolls off, a new piece of business comes on.
The capital has a different duration than the new premium base. If you look at the business insurance segment in the aggregate, as we did just yesterday in making sure that we could answer this question, the policy view, the return that we see in the business that we're writing today is very low double digits. That's where it falls. I don't want to get more specific than that because obviously it's reflecting of our pricing strategy. It wouldn't take much to move it into high single digits, but it's very low double digits.
Okay, that's great. Thank you.
Our next question comes from the line of Greg Locraft from Morgan Stanley. Please proceed.
I wanted to follow up on the holding company liquidity and get an update. I can't get the math to reconcile. I'm looking at the end of the second quarter. I added net income, took out share repo and dividends, it looks like liquidity was almost $200 million higher. Could you help reconcile the math there? Was there a dividend in the quarter, perhaps?
Let's see. At the beginning of the quarter, for holding company liquidity, we had about $2.4 billion. The mechanics of this thing are taking dividends out of the operating companies and bringing them up to the holding company. There are other things that impact it, like stock option and equity awards. That's a relatively small number. Taxes come into play. There's going on the other way, interest on the corporate debt, shareholder dividends, the stock repurchases themselves, any pension plan funding we do. Those are the components of it. I'm not sure you can really have full visibility to all those components. What ends up happening is that the $2.4 grows to $2.8, and the two major drivers of that are the size of the dividends coming up from the operating companies versus the amount of share repurchases that go out.
Okay. There was no special dividend. If I recall, in the first quarter, did you take a special dividend up and there was nothing in the quarter?
No, we had a dividend in the quarter. Let me clarify what is going on with the dividends. I think in our 10-K at the beginning of the year, of course, we had indicated that we could take dividends up without special authorization of something around $3.5 billion, was my recollection. Somebody's going to look that up for me. What we also said during the first or second quarter, I'm not sure which one, was that based on the capital positions of the operating companies, which had gotten very robust, we were actually going to go to our regulators and say that we wanted to take capital out of the operating companies at a higher level than that would have indicated and get their authority to do that, which we did. We did that in the first quarter.
When you do that, it also changes the dynamics going forward of what constitutes a normal dividend versus what constitutes a special dividend. In each of the quarters, we have gone to our regulators and said, "We'd like to take X out," and in each of the quarters, we've been given permission to do that. It's all in conjunction with a strategy, as Jay has talked about, of always right-sizing the capital to place. You've heard me say before that we manage to a certain level of capital in the operating companies to support the double A ratings, and that's all we're doing. Our regulators understand that, the rating agencies understand it, and it's just what's flowing through right now.
Okay. I guess just to be clear then, when we take your end of period holding company liquidity and all we get is kind of what net income is, and then we know what you're buying back and we know what you're dividending to shareholders. What you're saying is there is an amount that you guys are going to the regulators intra-quarter and requesting to take up from the subsidiaries, and that doesn't necessarily correlate one-to-one with net income.
That's correct. We take monies out of the operating companies each quarter because we're making money in the operating companies each quarter. If you take a big picture view of what we've been saying for the year, our guidance is for share repurchases of $4.5 billion-$5 billion. You look at what our year-to-date net income is, and that certainly doesn't equate to a picture for the year of $4.5 billion-$5 billion. Those two numbers alone are showing that we're taking capital out of the place.
Okay. I guess to push one more level on that then is, therefore the payout can sustainably be ahead of net income for the foreseeable future?
You get to a point where you've taken out the excess capital, you eventually get to a steady state where the payout is going to be based upon what is your net income, what are your capital needs in the company, and what are your various targets for holding company liquidity and debt and everything else. Eventually you do get to a steady state.
Greg, we had talked previously, you all know that we're going to stop the practice of giving guidance when we get to next year. One of the things I think that we probably do have to provide some continuing visibility on is our projected share repurchases, because to exactly your point, you really can't independently make an assessment of what our capital position is and what's available. My guess is that when we get to the fourth quarter and we've got our plans and budgets all squared away for next year, notwithstanding that we're really not going to speak to EPS, we will give you a robust understanding of what our capital management plans are.
Okay, great. That's helpful. Thanks. The other one on capital management is just the dividend policy. Could you just remind us how you set that? Because the dollars allocated to dividends obviously been flat for a while the share count's gone down. Just how do you think about that going forward? How have you thought about historically, whatever you're comfortable on?
We take a look at what the dividend yields are for comparable companies in the property casualty space. We look at payout ratios. We recognize that we're in a business where the wind blows and the earth shakes, so we take that into account in the payouts. What we've been fortunate enough to be able to do over the last couple of years is look at, well, we have a very solid earnings stream, and we've paid out roughly $700 million. If you look at the dollar amount in each one of the years, you're absolutely right, our share count has gone down as a result of the repurchases. We've been increasing the dividend to bring the dollar payout roughly back up to the same level, which has basically kept the payout ratio and the yields very competitive.
Okay. Totally shifting gears from capital management. Can you comment at all on the workers' comp pricing environment? We're hearing stuff out of Florida and others. What are you seeing at the margin there?
Sure. This is Brian. If you look at our numbers, you'll see that we've been growing our workers' comp business, and actually we've been growing it for about 10 years. We've been growing it fairly gradually. In the aggregate for our book, we feel very good. While the economy's clearly impacted the exposures and payroll changes in some places have been dramatic, we've been adding accounts and seeing some moderate growth in the line. The bottom line is comp, maybe more than any other product we do, is a state-by-state, industry-by-industry, account size by account size kind of dynamic. We look at a very granular level and really believe it's the classic risk selection game, and feel good about it.
One thing I would comment on, maybe I'll throw it to William Cunningham, is the AM Best data on comp got a good bit of play, that's just a slice of the industry, and you need to understand what it is that's looking at.
Right. The AM Best study that Brian's referencing was a composite of the state fund business and the monoline workers' comp market business. As you look at that, many of those state funds are markets of last resort. Obviously, as we talk about selection and the things that we have in place, selection is not possible when it's a market of last resort and the pricing is not always appropriate. As we look at the profile of our book and our results are much different than that.
Yeah. Greg, we would agree that there are certain states and certain industries where we'd be very concerned with comp and be pulling back. In the aggregate, we feel good about our book.
Okay. Are rates at the margin, is pricing going up or down for that line?
Again, it's a state by state. For competitive reasons, aren't going to get into state specifics. I would say on an overall basis, our workers' comp pricing has been fairly consistent with approach we've been taking over the last few quarters. We have not seen a dramatic change.
No shift. Okay. Last one for me on direct auto.
What is it, Greg?
Just on direct auto, just because there's a lot of money that's being spent in that direction. Just so I understand, you're pricing that business the same to both agents as you are to the direct customers. There's no benefit for going direct to Travelers at this point, to the customer?
Relative, yeah. Relative to the agent. They get the same price whether they go to the agent or they go to come direct. It's the only pricing track we have. We have no experience in the direct channel, so we don't have the ability to create a pricing track yet based on experience.
Okay. I'll follow up with that more offline. Thanks, guys.
We have time for one more questioner.
Our last question comes from the line of John Hall from Wells Fargo Securities. Please proceed.
Great. Thanks very much. I'll have only two questions. I'll give them straight up. The first one has to do with whether you're utilizing any enhanced commission structure as you go after any new business, whether that's part of your program on the commercial line side. The second one has to do with the comment that Brian made. You talked about the economy, or if and when the economy improving it having a very positive effect on exposure and potentially premium trends. I was wondering if that notion is factoring into your retention strategy, and how so?
First on the commission side, nothing unusual. We pay base commissions and we pay supplemental-
We don't have any. I'm asking do-
Yes.
We don't have any specials on with respect to growing or anything else.
What we're doing now is not a change from what we've-
Unchanged from what we've been doing.
Our fundamental programs there over the last several years have remained very constant, both in structure and in amount.
We will obviously, through predominantly our fixed value-based commission, differentiate one producer, one agency versus another. Those that grow more will have a higher fixed value-based supplement than others. There's nothing, no special arm with respect to growing this month or anything like that.
Right.
Okay, you're not utilizing any of the platform savings you talked about in that direction?
In small commercial, I don't think we changed the commission structure with Express, did we, when we put it out? We did for a little, maybe years ago, but not currently.
Yeah.
Yeah. No is the answer.
No.
Okay, great.
The second question was?
On the economy and retention.
Yeah. It's been an interesting phenomena that we speak about a lot, and we've talked about it for literally years, that the renewal market has had a remarkable stability to it that actually is consistent with those of us who were in the business in the '90s experienced then. There is a robust competitive environment for new business, but yet the renewal book seems to be, and this is universal by the way, it's not just us. You look at any carrier of any quality carrier, you speak to agents, you speak to brokers, they will all comment, the ones we speak to, will observe that the renewal market has a remarkable stability. Customers are just happy. Most customers are happy just staying where they are.
In effect, as long as the premium doesn't go up, they're happy with the broker, they're happy with the carrier, and the experience is good, and the business just is there.
Yeah. On the exposure change, if your comment is getting at is there anything, let me answer it directly. There's nothing explicitly in how we're evaluating business or pricing business that contemplates a growth in exposure and therefore a different profitability dynamic on the account going forward. Obviously, we're hopeful in these conditions that if you've got accounts and you feel comfortable with them today, as the economy expands, they're going to get bigger and that'll be a better thing. We're not building it into the economics of how we're viewing the trade.
The notion in today's environment, I'm sure like most of your employers, the willingness of your employers to take on added costs that aren't justified with underlying activity is a substantive issue for every business, and it's a substantive issue for our customers. It's not particularly easy nor well-received to go to a customer that's getting through a challenging economic environment and suggest that we just raise the price for insurance. Our retention strategy is molded by the environment we're in, by the economic environment we're in, and a realization that there are real life customers on the other side of the transaction. You've got to be responsive to the overall economic environment and manage the business for the long term. That really defines the approach.
Great. That answers the question. Thank you.
Ms. Nawi, I will now turn the call back to you.
Great. Well, thank you all for listening. If you have any follow-up questions, please contact myself or Andrew Hersom in the investor relations department. Thanks very much and have a good day.
Ladies and gentlemen, that does conclude the conference call for today. We thank you for your participation and ask that you please disconnect your lines.