旅行家集團 (TRV) 2020 Q3 法說會逐字稿

Good morning, ladies and gentlemen. Welcome to the third quarter results teleconference for Travelers. (Operator Instructions)

As a reminder, this conference is being recorded on October 20, 2020. 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 2020 results. We released our press release, financial supplement and webcast presentation 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, CFO, and our 3 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, and then we will take questions.

Before I turn the call over to Alan, I'd like to draw your attention to the explanatory note included at the end of the webcast presentation. Our presentation today includes forward-looking statements. The company cautions investors that any forward-looking statements involve 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 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. And 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. We are very pleased to report third quarter core income of $798 million or $3.12 per diluted share and core return on equity of 13.5%. Our bottom line result this quarter reflects strong underlying underwriting income, resulting from record net earned premium of $7.4 billion and an underlying combined ratio that improved 2.6 points to a strong 91.5%.

We're pleased with the underwriting results in all 3 segments with improved underlying profitability in both business insurance and personal insurance. In Business Insurance, the underlying result improved due to margin expansion as earned rate exceeded loss trends. In Personal Insurance, the benefits from lower frequency in the auto business more than offset higher levels of non-catastrophe weather and wildfire losses. In Bond & Specialty Insurance, the underlying combined ratio was elevated, consistent with the outlook we shared on the call last quarter. I'll note that the combined and underlying combined ratios were still under 90%, generating a solid return in a challenging environment.

Profitability in all 3 segments continues to reflect the benefit of our strategic focus on productivity and efficiency resulting in a [subminus 30%] consolidated expense ratio. Core income for the quarter also included catastrophe losses of $397 million pretax, which were meaningfully above the 10-year average for quarter growth.

I want to acknowledge the devastation caused by recent catastrophes. We hope for a quick recovery for all of those who have been impacted. I also want to express my gratitude to our dedicated claims team for taking care of our customers during these extraordinary times. Of the 100,000 or so claim notices we've received so far this year arising out of a record number of PCS catastrophes in the U.S., our claims team has met our objective of closing more than 90% of the claims within 30 days. A quick resolution results in a better experience for our customers, and a more efficient outcome for us. I also want to acknowledge the actions our underwriting and risk management teams have taken over recent years to manage our exposure to weather volatility. In addition to property cat aggregate treaty, which has mitigated our losses, the recent actions to improve the balance of risk and reward, meaningfully reduced our exposure to wildfires.

To illustrate the point in the areas impacted by the 5 costliest California wildfires this season our exposure is about a 1/3 lower than it was 2 years ago.

Turning to our investment portfolio. This quarter, we again benefited from our well-defined and consistent investment philosophy. With our high-quality investment portfolio, generating net investment income of $566 million after tax. Lastly, before I turn to the top line, I'll share that the uncertainty surrounding business interruption claims continues to result favorably and consistent with our expectations, but we remain confident on that front. In terms of the top line and production, we continue to generate strong results. Net written premiums in the quarter grew by 3%, driven by strong renewal rate change and retention in all 3 segments.

In our commercial businesses, exposure change on renewed accounts was only modestly negative for both the quarter and year-to-date compared to a much more significant reduction in economic activity. We believe that in addition to generating a better underwriting result, our high-quality portfolio of accounts is more resilient to economic hardship.

In Business Insurance, we achieved record renewal rate change of 8.2%, 4 points higher than the prior year quarter, while retention remains strong. We achieved higher renewal rate change year-over-year and sequentially in each of our lines of business other than workers' compensation.

In Bond & Specialty Insurance, net written premiums increased by 4% as renewal premium change in our domestic management liability business rose to 8.1%, including record renewal rate change, while retention remained at an historical high.

In Personal Insurance, net written premiums increased by 8%, driven by strong retention and new business in both Agency Auto and Agency Homeowners. In our Agency Homeowners business, we achieved renewal premium change of 8.2%, its highest level since 2014. Across all of our businesses, we've made good progress achieving rate gains in managing other levers of profitability to improve the outlook for returns in those lines that need it, and we'll continue to execute to meet our return objectives. For all the reasons we've discussed previously, from the loss environment to the interest rate environment, we expect continued momentum in the marketplace. Notwithstanding our focus on successfully managing through the pandemic and addressing other headwinds impacting the industry. It's important to note that we haven't been distracted from pursuing our strategic agenda. We remain focused on leveraging our scale and resources to continue to invest and innovate. As we've said before, we believe the winners in our industry will be those with deep domain expertise that can deliver industry-leading results while innovating successfully on top of the foundation of excellence.

From a position of strength, we continue to focus our efforts on extending our advantage in risk expertise, providing great experiences and improving productivity and efficiency. In our commercial businesses, we continue to make progress in digitizing virtually every aspect of the value chain. While at the same time, enhancing our advanced analytics. Just as one example, our BOP 2.0 small commercial product, which we launched in 2019 benefit from both. In the states in which we've rolled it out, we've seen about a 15% increase in both submissions and new business premiums. This product uses AI and third-party data to improve underwriting segmentation, operational efficiency and the agent experience.

At that point, the artificial intelligence eases the burden on the agent and has resulted in a substantial improvement in classification accuracy.

In personal insurance, we're balancing sophisticated total account solutions with streamlined agent and customer experiences. For example, we've completely redesigned the experience of our IntelliDrive auto telematics offering and introduced a distraction reading. We rolled this out in 9 states during the second and third quarters and have plans to launch in additional 10 states in the fourth quarter. We're observing a nearly 30% increase in the rate of adoption for IntelliDrive and have received strong agent feedback. Also in the fourth quarter, we're rolling out an enhanced customer self-service tools in a new mobile app.

In our claim organization, we're advancing the rollout of virtual end-to-end claim service tools, embracing the pandemic-driven trends at accelerated digital adoption by individuals and businesses. Customer and agent satisfaction are up, while payout discipline remains strong. To sum it up, we've achieved better performance in the face of a pandemic and a challenging underwriting environment. It reflects the importance of a strong underwriting culture, the benefit of data and analytics and the franchise value we offer to our customers and distribution partners.

In all of that, together with our highly engaged and talented workforce, we're confident that we're well positioned to capitalize on opportunities as the economy continues to reopen. And with that, I'll turn the call over to Dan.

Thank you, Alan. Our core income for the third quarter was $798 million, generating core ROE of 13.5%, both up significantly from core income of $378 million and core ROE of 6.5% that we reported in the prior year quarter. These increases resulted primarily from this year's favorable third quarter PYD compared to net unfavorable PYD in last year's third quarter as well as a significant increase in underlying underwriting profit. More on both of those items in a minute.

Our third quarter results include $397 million of pretax cat losses compared to $241 million in last year's third quarter. This quarter's cats included Hurricane Laura, tropical storm Isaias, the severe straight-line winds that impacted the Midwest in August and several large wildfires in the Western United States. The increase in the level of cat activity was even more pronounced than those numbers suggest, as our net cat result in the quarter was tempered by recoveries under the aggregate catastrophe XOL treaty. We have recognized a full recovery under that treaty in our third quarter results with $233 million pretax benefiting the cat line and $47 million pretax benefiting non-cat weather in our underlying results.

Recall that last year, we did not have any recoveries under the treaty until the fourth quarter. Of course, the full recovery in this year's third quarter means that there is no coverage remaining from this treaty as we enter the fourth quarter. The underlying combined ratio of 91.5%, which excludes the impacts of cats and PYD, improved by 2.6 points from 94.1% in last year's third quarter. The underlying loss ratio improved by 2.4 points and benefited from favorable auto frequency related to COVID-19 and the impact of earned pricing in excess of loss trends partially offset by an increase in non-cat weather losses, including wildfires. The expense ratio of 29.3% is 0.2/10 of a point favorable to last year's third quarter results and reflects our strategic focus over a number of years on improving productivity and efficiency.

Setting aside quarter-to-quarter variability, our year-to-date expense ratio of approximately 30% is a figure we're comfortable with. Our top line proved to be resilient, with a 3% increase in net written premium as continued strong renewal rate change and retention in all 3 segments more than offset modestly lower insured exposures in the commercial businesses. For the quarter, losses directly related to COVID-19 totaled $133 million pretax, with $92 million in business insurance, driven primarily by workers' comp and $41 million in our Bond & Specialty business, predominantly driven by management liability. More than offsetting those losses were lower levels of auto claims and to a lesser extent, fewer non-COVID workers' comp and GL claims due to lower levels of economic activity.

The net impact of the COVID environment on the consolidated underlying combined ratio amounted to a benefit of about 2 points, mostly in Personal Insurance. Given the ongoing uncertainty in this environment, we continue to take a cautious approach in estimating the net impact of COVID-19-related losses. Consistent with my commentary last quarter, the majority of direct COVID losses that we booked year-to-date through September is still sitting in IBNR. Looking at the year-to-date impact of direct COVID losses, net of related frequency benefits and other underwriting items, our underwriting results have benefited by a little more than $100 million pretax or about 0.5 point on the consolidated underlying combined ratio, including the impact of premium refunds to policyholders. However, year-to-date net investment income reflects the significant adverse impact on our nonfixed income portfolio.

Turning to prior year reserve development. As previously disclosed, the third quarter includes approximately $400 million of pretax benefit from the PG&E subrogation. About 80% of that benefit is reflected in personal insurance, with the remainder reflected in business insurance. Setting PG&E to the side, PYD results in the quarter were as follows. In Personal Insurance, net favorable development of $40 million pretax was driven by auto results coming in better-than-expected for recent accident years.

In Bond & Specialty Insurance, there was no net impact from PYD. In business insurance, we recognized unfavorable development of $295 million pretax as a result of our annual asbestos review. While there was some slight improvement in several of our asbestos indicators. The overall level of paid losses and general claim activity have persisted at levels higher than we had anticipated. This year, as we do every few years, we reviewed certain macro assumptions underlying our actuarial analysis. Our updated view of ultimate asbestos-related losses resulted in an increase in the low end of the actuarial range. This year's asbestos charge is greater than last year's charge as a result of our updated view of the range for ultimate losses, not as a result of increases in paid losses or severity. There are some indications that the environment is improving in terms of the emergence of new asbestos claims going forward. Page 19 of the webcast presentation includes the most recent annual data from the centers for disease control and prevention, which shows that the total number of deaths from mesothelioma in 2018 decreased by nearly 6% compared to 2010.

Of note, as you can see on the bottom 2 lines of the table, the decline in mesothelioma deaths was much more pronounced in all of the younger age groups. This trend is directionally consistent with our expectation that over time, a high-risk group of people actually exposed to asbestos in the workplace prior to the late 1970s will get smaller and will not be replaced by younger people as those who entered the workforce sometime in the 1980s should not have been exposed to asbestos to nearly the same degree as their predecessors.

Excluding the impacts of the PG&E settlement and the annual asbestos review, there was virtually no net prior year reserve development in business insurance. Favorable development in workers' comp was offset by an increase to the reserves from legacy liabilities in our runoff book related to a single insurer arising out of policies issued more than 20 years ago. After-tax net investment income increased by 7% from the prior year quarter to $566 million. The increase was driven by our nonfixed income returns, where results for our private equity, hedge fund and real estate partnerships are generally reported to us on a one quarter lag. Because of that reporting lag, the recovery experienced in the broader markets during the second quarter benefited our nonfixed income results in the third quarter. Fixed income returns decreased by $31 million after-tax as the benefit from higher levels of invested assets was more than offset by the decline in interest rates, consistent with our comments on last quarter's call.

Also consistent with our prior commentary, we expect after-tax fixed income NII in the fourth quarter to be down $35 million to $40 million compared to a year ago. Looking ahead to 2021, our current expectation is for after-tax fixed income NII to be between $420 million and $430 million per quarter.

Turning to capital management. Operating cash flows for the quarter of $2.3 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 $2.3 billion, well above our target level. Recall that in April, we prefunded the $500 million of debt coming due in November with a new 30-year $500 million debt issuance so our holding company liquidity at September 30 is temporarily elevated by that month.

Investment yields decreased as credit spreads tightened during the third quarter and accordingly, our net unrealized investment gain increased from $3.6 billion after-tax as of June 30 to $3.8 billion after-tax at September 30. Adjusted book value per share, which excludes net unrealized investment gains and losses, was $94.89 at quarter end, up 2% from year-end 2019 and up 5% year-over-year. We returned $218 million of capital to our shareholders this quarter via dividends. We did not repurchase any shares during the quarter.

Looking ahead, there's no change in our approach to capital management. Until there's more clarity on the state of the economy, we may buy back some shares in the coming quarters or we may continue to choose to buy none. And now I'll turn the call over to Greg for a discussion of Business Insurance.

Thanks, Dan. Business Insurance produced $365 million of segment income for the quarter, a significant increase over the prior year quarter, with prior year development, underlying underwriting income and net investment income, all contributing to the year-over-year increase. The underlying combined ratio of 94% improved by almost 2 points, driven by more than a point of earned rate in excess of loss trends. A modest favorable net impact from the pandemic contributed about 0.5 point to the improvement. As for the top line, net written premiums were 1% lower than the prior year quarter. With strong rate and high retentions, mostly offsetting modestly lower insured exposures and lower levels of new business. As Alan mentioned, we are very pleased with the resilience of our top line in the face of the ongoing macroeconomic challenges.

Turning to the domestic production. We achieved record renewal rate change of 8.2%, up 4 points from the third quarter of last year and almost a point from the second quarter of this year, while retention remained high at 83%. We feel very good about the headline numbers but as we've shared before, the quality of the execution and segmentation underneath the headline numbers are just as important. To that point, while we achieved meaningful rate increases in all product lines, except workers' comp. Our underwriters are making deliberate and granular decisions with respect to rate and retention on an account-by-account or class-by-class basis.

Thanks to our focus on continuous improvement of data and analytics at the point of sale, the quality of the execution is as good as I've ever seen. New business of $505 million was 9% lower than the prior year quarter. We attribute the decline to lower levels of economic activity as well as careful risk selection by our underwriters. In our core middle market business, for example, while submissions are up, our quote ratio is lower as we are taking a disciplined approach, given our view of quality of new business in the market. As for the individual businesses, in select, renewal rate change increased to 2.9%, marking the seventh consecutive quarter in which renewal rate change was higher than the corresponding prior year quarter. Retention of 80% was down a couple of points from recent periods, largely driven by policy cancellations that were deferred through the second quarter due to our pandemic-related billing relief program. In middle market, renewal rate change increased to 8.3%, while retention remained strong at 85%. The 8.3% was up by more than 4.5 points from the third quarter of 2019, and we achieved positive rate of more than 80% of our accounts this quarter, up from about 2/3 in the third quarter of last year.

To sum up, we feel terrific about our results and execution in a challenging underwriting environment. We also feel very good about the investments we're making for the future and the benefits we're seeing from those investments. Those investments include enhancing the experiences for our customers and distribution partners, digitizing the underwriting transaction in creating efficiencies. For example, we've recently launched multiple pilots to automatically incorporate data from our distribution partners, agency management systems directly into our systems, substantially reducing the time and friction in the process while also improving data quality. We're as confident as ever that our meaningful competitive advantages position us well for long-term profitable growth.

With that, I'll turn the call over to Tom.

Thanks, Greg. Bond & Specialty delivered solid returns and growth in the quarter despite the ongoing headwinds of COVID-19. Segment income was $115 million, a $24 million decrease from the prior year quarter as the benefit of higher volumes was more than offset by a higher underlying combined ratio. The underlying combined ratio of 89% increased 5.4 points, primarily driven by estimated losses from COVID-19 and related economic conditions. Given the products that we write, we expect the results of this segment to be impacted in times of severe economic downturn. We experienced that during the financial crisis, and we're seeing elevated loss activity in the current environment. We contemplate economic volatility in our underwriting and in our pricing and as Alan said, in these circumstances, we feel good about the returns we generated in the quarter. We expected the underlying combined ratio will continue to be elevated at around this level over the near term.

Turning to top line. Net written premiums grew 4% for the quarter, reflecting strong growth driven by improved pricing in our management liability business, partially offset by lower surety production due to the continued economic impact of COVID-19 on public project procurement and related bond demand.

In our domestic management liability business, we are pleased that the renewal premium increased to 8.1%, driven by record rate. This marks the eighth consecutive quarter in which RPC is higher than the corresponding prior year quarter. Retention remained at a historically high 90%. These production results demonstrate the successful execution of our strategy to pursue rate where needed, while maintaining strong retention of our high-quality portfolio.

We will continue to pursue rate increases where warranted. Domestic management liability new business for the quarter decreased $14 million, primarily reflecting our thoughtful underwriting in this elevated risk environment. Similar to what you heard from Greg in business insurance, submissions are up, while quote activity is down. So Bond & Specialty results remained resilient, despite the challenges brought on by COVID-19. We continue to be pleased with our strong execution and feel confident about our ability to navigate through this challenging environment and continue to deliver strong returns over time. And now I'll turn it over to Michael to discuss Personal Insurance.

Thank you, Tom, and good morning, everyone. In Personal Insurance this quarter, we are very pleased with our continued execution in the marketplace as we delivered excellent profitability and grew net written premiums by 8%, achieving record levels of domestic policies in force.

Personal Insurance segment income for the third quarter was $392 million, up $261 million from the prior year quarter. Driven by the pretax impacts of an improvement of $163 million in the underlying underwriting gains and $343 million of higher net favorable prior year reserve development, partially offset by $174 million of higher catastrophe losses net of reinsurance.

Our combined ratio for the quarter was 86.4%, an improvement of 11.6 points from the prior year quarter, driven primarily by the increase in net favorable prior year reserve development. Higher catastrophe loss experience in the quarter was largely offset by improvement in the underlying combined ratio. The improved underlying combined ratio reflects the continuation of favorable auto loss experience in the quarter, partially offset by higher non-catastrophe weather-related losses. I'll discuss both of these dynamics in a bit more detail next. Agency Automobile profitability was very strong, with a combined ratio of approximately 80% for the quarter. The underlying combined ratio of 81% improved nearly 12 points, continuing to reflect favorable frequency levels. Approximately 8 of the 12 points of improvement relate to current quarter favorability. The remainder results from favorable reestimates of activity in the first half of 2020. We continue to observe lower claim frequency as a result of fewer miles driven in light of the COVID-19 pandemic.

For the third quarter, data from our IntelliDrive program indicates that miles driven increased relative to last quarter, but continued to be down from pre-COVID-19 levels. In response to this continued favorable loss experience, we filed modest rate reductions in a handful of states during the third quarter. We will continue to analyze and incorporate current trends into our underwriting and pricing decisions, as we balance business volumes and profitability. In agency homeowners and other, the third quarter combined ratio was 92.8%, an improvement of 9.2 points from the prior year quarter resulting from 26 points of higher net favorable prior year reserve development, mostly from the PG&E subrogation recoveries. Partially offset by elevated levels of catastrophe losses and an increase in the underlying combined ratio, driven by higher non-catastrophe weather-related losses. Our catastrophe and non-catastrophe experience reflects a very active quarter with a record 31 PCS events.

West Coast wildfires represented almost half of the total PCS events in the quarter. Consistent with Dan's comments earlier, the quarter's catastrophe losses for personal insurance were also impacted by the Midwest Derecho, Tropical Storm Isaias, and to a lesser extent, Hurricane Laura.

In addition to pursuing rate increases in property, as we have been for some time, we continue to review and modify terms and conditions and implement loss mitigation actions in response to the elevated loss activity. Our actions to date have enabled us to reduce or avoid losses we would have otherwise incurred and improve returns as we continue to grow the line.

Turning to quarterly production. Our domestic agency results were again very strong. Our retentions remained high. Quotes and new business were up versus the prior year quarter, and we remain pleased with our policies in force growth. Agency Automobile retention was 84%, and new business increased 9% from the prior year quarter, both contributing to accelerating growth in policies in force.

Renewal premium change was again lower as we continue to moderate pricing in response to favorable loss activity. Agency Homeowners and other delivered another very strong quarter with retention of 86% and a 22% increase in new business. Renewal premium change increased to 8.2% as we remain focused on improving returns in property while growing the business. During the quarter, we continued to respond to the needs of our customers and distribution partners.

First and foremost, let me add my thanks to our claim professionals for delivering on our commitment to be there for our customers and distribution partners in response to the significant number of catastrophe and weather events in the quarter. At the same time, we continue to deliver new capabilities in the marketplace. Alan already mentioned both IntelliDrive and our new MyT mobile app, both of which are key tools in helping us attract and retain customers. We also expanded the availability of our digital quote proposal that gives agents and brokers the ability to send a Travelers insurance quote to their clients' mobile phone and interact with them digitally about the terms of the proposal, making the transaction more seamless for both the agent and the customer. And after reaching our goal of planting 1 million trees for customer enrollment in paperless billing, we extended our partnership with American Forests to plant an additional 500,000 trees by Earth Day 2021.

We have already achieved that milestone well ahead of schedule, providing our customers the digital experience they seek while benefiting the environment. These examples and others illustrate our ability to develop and deliver the capabilities our partners and customers value. We are very pleased with our performance so far this year, and I'm proud of our team's efforts to continue to deliver results while investing in the business for the future. Now I'll turn the call back over to Abbe.

Thank you very much. Operator, we're ready to start Q&A.

(Operator Instructions) And our first question comes from the line of Mike Zaremski from Crédit Suisse.

First question, if we can talk about some of the reserve changes during the quarter. Thanks for all the commentary. I believe, Greg, when you were talking about business of the BI segment, you didn't mention any additions to general liability. And I believe that had been -- you'd been adding to that for -- in prior quarters. Could you confirm if that -- it was indeed the case? And maybe just overall to Alan or Greg, if you guys can talk about whether your view has changed at all on overall loss expense trend?

Mike, it's Alan. Let me start with loss trend, and then maybe I'll turn it over to Dan to talk about prior development. So in the quarter, putting aside for a second, the direct COVID losses. In the quarter, generally, what we saw was favorable loss activity, primarily related to COVID. And so I think people staying home, less economic activity, lower levels of frequency. As you heard us say last quarter and as you heard us repeat again this quarter, we've taken a very cautious approach to recognizing that favorable news. There is still a fair amount of uncertainty and so we're going to be cautious about that. But we didn't see anything in the quarter that surprised us or caused us to think that the loss trends were deteriorating. And Dan, why don't you comment on PYD?

Sure, Mike. So just to go through business insurances, PYD, just for a second, quick recap of the summary. So it ended up being about $220 million unfavorable all in, that's driven by $295 million, as we said on asbestos. Then there was favorability from BI's portion of the PG&E settlement. And I said that was worth about 20% of the $400 million, so that's $80 million of a good guy. And that gets you to about $215 million compared to the $220 million. So the other moving parts you could think about as net offsets and those are favorable in workers' comp unfavorable for a single account and runoff from policies that we issued more than 20 years ago. And we didn't mention the other lines because there was really very modest activity in the other lines, some pluses and some minuses that basically net to 0.

Okay. That's helpful. My last question is, I guess, in regard to the overall competitive environment. Alan, you -- in your prepared remarks, you talked about expect continued momentum in the marketplace. I mean I guess -- I think what we're trying to get comfortable with is the investment income levels, we can -- thank you for the guidance, right? Clearly, they're under a lot of pressure. And I think in our shoes, we haven't been in a kind of a firm-to-hard marketplace in a while. So we're trying to get comfortable that the pricing momentum and excess of loss trends is going to continue because it seems like to offset the NII pressure, we have to get to meaningfully better combined ratios in outer years. So just -- maybe just some more color? Alan, do you guys -- when you say continued momentum, do you expect pricing to continue moving higher or are these levels kind of good enough for you guys to start kind of chipping away at offsetting some of the investment pressure?

Yes. Well, clearly, my written rate is ahead of loss trend. So we're definitely making progress at these levels, and we're not going to give pricing outlook, but we do choose our words pretty carefully. And I'll tell you that the fact of the matter is that for us and for the industry broadly there are just many main lines of business that are not yet rate adequate, and the drivers are clearly environmental. So on the loss side, social inflation isn't going away. We've got weather losses. We've got wildfires that are burning. As you pointed out, interest rates are lower for longer, that's going to be a driver of the outlook for returns. We do have some capacity issues in some lines as there are some markets that are pulling back on capacity. So we think this broader environmental -- all of these broader environmental trends will continue to provide positive momentum in the marketplace.

Our next question comes from the line of Ryan Tunis from Autonomous Research.

I guess my first question is it sounds like the COVID benefits you're enjoying are frequency-based, mostly short-tail lines. So I should take that to mean that you've continued to book in lines like commercial auto, similarly to the type of loss trend that you thought coming into the year, which I would think would be above your average loss trend, correct?

Yes, Ryan, it's Dan. That's right. So we're -- as we've said, I think, last quarter and this quarter, trying to be cautious in our assessment of the net impact of COVID. There is some benefit in the shorter tail lines, but we're trying to be pretty cautious in terms of how much of that we recognize in the current accident year. But in terms of the casualty, coverages, as you mentioned, yes. Yes, that's correct.

Right. And I totally understand that. Just trying to get some perspective on -- clearly, we don't know what frequency is going to be. There could be delays in reported claims, et cetera. But just from your seat, looking at the type of claim activity that's coming through the door, if that were to hold true, what's the type of frequency benefit on commercial auto that, I guess, we're not at this point seeing any results yet?

Yes. We're reluctant to put a number on that. I think we're not going to do it, Ryan. But -- and frequency, it does develop, but it develops certainly a lot more quickly than severity does. I mean you know if the phone rings or the phone doesn't ring, but I think we're going to leave it as -- we've recognized some of it because there was enough of it that we couldn't not recognize some of it, but we've been pretty cautious in the assessment.

Got it. And then the other follow-up I had, I guess, probably for Dan, is thinking about just the aggregate treaty. I guess it kind of worked this year, but it also kind of didn't because we're going into the fourth quarter, and we've gone through the top. How are you thinking about the strategy about reinsuring cats, I guess, heading into '21? Are you thinking about doing more aggregate cover, maybe going lower? Is that something that we should expect potentially be a catalyst for next year?

Ryan, before Dan answered that question, I'll just insert that I think the treaty actually worked just fine, it was the weather that didn't work so well, but...

Ryan, I think we'll do, as we would do for reinsurance in any -- going into any year, we'll make an assessment of price versus risk and then decide what and how much we're going to buy. I think I'd say in summary, we're glad that we've had that treaty for the last 2 years.

To your question specifically about do we go to a lower attachment? I wouldn't think so. We've got a growing business. And if anything, we probably would go naturally to a higher attachment point just to reflect the growth in the business. But maybe we'll buy that reinsurance again and maybe we won't. It will depend how we feel about the risk transfer relative to the pricing that we can get because, again, we're writing with the idea that we're a gross lines underwriter, and we said a couple of years ago, it felt like there was a little more uncertainty in the weather environment. And so we wanted to have that treaty, and we're glad we had it. And we'll see how that plays out going into next year.

And the other thing I would add to that, Ryan, is -- and Dan is exactly right. We bought that because we were -- wanted to acknowledge the incremental uncertainty that we felt after some wildfire losses, I guess, in '17 and '18. But as I mentioned in my prepared remarks, we've taken pretty substantial steps to try to get that risk reward in balance. So that would also factor into our view of the value of that reinsurance treaty and the terms we can get it on next year.

Our next question comes from the line of Brian Meredith from UBS.

My first one, I guess, is maybe some commentary about workers' comp. I mean I think that given where interest rates are, at some point, we should start to see that line see some positive rate momentum and that's an important component. But what are you seeing, kind of what are your thoughts on that line right now?

Greg, do you want to take that?

Yes, you bet, Brian. First, we like our long-standing competitive and economic advantages that we've got out of that line, and that's why we're the largest writer in it -- in the U.S. and I'm sure there's a little bit of pricing reductions going through that line right now, and that always puts some pressure on existing margins, but we aren't standing still. We continue to invest in our claim practices, specifically around medical management sophistication. And our underwriters, were always giving them new tools and new insight to make sure our risk selection and our renewal segmentation offset some of those price reductions. And workers' comp is just one piece of our portfolio. As we remind you all the time, we're more of an account solution than an individual monoline writer, and we'll continue to write our customers with accounts in addition to workers' comp. So we continue to feel good about the portfolio and the future of that line.

I mean was that improvement in Select, was that due to comp maybe getting a little less competitive? The rate in Select?

Well, all -- both Select and middle market are clearly going to follow the linear trend of the bureau's loss cost recommendations. And so in the Select business, where it's more of a flow business where you file that rate structure, you feel it more there than you would in middle market, where you have experienced mods and discretionary pricing and you look at the individual exposure rather than file at a state level. So a little more sensitive to the reductions in Select than you would see in middle market.

Great. And then just one quick follow-up for Dan. The aggregate cover this quarter, is it possible to break out how much of that benefit is kind of non-cat weather versus cat losses?

Yes, Brian, I gave it actually in my prepared remarks. So...

I'm sorry, I missed it.

That's quite all right. We said -- so $280 million was the total recovery, $233 million, one to cat and $47 million underlying.

And that's largely in the personal lines?

The split is a little different between the 2. The cat split is about 50-50 PI/BI the noncash split goes more like 70-30 PI/BI.

Our next question comes from the line of Elyse Greenspan from Wells Fargo.

My first question is also on the aggregate cover. So given that you guys have exhausted that cover and this is a comment that is specific to business insurance. How can we think about the improvement within the underlying loss ratio that you could see there in the fourth quarter?

I guess we could expect some kind of headwind from non-cat weather just given the absence of the aggregate? Or how should we think about the fourth quarter?

Elyse, it's Dan. The aggregate -- so as I just said in responding to Brian's question, not a big deal to the underlying for business insurance in the third quarter. So it got about 30% of the $47 million that was non-cat. So you could think about that for BI as not being more than 20 or 30 basis points in terms of its impact on the quarter.

Okay. So if we adjust BI for that and then also for like the COVID benefit that you guys pointed out earlier, is that a good type of kind of run rate to think about for the margin improvement that we could see in the fourth quarter?

I don't think I'd expect the COVID impact to cease as of September 30. Certainly, the cat recovery, the treated recovery is all contained in the third quarter, and you could take that out for sure. I'm not sure anybody knows what exactly the impact of COVID is going to be in the fourth quarter.

Or what weather volatility is going to look like. And I think we're going to get away generally from trying to forecast margins. But we're happy to take some modeling questions offline, if that's helpful, least we can.

Yes. And then my second question was on the expense ratio. Dan, I think you said that 30% was kind of a good run rate. But as we're thinking -- and like maybe this is less guidance, right, but you guys did guide with at least that. As you're getting a good amount of price and that's turning in, should we expect that we could see some improvement in the expense ratio just due to the denominator? Or is it counterbalanced by some investments that you pointed out that you guys have been making internally?

I think those are the considerations, Elyse, that will go into what it ends up being. I do think to the degree that we can get some bottom line growth, we would to be in a position to continue to see some leverage on the expense ratio. But as we've said a number of times over the last couple of years, what we like about productivity and efficiency that we've generated as it gives us the optionality of whether we make additional investments in places that we feel are going to provide a bigger bang for the buck down the road or let some of that come through the bottom line. All I was really trying to communicate was that we don't -- we're not seeking a dramatic change in the expense ratio from where we are right now.

And our next question comes from the line of Josh Shanker from Bank of America.

Two unrelated questions. The first one is about surety. And can you tell us a little bit about the premium volume looked pretty reasonable in the Bond & Specialty business? What are we seeing in terms of COVID economy for construction starts and whatnot?

Yes. This is Tom, Josh. When you think about surety, if you break it up to the nonconstruction and the construction side, our regular nonconstruction surety is holding up quite nicely. But with all of the decline in tax receipts and the difficulties that various public bodies are having funding public works projects, that's what is really driving the drop-off in Surety written premium. And so we will have to wait and see what the federal government does with any stimulus that might involve construction because certainly, states and other entities have shovel-ready projects.

Is there an implication for things that have already begun construction? Is there a risk that we start seeing issues with the Surety claims because they're not able to be finished?

There are 2 different issues. As far as the projects stopping, unless they're ordered to stop by government authority, I wouldn't expect to see a project that is ongoing stopping. And then as far as a slower economy causing additional defaults, that certainly is possible. It really depends on the contractors in the Surety's portfolio and how sound their financial position was going in and how they handle pursuing work or not pursuing work and shrinking the size of their company when times are lean. So we take a lot of pride in our contractors and we manage that portfolio. And we think different sureties will perform differently throughout this.

Josh, I'd just add to that, that we're underwriting that commercial Surety but the construction Surety book today is the same way we did in financial crisis and it performed pretty well through that. So we can't give any assurances who knows, but we feel pretty good about our underwriting practices and the people who are making the decisions to put business on the books.

Yes.

And then sort of a philosophical question. If we go back in time, maybe 10 years, I don't know when you came up with a Travelers definition for catastrophes. And of course, it varies from the PCS definition. It seems as time goes on, the risk associated with non-cat weather gets greater and greater in terms of volatility on your margins. Is the variance between your definition, the PCS definition still useful? Or have we gotten to a point where the reasons for that difference are starting to create a certain volatility in results that we wouldn't otherwise hope to see?

Yes, Josh, we think about that a lot. So I appreciate the question. And maybe someday offline, we'd love to get your thoughts on it. I mean our view is, from an economic perspective, the losses are going to be what they're going to be, whether they get reflected in a cat line or in an underlying line. And frankly, we report the numbers the way we manage the business. And we think that there is a level of loss activity that is defined as a PCS event that is in a working layer, if you will, and that ought to be managed day in, day out, year in, year out by the underwriters that are managing it as compared to a catastrophe loss, which we obviously define as a higher level, which get managed and addressed over a much longer period of time.

So we think it's the right way to manage the business. We think it's the right way to talk about the business internally, and we make the decision to talk about it externally for the same way we manage it internally. So I get that it might from time to time, cause some modeling issues or definitional issues, but we think the consistency and the thoughtful approach to managing it is sort of the right way to run the railroad.

And our next question comes from the line of Paul Newsome from Piper Sandler.

I was hoping you could home in on the competitive environment of the auto insurance business, in particular, seeing in the market different things from rate environment, obviously, you're doing what you're doing. But is it a materially more competitive market today than it was recently. And then I'll just give you my follow-up question, which is I'd like to know how that competitive environment links up with what you're doing in the home insurance business, where it's clearly growing, and my sense is the home insurance business is pretty much an outright hard market. So a different situation. So those are 2 questions.

Michael?

Thanks, Paul. It's Michael Klein. So on the competition in auto, certainly, we are seeing market rate drop in, I think, it's fair to call it an intensifying competitive environment. That said, for the most part, notwithstanding some of the headline numbers you see from some key competitors. It's still a relatively moderate adjustment in personal lines auto pricing. And from our perspective, largely reflective of the improvement in experience in the line. Some of it COVID related, but frankly, some of it consistent with longer-term trends of favorable frequency that we and other competitors have talked about.

In terms of our ability to compete in that environment, as I mentioned in my prepared remarks, and you see in the webcast script, we're pleased with our improving PIF growth in auto as we moderate pricing, again, intentionally to keep pace with those loss trends. And again, as I mentioned, in the prepared remarks, that does include taking some price decreases in a handful of states to, again, match premium to loss, pricing with loss experience. So that's sort of our view on the PI competitive environment.

Jumping to your second question about home, I would agree. If you look at the personal insurance industry right now, it's a little bit of a tale of 2 cities, right? You've got improving loss experience and moderating frequency in auto, putting downward pressure on prices. You've got a lot of what we've been talking about here this morning, catastrophes, non-catastrophe weather. There are still folks in the industry talking about nonweather water losses as well. But the pressure on loss costs in the property business is upward. We're pleased with our progress in RPC, in property. We're pleased with the performance of our Quantum Home 2 product as we've rolled that out. And again, as we look at the property market and personal insurance, as we've described it, our objective is to grow property, while improving profitability. And that's the path we've been on and will continue to be on.

I guess what I was trying to ask was, is there -- how much is the link between the willingness to grow in the home versus the auto?

I would say we look at it as a portfolio. Our strategy is to be a portfolio provider and work with our partners to provide total account solutions to customers. And so we are actively seeking to grow both lines and again add target returns.

Our next question comes from the line of David Motemaden from Evercore ISI.

Just had a question just on NII and the impact it has on ROE for the firm. And so the NII guide of $420 million to $430 million after-tax per quarter. I guess do you think that you can get enough margin expansion to offset the NII drag to keep ROE stable in this environment?

And is there enough pricing and margin expansion underneath where you can actually improve ROEs in this interest rate environment?

Yes. Well, we think the answer to that is yes. And if you look at written rate versus loss trend, margins are improving from here. And we're not going to give guidance, so I'm not going to predict whether we will or whether we won't. But I mean, certainly, interest rates and investment income are an input into our pricing model. And so there's certainly a very direct link between the investment return we're getting on the product and the price we're charging for it.

Okay. Great. And then just another follow-up. You mentioned, I think Greg mentioned getting rate in excess of loss trend of over a point in BI. I guess I'm just wondering if you could elaborate a bit on what gives you confidence that you're getting rate in excess of trend, just given the uncertainty around the loss environment, courts being closed, and just generally, the unpredictability of losses in the current environment.

Yes. So just to be clear, we look at rate versus loss trend on a written basis, and that's the rate number you see in production statistics, and we've shared before our view of loss trend. That one point was sort of the earned impact of rate and earn loss trends. So 2 different bases there. At this point, where we've got rate in excess of 8%. And we've got loss trend, which we shared last quarter, we view on the long-term basis at 5%. Yes, there's some uncertainty, but that gap is big enough that we feel some confidence in that. And it's true that in a circumstance like this, where you've got a disruption in the economy and some disruption in the data coming through. And in claim process, there's a little less -- there's a little more uncertainty than usual but as I've shared previously, what we're seeing in losses is actual favorability.

And again, not perfect clarity into it, and it's going to take some time for us to get more clarity and for that uncertainty to resolve. But the thing that we're not reacting to is favorable news, not unfavorable trends.

Our next question comes from the line of Yaron Kinar from Goldman Sachs.

First question around workers' comp. So clearly -- or I'm assuming there is a frequency benefit from non-COVID claims. On the other hand, I would think that, that frequency benefit could ultimately result in some severity increase further down the road. Is that a fair way of thinking about it? And if so, how are you thinking of the loss PIKs with that come dual dynamic?

It's certainly a possibility, and we contemplate that. So yes, there is some favorability in non-COVID related frequency, but we do contemplate when we think about longer-term trends and when we think about the profitability of the product that there could be down the road some increased severity due to really delayed access to health care. So we do contemplate that as a possibility.

And is that reflected in the PIKs already or is that something that you'll be thinking about as you move into 2021 and beyond?

It's Dan. I'd say both, which goes to the comment that we've made a couple of times about being cautious in our view of the net impact of COVID and we'll continue to consider that as we think about the product on a go-forward basis.

Got it. And then, Alan, I know you and in the firm, you've been vocal about social inflation trends, particularly in commercial auto the last couple of years. With courts being shut down, now starting to reopen, like are you seeing any change in the trend or is it too difficult to tell just given all the COVID-related noise today?

Yes. I guess what I can share is we didn't see anything that caused us to believe that there was any deterioration in those trends. And what we saw was, again, as I've said a few times, favorable. We think the extent to which it's favorable is a temporary phenomenon. So I guess that's way to answer the question. There is nothing in the quarter that caused us think anything was deteriorating.

Yes. At the same time, we certainly don't think that social inflation has gone away. And so our long-term expectation is that the sort of elevated levels of losses that we've seen related to social inflation probably persists.

Our next question comes from the line of Jimmy Bhullar from JPMorgan.

So first, I just had a question on pricing in the workers' comp line. Obviously, it's been declining. Are you seeing any signs of stability in that? Or any sort of -- what are your views on where that's headed?

I think I'll -- Jimmy, I'd repeat what we said last quarter, which is when we look at our data, we look at the data coming out of the rating bureaus, we would think that we are somewhere near an inflection point, whether it's now or in the coming quarters, not really sure, but we would expect in the relatively nearest term for that pricing to come back to negative and then turn positive. It's impossible to pick the exact quarter or date that's going to happen, but that would be our near-term outlook.

Okay. And then on sort of -- you mentioned you'll evaluate price versus risk on reinsurance. It seems like the reinsurance prices are obviously going up a decent amount. What's your -- any sort of initial indications of your views on sort of price versus risk given what's going on in the reinsurance markets versus last year?

Yes. I mean we certainly expect reinsurance pricing to go up. That's one of the factors that's driving our pricing and the pricing generally in the primary market going up, and we think that's going to persist. I guess the thing I would add as it relates to us is we are, as we've said many times, a gross line underwriter. We like our underwriting, we tend to keep it. And so compared to many of our -- many in the industry, we buy on a relative basis, less reinsurance. And frankly, we could buy in less than we do. We could buy a little more. We could buy a little less, depending on the risk and reward that we see in that purchase. But to the extent that we're relatively less reliant on reinsurance than others. The fact that reinsurance pricing going up is probably a benefit to us.

Okay. And then just lastly, what are some of the things that you're watching to determine whether to start buying back stock again?

I think, Jimmy, still overall economic uncertainty, right? So even as we sit here today, what's the depth and the duration going to be of economic downturn? Is there going to be a second dip in the economy? How quickly are COVID cases going to be brought under control? How quickly are things going to reopen? There's just still a lot of uncertainty that could impact both our top line and the loss environment in which we're operating. So we're choosing to be a little cautious here still. And clearly, through the third quarter, on top of all that, we had a very active cat quarter. So it just didn't feel like the right time to be restarting buybacks.

And we have one -- time for one further question. Our final question today will come from the line of Meyer Shields from KBW.

Great. I think it was Dan or Alan, I apologize for not knowing for sure. They talked about how you were taking a conservative view of accident year loss PIKs in light of uncertainty and only recognizing the frequency benefit that you had to. Was there a sort of similar approach taken to reserve reviews in the quarter?

Meyer, it's Dan. I'm not sure I quite understand the question. So the reserves will be a result of the loss PIKs that we made in the current accident year during the quarter?

Right. But when your -- so if I understand correctly, there is an overall conservatism because of the fact that things are developing differently as we work our way through the COVID and the economic ramifications. And I'm wondering whether that conservatism had sort of an unusual impact on the selections you made with regards to reviewing car period reserves as well?

Yes. I wouldn't think so. I think the comment that Alan just made a few minutes ago around the impact of courts being closed and settlements being down, we see some disruption in the data, but we're not taking that as a favorability in the development patterns that we're assuming on a long-term basis.

Okay. That's helpful. Second follow-up, if I can, really quickly. Just how should we think about the impact on return on equity or on indicated pricing if corporate tax rates go up?

Yes. Taxes are an input in the income statement like any other expense. And so it flows through the models. Yes. So that would be one of the many things that gets factored into our pricing. Obviously, it wouldn't just impact us. It would impact the economy more broadly. And as Alan said a bunch of times. I think if we feel like we've got a level playing field with our experience and our risk selection expertise we will do just fine. But for sure, to be an input into what the pricing environment needs to be going forward.

I would now like to turn the call back to Abbe Goldstein for closing remarks.

Thank you all very much for joining us this morning. And as always, if you have any follow-up, please feel free to reach out to Investor Relations. Hope everybody has a good day.

Thanks. Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect.