Good morning, ladies and gentlemen. Welcome to the third quarter results teleconference 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 conference is being recorded on October 22nd, 2019. At this time, I would like to turn the conference over to Ms. Abbe Goldstein, Senior Vice President of Investor Relations. Ms. Goldstein, you may begin.
Thank you. Good morning and welcome to Travelers' discussion of our third quarter 2019 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 travelers.com under the Investors section. Speaking today will be Alan Schnitzer, Chairman and CEO, Dan Frey, Chief Financial Officer, and our three segment presidents, Greg Toczydlowski of Business Insurance, Thomas Kunkel of Bond & Specialty Insurance, and Michael Klein of Personal Insurance. 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. Then we will take questions. Before I turn the call over to Alan, 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 expressed or implied in the forward-looking statements due to a variety of factors. These factors are described under the forward-looking statements 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 available in the Investors section on our website. Now I'd like to turn the call over to Alan Schnitzer.
Thank you, Abbe. Good morning, everyone, and thank you for joining us today. This morning, we reported third quarter net income of $396 million or $1.50 per diluted share. Our core income was $378 million or $1.43 per diluted share. Our results this quarter were impacted by net unfavorable prior year reserve development in Business Insurance. Continued favorable development in the workers' comp line was more than offset by a strengthening of asbestos-related reserves and reserves for the GL and commercial auto lines. As I shared at an industry conference last month, GL and commercial auto were impacted again this quarter by a tort environment that has deteriorated beyond our elevated expectations. As I'll explain in a minute, we remain confident that this is an industry-wide issue and that we're responding appropriately to recently emerging data.
Underwriting income in the quarter benefited from record high net earned premium of $7.2 billion, driven by higher pricing and stronger retention over the last four quarters. Also contributing was the continued execution of our strategy to invest in the capabilities we believe are required for success, given the forces of change we've previously identified as impacting our industry. In addition to investments in talent, technology, and workflow, this includes progress in terms of our strategic focus on productivity and efficiency, as reflected in our sub-30 expense ratio for both the quarter and year-to-date. As you've heard us say, improved operating leverage gives us the flexibility to invest further in our strategic priorities, let the benefit fall to the bottom line, and/or be more competitive on pricing without compromising our return objectives.
In terms of the underlying underwriting margin, as you can see on page four of our earnings presentation, the underlying combined ratio was solid at 94.1% but up 1.1 points over the prior year quarter. Favorable items impacting the underlying result included improvement in our workers' comp loss ratio and, as I mentioned, continued excellent work improving operating leverage. More than offsetting those items were higher non-catastrophe weather-related losses and headwinds from a more challenging tort environment. Dan will provide more detail on the pieces, but I'll spend a minute on the tort environment. The aggregate impact of the adjustments we made this quarter to reflect our latest view of the tort environment, together with those from the second quarter of this year and the fourth quarter of last year, added about two-thirds of a point to the underlying combined ratio on a year-to-date basis.
To give you some perspective on the loss environment, on page 21 of the earnings presentation, we've included a chart of ISO data showing a measure of commercial auto bodily injury severity for the industry over the past four years. The full year 2018 data just recently became available. As you can see, during 2018, the observed average paid claim cost shifted significantly compared to the prior years as well as throughout the year. This view of industry data for other commercial liability coverages is not yet available for 2018, but we believe that the more aggressive tort environment is to one degree or another impacting those as well. While the chart reflects industry data and is not Travelers specific, the shaded area of the graph illustrates the challenge we've been addressing as part of our current and prior year reserving process over the past four quarters.
Developing a stable view of ultimate losses in this case is made difficult by the magnitude of the shift in the data. Combined with the fact that it is so recent, the long tail nature of the exposure, and a lengthening of the claim development pattern, which is difficult to assess as it emerges. At the heart of the issue is a higher and more aggressive level of attorney involvement on claims. Taking a step back, we'll manage through this, but the bigger issue for all of us is that the broken tort system imposes a tort tax across society. According to a recent study by the Institute for Legal Reform, tort costs amount to approximately 2.3% of U.S. GDP, or more than $3,300 per year U.S. household, and considerably more than that in some states.
There's also evidence that some claimants can be disadvantaged. According to a study by the Insurance Research Council, private passenger claimants with attorneys receive less on average in settlement from insurers after deducting legal and other related costs as compared to claimants without attorneys. On average, it also takes claimants with attorneys longer to receive payments. Tort reform has been a public policy focus for us, and we're stepping up our efforts. Turning to the top line, our production results were again excellent in the quarter. We grew net written premiums by 7% to $7.6 billion, a record increase of more than $500 million, with each of our business segments contributing. Our premium growth reflects high levels of retention and improving renewal premium change broadly across the portfolio, as well as a higher level of new business.
We're growing our business with confidence in terms of accounts, geographies, and industries that we know well. In Business Insurance, net written premiums increased 7% as we achieved renewal premium change of 7.4%, including renewal rate change of 4.3%. In both cases, the highest levels in more than five years. At the same time, we maintained strong retention and grew new business. As you'll hear from Greg, new business in our national property business benefited from a level of disruption in the market as we picked up some accounts on our terms. New business in middle market benefited from success with our business center strategy, which you've heard us discuss. Across the board, new business levels benefited from stronger new business pricing.
In Bond & Specialty Insurance, net written premiums increased by 13% with strong production across our management liability and surety businesses, including historically high domestic management liability retention with improving renewal premium change and record new business. In Personal Insurance, net written premiums increased by 7%, with agency auto up 3% and agency homeowners up 11%, both benefiting from strong production. Given the headwinds that the industry is facing from continued low interest rates, a challenging level of social inflation, and ongoing uncertainties surrounding weather-related losses, we will continue to seek higher prices and improve terms and conditions. You'll hear more shortly from Greg, Tom, and Michael about our segment results. To sum it up, we feel terrific that the strategy we laid out in 2017 is contributing to top-line success in each of our segments.
In terms of the bottom line, we would've hoped that the tort environment would not have deteriorated further. We're in a business where from time to time, loss costs are going to vary from our expectations. As we've explained, what's important is recognizing it quickly and taking action. We have a long track record of successfully managing our diversified businesses through challenging circumstances. With insight from leading data and analytics driving our execution, the best talent in the industry, and deep relationships with our agents and brokers, we'll continue to leverage the power of our franchise to deliver industry-leading results over time. With that, I'll turn it over to Dan.
Thank you, Alan. Core income for the third quarter was $378 million, down $309 million from the prior year quarter, and core ROE was 6.5%, down from 12%. Both measures reflect the elevated levels of general liability and commercial auto losses that Alan described, which impacted both prior year reserve development and the current accident year, as well as an elevated level of non-CAT weather losses. Our third quarter results include $241 million of pre-tax CAT losses, compared to $264 million in the prior year quarter. Prior year reserve development, which I will discuss further in a few minutes, added 4.1 points to the combined ratio, compared to favorable 0.2 points in last year's third quarter.
The consolidated underlying combined ratio of 94.1%, which excludes the impacts of CATs and PYD, increased by 1.1 points, driven by elevated general liability, commercial auto, and non-CAT weather losses, partially offset by favorable experience in workers' compensation. The adjustments we made this quarter to reflect our latest view of the tort environment added about half a point to the underlying combined ratio, including a catch-up related to the first and second quarters. In terms of weather, non-catastrophe weather-related losses added about two-thirds of a point to the quarter-over-quarter change in the consolidated underlying combined ratio. For context, total weather losses in the aggregate, CAT and non-CAT, were somewhat worse in the quarter than what we planned for, but year-to-date, total weather losses were well within the range of what we would consider normal.
The third quarter expense ratio of 29.5% was in line with recent periods and brings the year-to-date expense ratio to 29.8%, reflecting the terrific progress we've made in recent years. Both the third quarter and year-to-date expense ratios are about two points below where they were at this point in 2016. Our third quarter pre-tax underlying underwriting gain of $386 million declined by 14%. As the elevated losses in GL, commercial auto, and non-CAT weather were partially offset by the benefits of higher levels of earned premium and favorable experience in the workers' compensation line. After-tax net investment income decreased by $19 million from the prior year quarter to $528 million. After-tax fixed income NII increased by $17 million, as expected, and our non-fixed income NII, while strong, was below last year's result.
Fixed income NII benefited from an increase in average invested assets resulting from continued growth in net written premiums and slightly higher yields compared to a year ago. We expect after-tax fixed income NII in the fourth quarter to be about flat when compared to the fourth quarter of 2018, as the benefit from higher levels of invested assets will be largely offset by lower reinvestment rates. As we look ahead to 2020, we currently expect quarterly after-tax fixed income NII to be about $10 million-$15 million lower than our results for each of the first three quarters of 2019, as the impact of lower interest rates is expected to more than offset the benefit from higher levels of invested assets. Net unfavorable prior year reserve development was $294 million pre-tax compared to $14 million of net favorable PYD in the prior year quarter.
Bond & Specialty Insurance had net favorable PYD of $3 million. In Personal Insurance, favorable PYD of $19 million resulted from improvements in both domestic property and auto. Business Insurance experienced net unfavorable PYD of $316 million pre-tax, including a $220 million increase to asbestos reserves, a $134 million strengthening to general liability reserves, and a $114 million strengthening to commercial auto reserves, driven by similar factors to those we experienced in the second quarter. You may also notice that we have a small charge for environmental this quarter. As the number of claims and related reserves have decreased over time, we've now moved to a quarterly evaluation rather than the annual evaluation we had historically completed during the second quarter. These reserve developments were partially offset by favorability in the workers' compensation and property lines.
The asbestos reserve strengthening resulted from the completion of our annual asbestos review during the third quarter. As has been the case in recent years, while there was some slight improvement in several of our asbestos indicators, including a decrease in deaths from mesothelioma as a percentage of the population over time, our expectation had been for more of an improvement. This phenomenon has continued to drive periodic reserve strengthening for us and for others in the industry. As we've discussed before, we added a new catastrophe reinsurance treaty for 2019, providing coverage for PCS-designated events for which we incur $5 million or more in losses above an aggregate retention of $1.3 billion. Through September 30th, we've accumulated $1.2 billion towards the $1.3 billion retention. Should we reach $1.3 billion, 86% of the next $500 million of qualifying losses will be covered by the treaty.
Remember that for purposes of the treaty, hurricane and earthquake events have a cap of $250 million per occurrence. Let me take a minute to address the topic of our subrogation claims against PG&E for the California wildfires of 2017 and 2018. On September 22nd, PG&E and the Ad Hoc Subrogation Group, of which Travelers is a member, agreed to a proposed $11 billion settlement for subrogation claims against PG&E. Because of the uncertainties that remain regarding our ultimate realization of subrogation benefits from PG&E, we have not reflected any benefit in our financials as of September 30th. If PG&E ultimately emerges from bankruptcy pursuant to a plan of reorganization which includes the proposed settlement, we estimate that Travelers would recover between $400 million and $450 million pre-tax, which is net of expenses and net of amounts that would inure to the benefit of our reinsurance. Turning to capital management.
Operating cash flows for the quarter of $2 billion were again very strong. All our capital ratios were at or better than target levels, and we ended the quarter with holding company liquidity of approximately $1.5 billion. Recent decreases in interest rates have increased our net unrealized investment gain. As of September 30th, we had a net unrealized investment gain of $2.4 billion after tax. Adjusted book value per share, which excludes unrealized investment gains and losses, is now $90.09, 3% higher than at the beginning of the year. Consistent with our historical capital management strategy, our first objective for the capital we generate from earnings is to invest it back into the business where we think we can do so at attractive returns. Beyond that, we'll continue to return excess capital to shareholders.
Accordingly, this quarter, we returned nearly $600 million to our shareholders, comprising share repurchases of $375 million and dividends of $215 million. On a year-to-date basis, we have returned $1.8 billion of capital through dividends and share repurchases. As we grow premium volumes, with all else being equal, we'll need to retain more capital. To the extent we continue to grow, we would expect that over time, the amount of capital returned to shareholders relative to earnings will be somewhat less than it otherwise would have been. With that, I'll turn the microphone over to Greg.
Business Insurance produced segment income of $179 million and a combined ratio of 107%, both impacted by net unfavorable prior year reserve development of $316 million pre-tax, as Dan mentioned. The underlying combined ratio of 95.9% was a half a point higher than the prior year quarter. There are several moving pieces underneath that variance. We've included information on slide nine of the earnings presentation to break down the components of the change. I'll start with the notable items that increased the underlying combined ratio year-over-year. The first three are items that we've discussed with you previously, with the first of those being the half point impact from the lower earned premium due to the new CAT treaty. The remaining two items relate to actions we took last quarter and in the fourth quarter of last year related to the challenging trends in the tort environment.
The last of the unfavorable items relates to adjustments to our loss estimates resulting from an even further deterioration in the tort environment, as Alan and Dan have discussed. The impact of these latest adjustments was about a point, with about a third of that relating to the current quarter and the remaining two-thirds of that being re-estimation of the first and second quarters of the year. Similar to last quarter, this quarter's adjustment included both commercial auto and general liability. I'll note that unlike commercial auto, where returns are clearly inadequate, the returns in the general liability line, both primary and excess, are much healthier for us. Turning to the favorable year-over-year items. First, there is about a point of favorability relating to adjusting our workers' compensation losses. About half of that impact relates to re-estimating the first two quarters of 2019.
Second, our expense ratio benefited by about a half a point due to the continued successful execution of our strategy to improve productivity and efficiency. Importantly, we continue to achieve this expense leverage while investing in strategic initiatives that position us for the future. In addition to these individual favorable items, there was about a half a point of favorable impact from other items that caused normal variability from period to period. Before turning to the top line, I'll make a comment on weather-related losses. While non-CAT weather losses in the quarter were elevated relative to our expectations, it was not highlighted in the quarter-over-quarter underlying combined ratio comparison, as they were similarly elevated in the prior year quarter. Catastrophes for the quarter were below the long-term average in Business Insurance. All in, weather losses for the segment were favorable to what we would consider a normal level.
As for the top line, net written premiums for the quarter were up 7% over the prior year, driven by strong underlying production results, with renewal premium change being the biggest driver. In terms of domestic production, we continued to execute on our strategy to improve the margins in the book while retaining our high-quality portfolio. We achieved strong renewal premium change of 7.4% in the quarter, including renewal rate change of 4.3%. The renewal rate change was up seven-tenths of a point from the second quarter and more than two and a half points from the third quarter of last year, demonstrating continued momentum in our execution even after the continued downward pressure in workers' comp pricing.
The higher rate levels are being achieved broadly across the book, with about three-quarters of our Middle Market accounts getting positive rate increases this quarter, which is up from about two-thirds in the third quarter of last year. While the commercial auto and property lines continue to lead the way, our general liability, both primary and excess, as well as commercial multi-peril lines, also saw meaningful increases. Importantly, at the same time as we achieve these price increases, we maintain strong retention. New business of $568 million was up 21% from the prior year quarter, driven by strong results in our National Property Business, which I'll touch on in a moment, as well as higher new business pricing and deliberate execution across the other markets.
As for the individual businesses, in Select, renewal premium change and renewal rate change both ticked up from the second quarter and year-over-year, while retention remained strong at 82%. New business was up 5% over the prior year quarter, reflecting strategic investments in product development and ease of doing business. To that end, we're excited to have launched a completely redesigned small business owners policy product in three states during the quarter that includes industry-leading segmentation and a fast, easy quoting experience. The full rollout of this product will take some time, but we're encouraged with the early returns. In Middle Market, renewal premium change was 6.3%, with renewal rate change of 3.8%, up more than two points from the third quarter of last year, while retention remained high at 86%.
New business premiums of $308 million were up 15% from the prior year quarter, driven by higher new business pricing, as well as success in writing larger accounts during the quarter. While impacts from large accounts will be lumpy from time to time, we're pleased to see the results this quarter as further evidence that the Business Center Strategy that we've discussed with you is paying off by freeing up our local underwriters to spend more time on larger accounts. I'll touch on our National Property Business, which has had two consecutive quarters with substantial year-over-year increases in new business. As I mentioned last quarter, in this space, we are seeing more opportunities given the level of disruption in the marketplace, and we're selectively writing attractive accounts at our prices and on our terms and conditions.
In point, for the new accounts of size that we wrote in the second and third quarters, we'd previously written or quoted 80% of them and let them go when we couldn't write them at prices and on terms and conditions that met our discipline standards. To close, while we're managing through a changing environment, we couldn't be more pleased with our insight and execution. As Alan said, given the low interest rate environment, continued uncertainty around weather-related losses, and an upward pressure from the tort environment, we'll continue to seek rate and use all other available levers to meet our return objectives. With that, I'll turn the call over to Tom.
Thanks, Greg. Bond & Specialty delivered another quarter of strong returns and growth. Segment income was $139 million, a decrease of $57 million from the prior year quarter, primarily due to a lower level of net favorable prior year reserve development, and to a lesser extent, modestly higher management liability loss estimates for the current year. The underlying combined ratio increased 5.3 points from the prior year quarter. As you can see on slide 14, about three points of that is a third quarter adjustment for management liability coverages, about two points of which reflects a re-estimation of losses from the first two quarters of 2019. Despite the increase, the combined and underlying combined ratios remain strong at 83.3% and 83.6% respectively. Net written premiums for the quarter were up 13%, with strong growth across all businesses.
In domestic management liability, we are pleased that retention remained at a historically high 90%, with renewal premium change higher at 4.8%. New business for the quarter was a record $67 million. Domestic surety growth reflected higher bonded contract sizes, and our international growth was primarily in our U.K. management liability businesses, including from new product offerings. Our production results reflect the profitability of our high-quality portfolio, strong field execution, and our strategic long-term investments in our suite of products and services, together with our commitment to thoughtful and disciplined underwriting and risk selection. As reflected in our improving renewal premium change, we are pursuing price increases for coverages where we are experiencing elevated levels of loss activity, and we will continue to do so. Bond & Specialty results remain strong, and we feel terrific about our ability to continue to deliver excellent results over time.
Now I'll turn it over to Michael to discuss Personal Insurance.
Thanks, Tom, and good morning, everyone. In Personal Insurance this quarter, we're pleased with our continued execution in the marketplace as we grew premium and policies in force in both our agency automobile and homeowners and other product lines, continued to achieve excellent profitability in agency automobile, and reported good results in agency homeowners and other, particularly considering a high level of non-catastrophe weather-related losses. The combined ratio for the quarter was 98%, slightly higher than the prior year quarter of 97.2%. For the segment, catastrophes and net favorable prior year reserve development were comparable to the prior year quarter, while the underlying combined ratio was higher by 1.1 points, primarily driven by higher non-catastrophe weather-related losses and the impact from the new catastrophe reinsurance treaty, partially offset by lower other loss activity. Net written premiums for the quarter grew 7%, with strong retention, renewal premium change, and new business.
Agency automobile delivered another strong quarterly performance with a combined ratio of 93%. This is our seventh consecutive quarterly combined ratio under 96. The underlying combined ratio for the quarter of 92.7% was comparable to the prior year quarter and consistent with an excellent year-to-date result of 92.8%, two points lower than the prior year. In agency homeowners and other, the third quarter combined ratio of 100.2% increased 1.7 points from the prior year quarter, driven primarily by a higher underlying combined ratio, partially offset by lower catastrophe losses and a favorable change in prior year reserve development. The underlying combined ratio for the quarter of 93.5% was five points higher than the prior year quarter, as approximately four points of higher non-catastrophe weather-related losses and one and a half point impact from the new catastrophe reinsurance treaty were partially offset by lower other loss activity.
The higher non-catastrophe weather-related losses this quarter are on top of an elevated level in the prior year quarter and are reflective of an increased number of events compared to historical experience. As a point of reference, data from NOAA, the National Oceanic and Atmospheric Administration, reflects approximately 8,500 severe weather events in the U.S. for the third quarter of 2019. That's a 28% increase when compared to the average third quarter for the period 2013 to 2018. As we've mentioned in the past, we've increased our loss expectations to reflect higher levels of loss activity and property, and that's reflected in the pricing gains we're achieving and our plans to further improve pricing and property going forward. Before I shift to discussing production, I'll remind you that looking ahead to the fourth quarter, there tends to be a good amount of seasonality in our combined ratio results.
On a relative basis, fourth quarter losses are typically higher than the full-year average for automobile and lower for homeowners and other. Shifting to quarterly production, agency automobile premiums grew 3% with modest growth in new business and policies in force, while retention remained strong at 84% and renewal premium change was 3.6%. We remain focused on actions to grow our auto business at returns that continue to meet our objectives and are encouraged by the increase in policies in force this quarter. Agency homeowners and other delivered another strong production quarter with retention of 86%, even as renewal premium change rose to 6.9%, up more than three points from the prior year quarter. This marks the seventh consecutive quarter where renewal premium change is higher than the corresponding prior year quarter.
Consistent with our objectives, new business for agency homeowners and other was up over the prior year quarter, with increases in both issued policies and average new business premiums, largely driven by our ongoing successful rollout of Quantum Home 2.0. We remain pleased with the quality of the new business we're writing. The profile of the business is consistent with our expectations, and most of the increase in issued policies results from higher quote volume at a consistent close rate. As we intended, sales and marketing activities associated with the rollout of Quantum Home 2.0, along with its granular pricing segmentation, customizable coverages, and ease of quoting, are leading contributors to higher new business volumes. Wrapping up, while profitability for the quarter was down slightly from the third quarter of 2018, we're pleased with our year-to-date results.
Segment income of $497 million and the combined ratio of 96.2% are both significantly improved relative to the prior year. Given our pricing actions in property and early signs of policy growth in auto, we remain confident in our ability to generate profitable growth going forward. The competitive advantage of a balanced and diversified portfolio and our focus on being a complete property and casualty solution for our agents and customers continues to position us well in the marketplace. Now I'll turn the call back over to Abbe.
Thank you very much. We're ready to take your questions.
At this time, I'd like to advise everyone, in order to ask a question, please press star, then the number 1 on your telephone keypad. Again, that is star, then the number 1 on your telephone keypad. If you wish to withdraw your question, you may press the pound key. Our first question comes from the line of Jay Gelb with Barclays. Your line is open.
Thanks, and good morning. I want to touch base on the liability claims inflation issue across GL, commercial auto, and D&O. I realize that this topic was discussed and essentially warned about at the Barclays conference last month. What I'm trying to get a perspective on is whether the company can say we have the all clear on reserve issues going forward, or there's still a risk of more reserve strengthening going forward, given this has been an issue in three out of the past four quarters.
Jay, good morning. It's Alan. Thanks for the question. I don't think that we or any company could ever give that kind of assurance. We take our best people, our best processes. We look at the data that we have, and we come up with our best estimate every quarter. When we look back over the last four quarters at the decisions that we've made and the estimates that we've published, we look at it and say, "Gee, based on the information we had at the time, we think we made the right call." We weren't too aggressive in any of our assumptions. We, in fact, picked close to the top end of the range in many of those cases. Again, this quarter, we look at the data, and we think we've done the right thing. That's one of the reasons why we wanted to provide slide 21.
You can see this is an industry-wide phenomenon, and you can see the extent of the shift, at least from an industry perspective. This isn't our data, but it's illustrative and relevant in thinking about our data. It's quite a significant shift, and one of the associated factors is this lengthening of the claim development pattern, and that's a difficult thing to assess as it emerges. With every quarter, we get a new data point. With every new data point, we get a different trend line, and we put our best estimate behind it. We continue to think running the business by the numbers, being very disciplined about that is the right way to do it. Jay, the answer is, we've got a long track record of managing that way over time.
We think we're running it the right way, and we think we've got the right estimate now. I don't think we or anybody could give you a guarantee.
Right. Essentially, you're saying you're as confident as you can be that this issue is addressed, right?
I'm as confident as I can be that it's our best estimate.
Right. Okay. Then separate topic with regard to capital return. I believe there was commentary in the prepared remarks around potentially less capital being returned as a percentage of earnings as the company's growth rate continues to accelerate. Would you be able to give us some parameters around that? Historically, it's been right around 100% of earnings being returned annually in the form of dividends and share buybacks. Would that number be something closer to 75% or should we think about another number?
Jay, it's Dan. I don't think we would target a number. I don't think you should think about it as a target, per se. That's going to be an outcome of the things that we always talk about, which is how much capital do we have? How much capital do we think we need to adequately support the business that we're writing and that we expect to add on a go-forward basis? What are all the investments we'll make in the business that we think can generate an adequate return? Then we'll look at what's left and return it. The comment was intended to
To raise awareness, as you say, from this point forward, as we continue to grow premium to support a business that's got a bigger premium base, you're going to need a bigger capital base. If you perpetually return to 100% of earnings, your capital base wouldn't change. It's simply to give you an indication that as we continue to grow, if all else remained equal, you should expect something lower than it otherwise wouldn't have been. We're actually not going to target a particular percentage of earnings that we're looking to return.
Jay, I just add two things to that. One, there's no change at all in our thought process or approach towards capital management. This is ordinary course and just a function of what we've been able to achieve in terms of growing the premium base. Two, one of the reasons why we can't give you a number is it's going to depend on what lines we're growing in and what the geography is, and it's just not linear math behind it. We couldn't actually give you that target.
Appreciate that. Thanks very much.
Thank you.
Your next question comes from the line of Elyse Greenspan with Wells Fargo. Your line is open.
Hi, good morning. My first question goes to your outlook for Business Insurance. Broadly, really unchanged. You guys are looking for margin improvement, which given the commentary in the queue, seems to be a function of price exceeding trend and some improvement in your international business. What I'm trying to get a sense of is what's the forward view that you guys have for price, then also for trend, can you give us a sense, and I know it's a compilation of all your different businesses, where trend sits today, then the outlook for that as we think about improvement coming into your margins from here in the Business Insurance segment.
Yeah, Elyse, in terms of the outlook for price, you see the price that we've written and that's going to earn in. We give you an outlook for RPC. It's those two components that are going to earn in over time. In terms of loss trend, we always talk about that in BI from a segment perspective, we showed last quarter that we had increased our view of loss trend to 4.5%. That's obviously on top of the current picks, which have been adjusted throughout the course of the year. It really is the math from those pieces.
Okay. If trend was 4.5% last quarter, it's somewhere within the vicinity of 5% right now?
No. The 4.5 was up a half a point from what we had historically talked about at four. Elyse, our caution is that the 4.5 is our long-term outlook on loss trend. In any given period, there's going to be some volatility in actual losses. That's sort of the math and the way we think about it over time.
Okay. My second question, within Business Insurance, you guys released reserves within workers' comp. You also lowered your loss picks there this quarter. I guess I'm just trying to get a little bit more color given the weaker pricing in comp combined with the higher loss trend you're now seeing in other businesses within that segment. What gave you guys the conviction to change your picks there? If you could also let us know the prior year development on comp in the quarter. Thank you.
Elyse, it's Dan Frey. I'll start thematically. What gives us the conviction to move on workers' comp is as we see results come in, again, there's a long-term assumption around what's going to happen to the costs to settle workers' comp claims. A lot of that's driven by medical cost inflation. We see in our data, and there's been discussion in the industry that medical cost inflation has continued to be pretty benign relative to our expectations. That drives the prior year release as another period of data comes by where again, the cost of loss settlement comes in favorable to what our expectation was. In this quarter, in round numbers, order of magnitude is around $100 million pre-tax, a little less than that. That flows through into what we're seeing in the current period as well.
Obviously, in the current period, you're reacting to both what are you seeing in terms of the number of claims coming in and those settlement trends. Workers' comp has been a pretty steady level of good news relative to our expectations, and that's why we adjusted both numbers.
Elyse, I'd just add, it's just ordinary course. We treated workers' comp this period like we've treated it every other period, and we didn't actually make a change to either our frequency trend or our severity trend. It's the impact of adjustments to prior year reserve that's rolling through the current year.
Okay. Thank you very much.
Your next question comes from the line of Mike Zaremski with Credit Suisse. Your line is open.
Hey, good morning. First question is on the catastrophe reinsurance treaty because I have a feeling it's going to trip us up in our models. I believe you said it's up to $1.2 billion, so it's very close to the $1.3 retention. You've reported $800 million of catastrophes, which is a separate definition. If we trip the $1.3, which it seems likely from a probability standpoint, does that start benefiting the underlying loss ratio in both the commercial and homeowner segments, or just if you can help us think about that?
Mike, it's Dan again. I think as we discussed the treaty from when we first talked about it, going back to the first quarter, it will depend on what weather actually comes in over the remainder of the year. If we have a treaty benefit, how the treaty benefit will be allocated back between what we would call major CATs versus minor CATs. If all the losses that come in from this point forward are a major CAT, and that's what causes us to trip the treaty, and then those are the recoveries, most of the benefit would be attributable to major CATs. If on the other hand, most of the weather that comes in is $10 million-$15 million PCS events that don't meet our major CAT threshold, that's where the money will go.
It really is too early to tell, and it will depend on what actual weather we experience in the fourth quarter.
Okay. It could benefit the underlying loss ratio as well?
That's correct.
Lastly, if I think about the past two to three years and what sticks out, top-line growth clearly has improved for the company. It's also coincided during a period of time of increasing non-CAT weather losses. You guys have said catastrophe levels are, you changed your view on that being a little higher, and now we're seeing kind of higher liability loss trends in the last year or so. Is this all telling us that top-line growth and maybe retention rates need to maybe fall a little in the coming year and take a step off the gas a little bit?
No, Mike, I don't think that's the way we think about it at all. We don't ring a bell and say grow or ring a bell and say, don't grow. We execute one account, one class of business at a time, and we do that with a return focus. We're out there executing on our accounts, trying to retain the ones we want on the prices and terms and conditions that we want. We're out there hustling on new business that we find attractive and that meets our return objectives. When the sum of those things adds up to growth, we'll grow, and when it doesn't, we won't. There's nothing about the growth that's contributed to the loss environment, and we are pretty good about reflecting the loss environment into our pricing models. The answer is, I don't think so.
The other thing I would say is the growth that we've seen is, and I don't want to diminish the strategic efforts that's led to new business. I think that's important, and it's been successful. A very big component of the growth that we've reported has come from a very strong foundation of retention and price on top of that. That's important to keep in mind as well.
Thank you.
Thank you.
Your next question comes from the line of Larry Greenberg with Janney Montgomery. Your line is open.
Good morning. Thank you. On commercial auto, I'm just trying to compare what you're experiencing with the issues that you had a while back in personal auto. I think then you talked about taking a higher loss pick or adjusting your loss pick for severity. The actual trend really hadn't changed in your view. If I'm hearing correctly, in this situation, we're talking about higher kind of loss pick foundations and then an accelerating trend on top of that. Is that fair?
We've definitely increased the base year, and we also, last quarter, as we told you, increased the loss trend from there. We have increased both in commercial auto. Is that responsive, Larry?
Yeah. Okay. That's great. I think Greg talked about pricing and terms and conditions. Can you talk about how terms and conditions might be changing in response to what's going on?
Sure, Larry, good morning. This is Greg. Most of the terms and conditions activity are happening in the national property space, and that certainly is leaking down into property across middle market. I think our underwriters start with the first thing is they look at the total insured value, making sure that we've got the right values on the properties. From there, we'll get into discussions with both distribution and the customer around co-insurance, and that includes both deductibles and sub-limits. You can think of sub-limits on both flood and earthquake. Those are some of the items that are in the marketplace right now, and our underwriters are very involved in pulling those levers.
Yeah, I was thinking more in terms of commercial auto.
Yes, Larry, in response to that, given the environmental nature of that, most of the fix to that is going to come from price. To some lesser degree, it'll come from some risk selection, and that's a factor as well. Then we'll calibrate our litigation strategy. Those would be primarily the three levers we'd use to address it.
Thank you.
Your next question comes from the line of Paul Newsome with Sandler O'Neill. Your line is open.
Good morning. Thanks for the call. I was hoping we could revisit sort of the view in commercial insurance between the relationship between pricing and retention. I guess I'm a little surprised at the strength of the retention given you are raising prices. I'm just wondering if that means that you're not stepping on the pricing as hard as you could, or if there's some other dynamic going on there.
Yeah. Paul, again, we're executing one account at a time. We're making the right decisions on price and retention, we think. I think one takeaway that we take from the retention is that there's certainly more room to go. The market's not pushing back on us, which I think is evidence that these really are environmental issues. We think the execution's terrific, and we think there's more room to go. I think also in terms of our customers and our distribution relationships, doing this thoughtfully, slowly, steadily, I think there's a lot of benefit to that. That's part of our execution strategy.
Paul, one thing I would add, this is Greg, is those are both headline numbers, both the rate and the retention. Our underwriters are equipped with very granular segmentation. We spend more time watching the mix distribution of some of the distressed accounts and the healthy accounts than we do on the aggregate numbers. We're pleased with that execution when we look at that segmentation.
Separately, I was wondering as we look at the expense ratio, if there's any impact this quarter from change in contingent commissions that would affect either this quarter or maybe even next year, the level of the expense ratio.
Paul, it's Dan. I'd say across the place, generally not. It's been pretty consistent, contingent commission is a relatively small component of the overall commission mix to begin with.
Great. Thank you very much.
Our next question comes from the line of David Motemaden with Evercore ISI. Your line is open.
Hi. Thanks for taking the question. Just to follow up on the changes that were made in general liability. Are the changes really still commercial auto-related excess, or has it spread to other parts of GL? If you could talk a bit about the changes you made to each portion of that.
David, it continues to be both. When, again, I think the right way to look at this is broadly over the last years we've been addressing this issue. If you look at it in that context, commercial auto and the excess GL where the underlying event is auto-related continues to be the majority of it. GL has been a not insignificant factor, and in fact, in this quarter was a bigger factor than commercial auto.
Got it. Okay. Just in terms of just putting numbers around it, a peer had come out and said they were seeing severity on the GL side up in the high single digits. Just wondering, is that sort of what you guys are seeing and is that what you guys are assuming now going forward in your loss picks?
We've just never given the severity at that granular level. I don't think we're inclined to do that. We stop at the segment level where we've given you the 4.5%. I also think there can be some confusion when we're talking quickly about loss trend, because ultimately we're trying to get to a loss pick, and there's two things that contribute to that. One is base year changes, and two is a change in loss trend. As I said earlier, over the course of this year, as it relates to liability lines, we've raised both the initial loss pick, which we refer to as base year, and the loss trend. Those are both up.
Got it. Okay. Just to follow up, the slide in the back, that would imply that what you guys are seeing is more severity-related. Are you also seeing just a spike in frequency from higher attorney involvement as well?
No, it's definitely more of a severity issue than a frequency issue.
Great. Thank you.
Thank you.
Your next question comes from the line of Josh Shanker with Deutsche Bank. Your line is open.
Yeah, good morning, everyone. Thank you for taking my call. I wonder if there's any way of getting to a little more granularity on the tort issues. As you've acknowledged you're experiencing, you do look at your competitors' numbers, and they're not. Is there something going on in the account size that's different between large, middle market, and small accounts on the tort severity side that might make Travelers' book look different than some others?
Josh, my instinct is no. That's one of the reasons we shared the industry data is we believe this is environmental. We don't think there's anything about our book of business that makes us more susceptible to these types of claims. You turn on the TV, you see attorney advertising. The rate of attorney involvement and the aggressive behavior is up, I think that's going to impact anybody writing liability coverages. We don't think there's any reason to think that our book of business is more susceptible than anybody else's.
Do you have anything you can add to the granularity where you're seeing the biggest tort severity issues from your own experience?
Josh, it's broadly across geographies, business, and accounts. It's always hard to generalize broadly in a sentence or two because this is a complex issue, and we've spent hundreds of hours on this internally. It's always difficult to try to summarize in a sentence or two. Broadly speaking, it's probably groups of homogeneous type claims where a potential plaintiff can be reached by television commercial. Whether that's commercial auto accidents or slips and falls or things like that. Once you get into the bigger stuff, you tend to have cases that are and always have been associated with claimants being represented and aggressive litigation. Again, that's a very high-level summary of a complex issue.
Elyse kind of asked it, but I just want to The workers' comp reserve release. Is third quarter typically a quarter where you do a comprehensive workers' comp reserve analysis? Should we expect typically that in the third quarter we could see resets of picks typically?
Josh, it's Dan. Not really. Comp is something that we have enough data and enough information that's in the regular flow that we're assessing work comp reserve levels every quarter, and I've commented on work comp and PYD in earlier quarters this year.
Okay. Thank you for the answers.
Thank you.
Your next question comes from the line of Brian Meredith with UBS. Your line is open.
Thanks. Just a couple quick ones here. First one, if I strip out some of the noise in the quarter on Business Insurance, it looks like the underlying loss ratio kind of improved, call it by 40 to 50 basis points. Is that a kind of fair representation of what kind of earned rate right now is coming through in excess of trend?
That's a Business Insurance question?
Yeah, Business Insurance. Sorry, yeah. Just making all the adjustments that you put in the slide. It looks like it's about 40 to 50 basis points of improvement on a year-over-year basis.
Brian, we're looking at the math. Yeah, I don't think it's quite so easy to look at it that way. You're getting to a very narrow view of rate versus loss trend, I understand why that's so appealing to look at. The fact is there's a lot of things that go into margin, whether it's mix or reinsurance or claims handling or risk. There's all sorts of things that go into that. So I think it'd be really hard to look at rate versus loss trend and think you've captured it.
Gotcha. Then just one other quick one here. As I look at your outlook for RPC, for the Business Insurance here over the next 12 months, what is your assumption with respect to exposure in that figure? Do you expect it to kind of moderate here? Be the same?
Yeah, we just don't break that out. There's a lot of estimation that goes into that number, and we have a view on the pieces, but to imply that we've got that level of granularity out a year, that would be a false impression.
Great. Thanks.
Thank you.
Your next question comes from the line of Michael Phillips with Morgan Stanley. Your line is open.
Thank you, and thanks for the time here. I guess I'll take the pressure off on Business Insurance guys, switch over to Personal, and specifically Personal Auto. Your outlook says for margins that are relatively in line with last year. I guess could you talk about what you're seeing there in pricing and loss trends? We're certainly hearing loss trends on the rise in some segments from competitors. Kind of want to see what you're thinking there and why we'd expect it to be flat next year.
Sure Michael, this is Michael Klein. I would say, we see the same industry data you're referring to and certainly are keeping an eye on severity trends both in physical damage and in bodily injury. As we've talked about this year, we've tended to see frequency going the other way and offsetting some of those severity increases. Our view of loss trend heading into the outlook is fairly consistent with the trends we've been seeing. The outlook includes an assumption that pricing will continue to moderate in auto, and that's consistent with our strategy to continue to work to return to growing policies in force. The broadly consistent outlook has a little bit less earned price in it than we've been running. Again, a consistent view of loss trend.
Lastly, I would just remind you that sort of rolling 4 quarters, we do expect a higher combined ratio in the fourth quarter than we saw in the fourth quarter of 2018, partly driven by normal seasonality and driven by the fact that we had an unusually good quarter in the fourth quarter of 2018, as well as some prior quarter good guy catch up in that quarter. It's a tough comparison, fourth quarter to fourth quarter.
Okay, great. No, thank you for that detail. I guess second question on Bond & Specialty. What are you seeing in the management liability and maybe some more granular details of what's driving that increase there?
Sure. Thanks for the question, Michael. If you look back at where we started reserving at the beginning of 2019, there were definitely some elements of normal loss trend versus rate that impacted how we were reserving there. When you look at that third quarter re-estimation we did, that was largely driven by an increase in frequency having to do with things like hashtag #MeToo, sexual harassment, and to a lesser degree, securities class action suits.
That's really what we've seen driving any movement that we've had in management liability there. By the way, I should point out that where we're having elevated loss trending, we're definitely pursuing price in the marketplace. I think you can see that with publicly traded companies, and you can see it in various other parts of the marketplace that are impacted by the scenarios I discussed.
Okay, great. Thank you very much for the color there.
Thank you.
Your next question comes from Ryan Tunis with Autonomous Research. Your line is open.
Hey, thanks. Just had a couple quick technical ones and then a broader one for Alan. First of all, just trying to size this. Within multi-peril, what % of the claims dollars are usually for liability type of stuff versus property?
Yeah, I don't know that we have that at our fingertips or that we vetted that number, Ryan.
Got you. That's fine. The other one I had was just, it seems like part of this is auto claims and excess umbrella, those type of things. Can we get a sense for maybe what % of overall casualty claims outside of workers' comp or something like that can be tied back to something auto-related? Is there some way that you think about that? Just trying to think about the auto exposure outside of what you explicitly call commercial auto.
Yeah, we definitely don't have that breakout at our fingertips here, Ryan. We can look for that and try to figure out whether we can share that outside of this call. We definitely don't have that in a vetted way that I think we'd be comfortable sharing.
Okay. No, fair enough. Then again, Alan, I guess this is sort of a broad one, but this is, I guess, the third quarter of the last four where we've had, I guess, an increase in the loss picks. What would you point us to this quarter that would give us more confidence that that's unlikely to happen going forward? What are you looking at that gives you more confidence, it should give us more confidence?
Yeah. Ryan, I think that was the first question of the call. I wish I could stand up and guarantee that we've got it. This will never happen again. Obviously, I can't do that. I don't think anyone can do it. I point to the magnitude of the shift. This is very recent information in a long tail line. The lengthening of the claim development patterns really does complicate our ability to assess it. It's a complicated process, and it's taken us a couple of quarters to get to where we are. We're going to get another quarter's worth of data in a couple of months here. We'll see where that comes in. There is an incremental level of confidence. It comes from the fact that we've now got a couple of quarters strung together and a view on what a new trend line might be.
That gives us a little bit more comfort. I can't tell you now where the next dot was going to come in on the chart or what that's going to do to the trend line. It's data we haven't seen yet. I can tell you that we now have four quarters of data and a new trend line. We certainly reacted to that. We're confident that we're taking the right swing at it. I don't know what I could tell you beyond that.
You'd say that the 2019 picks are now relatively reflective of some of the more seasoned accident years at this point.
Yes. For sure, it would be more reflective of the more seasoned accident years. You've got the 2017 and 2018 accident years, which I think the 2017 accident year is, on commercial auto, less than 50% paid. The GL would be even less than that, and the 2018 accident year would be less than that on both. We've got a couple of recent accident years that are still developing. The older accident years, yes, as they season, we certainly feel much better about those. As I said, we now have 4 quarters where it's not just another data point, it's another line. We've reflected that in a way that we feel is very responsible in terms of coming up with our best estimate. I can't, and nobody could tell you what the next data point's going to look like.
Okay. Thanks. I'll leave it there.
Thank you.
Our last question comes from the line of Sean Reitenbach with KBW. Your line is open.
Hello. For PG&E, would any benefit from there come through operating income?
Yeah, Ryan, it's Dan. It would. It would come through the same way we would record any prior year reserve development. It'll be in core income. It won't affect the underlying combined ratio because it relates to prior accident years.
Okay. Although it's still very strong, the retention rate slipped by a point 3Q. Would you expect that to tick down any further as a result of either any kind of non-renewing any business that you've had trouble with or pushing for more rate and losing anything there?
I would say the retention rates remain very strong by historical standards, particularly in an environment where we're getting rate at the level that we're getting it. We feel terrific about the retention. We don't project where that's going, so I don't have a comment on that. I will say that we like our book of business, and our objective would be to keep a lot of it.
All right. Thank you very much.
Thank you.
There are no further questions at this time. I now turn the call back to Ms. Abbe Goldstein.
Thank you very much for joining us this morning. As always, if there's any follow-up, please feel free to reach out to Investor Relations. Thanks, and have a good day.
This concludes today's conference call. Thank you for your participation. You may now disconnect.