旅行家集團 (TRV) 2018 Q2 法說會逐字稿

Good morning, ladies and gentlemen.

Welcome to the second quarter results teleconference for Travelers.

(Operator Instructions) As a reminder, this conference is being recorded today, July 19, 2018.

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 second quarter 2018 results.

Hopefully, all of you have seen our press release, financial supplement and webcast presentation released earlier this morning.

All of these materials can be found on our website at www.travelers.com under the Investors section.

Speaking today will be Alan Schnitzer, Chairman and CEO; Jay Benet, Chief Financial Officer; and our 3 segment Presidents: Gregory Toczydlowski of Business Insurance; Tom Kunkel of Bond & Specialty Insurance; and Michael Klein of Personal Insurance.

They will discuss the financial results of our business and the client 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 Allen, I'd like to draw your attention to the explanatory notes 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 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.

Thank you, Abbe.

Good morning, everyone, and thank you for joining us today.

This morning, we reported second quarter net income of $524 million and return on equity of 9.2%.

Core income was $494 million, generating a core ROE of 8.7%.

These results were significantly impacted by an active tornado-hail season.

Catastrophe losses of $488 million this quarter arose out of 9 storms.

To give you some context, that's around $50 million more than we would have expected.

To put a finer point on it, losses from PCF-defined catastrophes, that don't hit our threshold to qualify as cat losses as we report them, were about $25 million favorable to our expectation.

So all in, losses from PCF events were only about $25 million pretax more than we would have expected, that's well within the normal variability that we anticipate and price for.

Having said that, this makes a recent series of quarters of catastrophe losses that have exceeded our historical experience and our expectations.

Tornado, hail, Nor'easters, hurricanes, wildfires and mudslides.

We haven't seen a string like that in the last decade.

It also includes some unusual circumstances.

For example, the California wildfires were historic, and last year was the first time more than one category for hurricane has made landfall on the U.S. mainland in one season.

In terms of creating shareholder value over time, we don't want to overreact any more than we want to underreact.

And when it comes to something as inherently unpredictable as weather, we take a balanced approach to developing conclusions from what takes place over a relatively short period of time.

As always, the impact of weather on our business has our full attention.

And we'll continue to use our leading actuarial expertise and the latest in weather modeling to inform our underwriting and pricing decisions.

In addition to the weather, the underlying results in business insurance include a small number of large commercial losses, primarily fire related, that exceeded our expectations by nearly a point on consolidated combined ratio.

We believe this is normal variability in large loss activity, and Greg will share more detail on that with you shortly.

Whether it's weather or other large losses, there is no doubt that actual resorts are going to vary from our expectations some time significantly.

What's important in this business is that we have the data analytics and expertise to see and evaluate the trends, so we can manage for success over time.

We're confident that we do.

On an underlying basis, the underwriting results benefited from record earned premium.

The underlying combined ratio of 93.6% was strong and consistent with the prior year.

Underneath that, our businesses continue to perform well.

In Business Insurance, the underlying combined ratio of 96.5% was solid, particularly in light of the large loss activity I just mentioned.

Bond & Specialty Insurance produced another impressive quarter with an underlying combined ratio of 80.5%, a 1.5% improvement from a strong result in the prior year.

In Personal Insurance, the underlying combined ratio improved about 2 points as a result of the successful execution of the pricing and underwriting actions we've taken in our auto business.

Our consolidated expense ratio improved by 40 basis points in the quarter and 30 basis points on a year-to-date basis, as we continue to benefit from expense discipline and strategic investments in technology and workflow to improve productivity.

In terms of our investment results for the quarter, we were pleased that income from our fixed income portfolio was up, driven by higher short-term rates and a higher level of invested assets.

Turning to production.

We were very pleased with our continued successful execution in the marketplace.

Net written premiums increased by 7% to a record $7.1 billion with each of our business segments contributing.

As we saw this quarter and at recent quarters, we would expect underwriting results going forward to benefit from higher levels of earned premium.

Once again, premium growth reflects to a large degree high levels of retention and positive renewal premium change.

As I pointed out in the past, that speaks to the quality of the business we are writing.

In addition to that, we explained at our Investor Day last fall that the strategic investments we're making are designed in large part to create top line opportunities for us.

The early success of some of those initiatives contributed to a healthy level of new business in the quarter.

In Business Insurance, renewal premium change reached 5.3 points, its highest levels since 2014, while retention remained at historically high levels.

Excluding workers' comp, pure renewal rate change for domestic business insurance was 3.6 points compared to a little less than 1 point in the same quarter last year.

In Bond & Specialty Insurance, record retention, higher renewal premium change and higher new business led to a 6% increase Management Liability net written premiums.

Surety premium was up double digits.

In Personal Insurance, as you know, we grew our PI auto business considerably from late 2015 to the first part of 2017, including during a period of time, in which the rate we were charging was not sufficient to cover an unexpected increase in bodily injury severity.

We made the point at the time that the volume of business we wrote during that period would nonetheless be a positive contributor to economic value, if we were able to keep that business and achieve rate adequacy.

We have, in fact, been able to do that.

Greg, Tom and Michael will provide more detail on production at the segment level.

To sum it up, weather and large losses unfortunately color the quarter.

As I said, it has our full attention.

But looking at the quarter and looking forward, that doesn't define the strength of our business.

We see that clearly in our underwriting -- underlying underwriting results and in our success in the marketplace.

That is, in part, a reflection of the competitive advantages that we've developed over decades and continue to service well and new capabilities were developing to enable us to continue to lead in a rapidly changing world.

We have and will continue to invest in our franchise by extending our lead and risk expertise; improving the experience for our customers, agents and brokers; and enhancing productivity and efficiency.

Combining that with our strong balance sheet, superior talent and capital management strategy, we remain well positioned to continue to deliver industry-leading results over time.

And with that, I'll turn it over to Jay.

Thanks, Alan.

Core income was $494 million, down from $543 million in the prior year quarter, and core ROE was 8.7%, down from 9.5%.

As Alan indicated, these changes were not driven by fundamentals in our operating performance, rather it resulted from a pretax increase of $85 million in catastrophe losses, $488 million compared to $403 million in the prior year quarter and an incremental pretext charge of $45 million related to a few large commercial losses that were primarily fire related.

PYD was also modestly lower, $186 million pretax compared to $203 million.

The higher cats and lower PYD accounted for 1.3 of the 1.4 point increase in our consolidated combined ratio, 98.1% versus 96.7% in the prior year quarter.

Our consolidated underlying combined ratio of 93.6%, which excludes the impacts of cats and PYD remained steady, changing by only 0.1 as a 1.7 point increase in BI's underlying combined ratio that primarily resulted from a fire-related losses was almost entirely offset by improvements in both PI and Bond & Specialty, 1.9 and 1.5 points, respectively, demonstrating the value of our diversified set of businesses.

Our pretax underlying underwriting gain of $392 million increased by $19 million, driven by increases in both PI and Bond & Specialty.

Greg, Tom and Michael will provide more details of our segment results shortly.

Pretax net investment income remained strong, $595 million in the current quarter versus $598 million in the prior year quarter, while after tax NII increased from $468 million to $507 million due to the lower U.S. corporate income tax rate for all the tax exempt investment income.

As in the first quarter, pretax fixed income NII of $510 million increased by $26 million compared to the prior year quarter, driven by the more favorable interest-rate environment, particularly for short-term rates as well as an increase in average invested assets that resulted from recent growth in net written premiums.

After-tax fixed income NII increased by $48 million, and looking forward, we expect after-tax fixed income NII for the remainder of 2018 will increase by approximately $55 million to $60 million in each of the third and fourth quarters compared to their corresponding quarters in 2017.

Nonfixed income continued to perform well, although not as well as in the prior year delivering $94 million of pretax NII.

And overall, core income benefited this quarter by $54 million due to the lower U.S. corporate income tax rate.

Turning to reserve development and on a pretax basis, BI's current quarter net favorable development of $84 million was primarily driven by better-than-expected loss experience in domestic workers' comp, partially offset by higher-than-expected experience in general liability for accident years 2008 and prior, including a $55 million increase in environmental reserves.

Importantly, the current GL loss trend remained consistent with recent periods.

Bond & Specialty's net favorable development increased to $89 million, driven by domestic Management Liability, and PI had $13 million of net favorable reserve development, driven by personal auto.

Year-to-date, on a combined statutory Schedule P basis for all of our U.S. subs, while accident years across all of our product lines in the aggregate and all of our product lines across all accident years in the aggregate developed favorably or had relatively small unfavorable development.

Operating cash flows of over $1.1 billion were very, very strong.

We ended the quarter with holding company liquidity of $1.4 billion, and all of our capital ratios were at or better than target levels.

The recent run-up in interest rates that's benefited fixed income NII has, for the first time in many years, resulted in a small net unrealized investment loss that impacted shareholders' equity.

After-tax net unrealized investment gains, which were $1.1 billion at the beginning of the year and a $133 million at the end of the first quarter, moved to a net unrealized loss position of a $112 million after tax at the end of the quarter.

This was a driver behind a 3% decrease in book value per share from $87.46 at the beginning of the year to $84.51 at the end of the current quarter.

I'll remind you that changes in unrealized investment gains and losses do not impact the manner in which we manage our investment portfolio or our business.

We generally hold fixed income investments to maturity, their quality remains very high and changes in unrealized gains and losses have little-or-no impact on regulatory capital.

Adjusted book value per share, which excludes unrealized investment gains and losses, is now $84.93 or 2% higher than the beginning of the year and 3% higher than the end of the second quarter of last year.

We continued to generate excess capital, and consistent with our ongoing capital management strategy, we returned almost $560 million of capital to our shareholders this quarter comprising dividends of $209 million and share repurchases of $350 million.

Year-to-date, we returned over $1.15 billion of excess capital to our shareholders with dividends and share repurchases.

Before turning the microphone over to Greg, I'll point you to Page 19 of the webcast and provide a brief update on our catastrophe reinsurance program, a significant component of our overall reinsurance protection program.

While the structure of our cap reinsurance is generally consistent with the prior year, we did take advantage of the current pricing environment to increase our cat bond limit by $200 million, while reducing our Northeast property cat excess of loss treaty limit by the same amount.

A new $500 million reinsurance agreement with Longpoint Re III has replaced the $300 million agreement that expired in May.

The new agreement provides Northeast hurricane, earthquakes, severe thunderstorm and/or winter storm coverage for certain property losses on specified lines of business through May 24, 2022.

The attachment point and maximum limit will be reset annually.

So through May 24, 2019, the full $500 million limit is available after covered losses from a single occurrence reach $1.9 billion and until such covered losses reach a maximum of $2.4 billion.

And effective July 1, we renewed our Northeast property cat excess of loss treaty, which now provides coverage of $600 million out of $850 million in excess of $2.25 billion.

A more complete description of our cat reinsurance coverage, which also includes a description of our Gen Cat aggregate excess of loss treaty, it covers an accumulation of certain property losses arising from multiple occurrences is included in our second quarter 10-Q, which we filed earlier today and in our 10-K.

So with that, let me now turn the microphone over to Greg.

Thanks, Jay.

Business Insurance produced segment income of $385 million and a combined ratio of 98.8% for the quarter.

The underlying combined ratio of 96.5% was 1.7 point higher than the prior year quarter, driven in particular by non-cat property losses, mostly a small number of large fire claims that Alan and Jay mentioned.

We reviewed every one of these claims looking for underlying trends.

And not seeing any correlation, we view this activity as normal period-to-period variability.

The underlying loss ratio was also impacted by the net impact of small amounts and movements in a number of usual things, including non-cat weather, base year adjustments, earned pricing compared to loss trend as well as business mix.

A couple of other points I'll make on the underlying results.

First, we've been discussing the impacts of earned pricing versus loss trends for some time now.

We're now at a point where the higher levels of the earned pricing about covered loss trend for the quarter.

Secondly, we remain focused on managing expenses thoughtfully, while making ongoing strategic investments and prudently growing the business.

Our expense ratio improved by about a 0.5 point for the quarter when adjusted for the industry-wide assessment from a Texas Windstorm pool related to Hurricane Harvey.

Turning to the top line.

Net written premiums were strong for the quarter at $3.8 billion, up 7% over the prior year quarter with domestic net written premiums up 6%, driven by strong production results across virtually all of our businesses.

Notably, Middle Market was up 9% due to the production results that I'll touch on in a moment.

International net written premiums were up 6%, excluding the impact of changes in foreign currency rates.

Turning to domestic production.

We achieved renewal rate change of 2.1 points and renewal premium change was 5.3 points, while retention remained strong at 85%.

New business of $532 million was up 8% from a year ago.

We're pleased with these production results, particularly considering the pricing pressure in workers' comp associated with strong industry profitability.

Outside the comp, we continue to achieve renewal rate gains broadly across the remaining portfolio.

Auto continues to be the line with the highest level of rate, while property increases continued to accelerate in the quarter.

We continued to execute our pricing strategy on an account-by-account and class-by-class basis with thoughtful balance towards retaining our best business, improving pricing where it's needed and pursuing attractive new business opportunities.

Our results for the quarter reflect our continued deliberate and successful execution in the marketplace.

Turning to the individual businesses.

In Select, renewal premium change was 4.8%, while retention remained strong at 82%.

New business was up 15% over the prior year quarter as we continue to leverage our investments in technology and workflow initiatives.

We're pleased with the returns in this business and our strategic direction.

In Middle Market, renewal premium change was 5.2 points with renewal rate change of 1.9, up from 1.6 in the first quarter and up by more than 1 point from a year ago, while retention remained historically high at 88%.

New business premiums of $315 million were strong, up 6% from the prior year quarter.

As Alan mentioned, we couldn't be more pleased with the impact our strategic initiatives are having on the business.

So all in for the segment, we continue to build momentum in the marketplace from our strategic initiatives and feel great about how we're positioned for the future.

With that, I'll turn it over to Tom to talk about Bond & Specialty Insurance.

Thanks, Greg.

Bond & Specialty's operating results were very strong with segment income of $204 million, up $41 million from the prior year quarter due to a higher level of favorable prior year development and higher earned premiums.

The underlying combined ratio was also very strong at 80.5%, 1.5 points lower than the prior year quarter, primarily reflecting improvements in the expense ratio, also due to the higher level of earned premiums.

As to the top line, net written premiums for the quarter were up 9%, driven by broad growth across our businesses.

These results reflect the impacts of strategic product, marketing and distribution initiatives to grow these profitable lines, and in the case of Surety, also reflect higher average bonded contract sizes and modestly increasing spending in particular public sectors.

Our growth in international was primarily driven by our U.K. Management Liability business.

Turning to production in our domestic Management Liability business.

Given the level of returns we are achieving, we continue to execute our strategy to retain a substantial percentage of our high-quality portfolio while pursuing attractive new business.

So we are pleased that retention again came in at a record of 89% for the quarter and that new business was strong, up 10% from the second quarter of last year.

Renewal premium change of 3.4 points increased from the prior year quarter.

So Bond & Specialty results were excellent, and we continue to feel great about our execution in the marketplace, our growth in returns and the opportunities that our strong market position and competitive advantages present for the future.

And now I'll turn it over to Michael to discuss Personal Insurance.

Thanks, Tom, and good morning, everyone.

The second quarter results in Personal Insurance reflects the elevated level of catastrophe losses already mentioned by Alan and Jay.

While we're disappointed with the resulting net loss for the quarter, we're very pleased with the underlying performance of both our Agency Auto and Agency Homeowners business.

The second quarter results continue to confirm that we are delivering what we intended, a stable volume of auto business with improved profitability and a steadily growing volume of homeowners business with attractive long-term returns.

Personal Insurance reported a quarterly loss for the segment of $17 million compared to income of $12 million in the prior year quarter.

The overall combined ratio was up slightly to 104.9% from last year's 104.1%, but the underlying combined ratio improved nearly 2 points, as catastrophe losses accounted for more of the total combined ratio.

The quarter also included a modest amount of favorable prior year reserve development, which Jay described earlier.

In Agency Auto, the combined ratio for the quarter was 95.4%, down 11 points from the prior year quarter, due, in part, to nearly 3 points of favorable prior year development and about 1 point of cat losses -- 1 point less of cat losses.

The underlying combined ratio was also much improved, down nearly 7 points to 95.5%, as a result of earned rates exceeding loss trend and more of an [earned] loss environment.

Stepping back to look at Agency Auto overall, we're very pleased to see premium growth for the year-to-date of 9%, with the combined ratio on both the total and underlying basis below 96%.

Normal seasonality were most likely result in combined ratio with higher than this for the full year, but we're pleased with the trajectory of this business and remain on track to deliver target returns over time.

In Agency Homeowners, the second quarter combined ratio of 113.6% include 26.2 points of catastrophe losses, related to the wind and hail activity Alan mentioned.

And the underlying combined ratio of 85.2% included about 1 point of impact from the Texas Windstorm pool assessment that Greg described.

Catastrophe losses for the quarter were above both last year's elevated levels and above our expectations, with the late May event in the Northeast and the late June Colorado storm being the most significant for us.

Given the magnitude of these losses and what is now a series of quarters, in which catastrophe losses have exceeded expectations, a few comments are warranted here regarding our analysis of these results.

In the quarter, the primary driver of the elevated cat losses is the nature of the events: their frequency, severity and location, including the localized impact that is characteristic of tornadoes and hail storms.

As Alan mentioned, as we look at this quarter and our experience over recent quarters, we're factoring it into our pricing and underwriting decisions.

As you'll note in our outlook, we expect renewal premium change for homeowners to be slightly higher over the next 4 quarters than the comparable periods in 2017 and '18.

Turning to the top line.

Agency Auto premiums once again grew by 9%, driven primarily by price increases as the volume of policies enforced stayed broadly flat.

We're pleased to see retention holding steady as we continued to achieve renewal premium change in excess of loss trends, although by design, the pace of price increases is trending down from the peaks we reached in the third and fourth quarters of 2017.

In Agency Homeowners and other, premium growth for the quarter accelerated slightly to 6%, with continued increases in both the numbers of policies enforced and renewal premium change.

Retention held steady at a very strong 86%.

The rollout of Quantum Home 2.0 continued with another 4 states during the second quarter.

Another 6 states will launch next month.

As we've stated in the past, the introduction of this new product will continue -- contribute to the sustained momentum, and we continue to be encouraged that its flexibility, sophistication and ease of use are being well received by our agents and customers.

And with that, I'll turn the call back over to Abbe.

Thanks, Michael, and we're ready to take your questions.

(Operator Instructions) Your first question comes from the line of Jay Gelb from Barclays.

My first question is on the investment portfolio.

Given the effect of the lower tax rate for corporations.

I was hoping you can just update us on your views, in terms of whether Travelers may consider reducing its major allocation to municipal fixed income investments in the fixed income portfolio?

Yes.

Good morning, Jay.

It's Alan.

I'm going to ask Dave Rowland to answer that question.

Sure.

Thanks, Alan.

We have not really seen a reason to change our strategy overall.

We continue to find relative value and points on the curve where municipals are more attractive to us on risk-adjusted after-tax return basis.

And so we continue to buy municipals and plan to do so in the future.

Okay.

And then my next question is on the recent news with Johnson & Johnson and their emerging asbestos exposure around talc.

They -- I mean, just last week, they had a $4.7 billion legal judgment against them that is tied to asbestos.

So I was wondering if you can give your broad thoughts on that issue as it relates to talc and J&J, and whether Travelers might have exposure there.

Sure, Jay.

Thanks for the question.

As you can imagine, we're always following -- followed the emerging issues, and talc has been on our listed issues to follow up for a long time.

We don't comment on any individual insured.

That wouldn't be appropriate.

But we will say as it relates to that issue and virtually any other issue, maybe not even virtually just every other issue.

All the news that's in the public domain and all the news that we're aware of in the public domain or not is reflected in our thought process that goes into reserves.

So I think I'll leave that there.

Your next question comes from the line of Randy Binner from B. Riley FBR.

I had a question just on top line, which was good in the quarter.

I think there were some comments on it in your opening remarks and script.

But I just like to understand better if this better net premium return growth is a reflection of better economic activity?

Or if you feel that your capturing share from competitors?

And if it's the latter, just kind of elaborate on what the market dynamic is there?

Yes, Randy, it's a broad question because I think you got us.

I'm just looking at it business-by-business across the place.

I mean, you can definitely see in the exposure numbers.

It's at a healthier clip than recent periods this quarter, and we do look at that.

And there is nothing really unusual going on underneath.

So we do attribute that to economic activity, but there are other strategic things going on in the business that we think are driving premium growth.

So we spent some time at Investor Day talking to about the things that Greg and his team are doing in Business Insurance.

All designs to create top line opportunities for us with no change in appetite, no change in approach to risk, but just the value proposition that we offered to our agents, brokers and customers and it's still reasonably early days, but the traction has been good and that's reflected in the new business numbers you see in the Middle Market there.

And I say the same thing for Tom's business.

Tom's lot of strategic things underway.

I think he also talked about most of them at the Investor Day last fall, and we think that's making a difference.

But these seismic changes in market share, I doubt it.

But I think it is contributing to really healthy production and really healthy earned premium results.

Yes.

I guess, the follow up there is -- and we'll see how this develops over the earnings season, but these are your very large market and these are large gains.

And I'm thinking more in Business Insurance in that Slide 9 in particular in the slide deck.

Is it -- it seems like it's a combination of things?

I appreciate that answer, but is there a market out there?

Is there another competitor who is back in a way from key lines where you are focused on?

Or is it really just an even mix of all the things you talked about?

Randy, I think it's an even mix.

And let me just reiterate something that I said in my prepared remarks, which is when we look at the change in written premium quarter-over-quarter, for example, there's a large part of that's coming from retained dollars.

So that's an equivalent retention rate on a higher base of expiring premium and rate.

And so we view that as a very positive thing and a reflection of a lot of hustle by a terrific field organization and great data analytics at the point of sale.

But we think that really speaks to the quality of the business that we're putting on, and again the fact that we've got no change in appetite, no change in our approach to risk, just really solid execution and again account-by-account basis, whether that's on a retained basis or a new basis.

There is some exposure on that 5.3 point surprise, and again, we do think that's a function of economic activity, and there is some reason there.

So I think in terms of what we're doing in the marketplace production top line, Randy, we feel great about it and firing on all cylinders.

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

My first question, in the 10-Q, in your outlook, you guys pointed the potential for the imposition of tariffs and other barriers to international trade could potentially lead to higher-than-expected levels of inflation.

Alan, if you could expand upon that comment?

What lines you potentially see being impacted over what time frame?

And how that translates into your margin outlook for your businesses?

Yes.

Good morning, Elyse.

Thanks for the question.

Obviously, tariffs and conversation about trade generally is in the news, and it is completely appropriate to point it out.

Tariffs presumably would impact the inputs that go into loss cost as there are lot of things by the way that impact the way we think about loss cost.

I think, generally, we would look at that as affecting short-term lines and things that we could react to.

So we don't look at it and -- there is no sort of outsized expectations here, but we're watching it.

There's a lot of rhetoric out there.

And so we're just anticipating what could happen.

I think the important takeaways are largely that short-term lines and we can react to it.

And you and anybody can have a view on what those tariffs are likely to do to economic activity, which we think has much leverage for us positively or negatively as on the inflation side of things.

So that's the way I think about tariffs, Elyse.

Okay.

And then my second question on the renewal rate change, if back out comp, was about 3.6% in the quarter, up about 30 basis point sequentially.

So obviously, a little bit of a slowdown quarter-over-quarter than we saw last quarter.

Can you just comment how you expect the rating environment to continue from here?

Maybe by [aligning], getting a lot of rate and commercial auto and property, but how do you see that all coming together?

And do you expect next quarter will the number be higher than 3.6%?

And then, also as we think about getting into 2019, just how do you see those numbers progressing from here?

Yes, Elyse, as always, we give you some texture on outlook in the 10-Q for renewal price change, and so we're probably not going to get more granular than that.

There's not much more competitively sensitive than our pricing strategies.

I'll make some observations, though.

We do see a steady trend over the course of the year.

Actually, probably going back to the beginning of 2017, we feel pretty good of that -- about that.

The breadth of where we are getting the rate gains is -- continues to widen.

We think that's a good thing.

Retention hasn't shown any stress at all.

I mean, where we are in the mid- to high-80s, we had a long way to go before we'd feel like we were sacrificing retention but we haven't.

Retention has held up.

And so we feel pretty good about it.

We are not -- as I always remind you, we are not executing for a headline number.

We're executing one account at a time to achieve a targeted return on that account and on our portfolio.

So all in all, we feel pretty good about that.

So maybe I'll leave it there and happy to respond to a follow-up if you have.

No.

That's helpful.

But just one point of clarification, in Business Insurance, did I hear the comment correctly that the higher you guys are insinuating that earned prices is covering trend or is it just at those lines getting the highest level of rate?

I think what Greg's comment was that -- and this is for all of BI that earned pricing is about covering loss trend.

And so we've had a bunch of quarters in a row where we've told you that loss trend outpacing pricing has been a contributor to the underlying combined ratio.

This is the first quarter and probably in a while we haven't mentioned that because the impact really is quite small.

And that's on an earned basis, not a written basis.

Obviously, written basis would be more favorable.

Your next question comes from the line Kai Pan from Morgan Stanley.

My first question is on the BI underlying combined ratio deterioration 170 basis year-over-year.

If you take out the, sort of, like, a $45 million of large commercial fire losses as well as a $9 million of assessment from Texas Wind pool, that's about a 1.5 points.

So in fact, you're flat year-over-year.

I remember last year's second quarter, you had a higher, basically, elevated level of non-cat weather losses.

I just wondered, can you quantify this quarter's non-cat weather losses relatively to both last year's second quarter as well as your historical average?

Kai, this is Greg.

Yes, we try to give you all the pieces there so you could do the math on that.

On the combined ratio, as we said, yes, the large losses were running in just over the 100 basis point, but we have got to the 30 basis points.

And so the gap that's in the remainder is the normal margin of variability that we would expect from quarter-to-quarter.

So we try to give you all the pieces on that so you could project that going forward.

So Kai, Greg mentioned all the -- a bunch of the sort of smaller items that -- those things move around every quarter.

They are not predictable.

And we try to do our best to give you a sense in the outlook section as good as we can, but there's obviously a lot of estimation in that.

And obviously, whether it's base year or small weather or whatever, those things are going to move a little bit positives and negatives, and you tally them up, and there's a net impact.

But it's that kind of thing.

So your outlook suggest that this second quarter, the non-cat weather losses is larger or higher than the normalized levels?

Say that again, Kai?

So your outlook said the second half assuming more normalized non-cat weather losses, which means, like, is the second quarter this year is actually have higher non-cat losses?

If I understand the question right, Kai, in Business Insurance if that's what you're asking, non-cat weather was -- are you talking quarter-over-quarter relative to our expectations?

Just relative to your expectations.

It was slightly favorable in Business Insurance this quarter.

But Kai, this is Jay.

What we're referring to is not just the non-cat losses.

We're just talking about non -- variability -- I'm sorry, return to whatever due is of normal for these things that will vary from quarter-to-quarter, non-cat losses being one of them but we sort all other loss activity.

So we weren't trying to just spike out the non-cats.

Okay.

I just wanted to, sort of, like, figure out that it is -- is it the randomness or is that the loss cost trend above your earned pricing?

It's a variety of things for which there's always going to be normal variability.

There is no fundamental change in our view of loss trend in the quarter.

Okay.

My follow-up question is on the inflation.

Aside for tariff impact, we've also seen higher litigations as well as settlement in defense cost.

And how do you think about loss cost trend going forward?

And have you discounted that in your reserves as well as pricing going forward?

I'd like Tom to comment on that as it relates to the Management Liability.

But Kai, as I imagine that, that might be your focus.

But I will tell you that we certainly see in our own book and read about all the trends that you see and read about that's reflected in, of course, our pricing and our reserves.

And there's nothing in the quarter that surprises us.

But Tom, why don't you comment on...?

Yes.

In our Management Liability book, and again, we read what you read, but actually we have nothing that is out of the ordinary compared to recent periods when it comes to our own legal expenses.

So normal inflation in there; we're definitely seeing that.

But we have not experienced anything that is outside of what we have experienced in recent periods.

And Kai, that's not to say that our view trend is 0, right?

We're aware of everything going on out there.

We think everything going on out there has a potential to impact those sorts of things, and it's reflected in our view of profitability and enterprising.

Your next question comes from the line of Amit Kumar from Buckingham Research.

One or 2 questions.

First of all, maybe I'm just a bit confused here going back to the answer to Kai's question.

So if we ex out the noise in BI, including fire adjustment and the TWIA adjustment, are you suggesting that the underlined margin improvement -- that there was underlying margin improvement and it was in line with what you said in Q1 10-Q or not?

No, that's not what we were saying.

If you take the underlying combined ratio and you eliminate, for example, the fire losses, the observation was there are a number of factors, and Greg mentioned the significant ones in his prepared remarks, that, that impacted unfavorably the period-over-period underlying combined ratio.

So the net of all those things -- some were good guys, some were bad guys, but the net of those things was a quarter-over-quarter negative.

Okay.

I guess maybe I'll tie that into the second question to that.

If you look at sort of the underlying pricing discussion and you talked about some of the broader segments, can you also talk a bit more about comp trends, and what's going on that front?

And how does that impact the underlying loss ratio going forward?

Let me just follow-up quickly on your -- my answer to your prior question.

I just want to make sure I'm clear about this.

There are always going to be things in a quarter that are variations from our expectations because our view of base year is going to change or the underlying weather is going to change or all the things that Greg mentioned.

There are always puts and takes in that process.

But what we're telling you is, there is no fundamental change in the underlying trend.

Those are just things that are always going to move one way or the other.

That's the color I was trying to give you.

Can I just follow-up to that, and we will drop that second question.

So what you were saying is, this was the volatility like ex that out.

So in the guidance and the discussion and all the questions, we should expect an acceleration in the gap between pricing and loss cost, which will be at a faster pace in Q3 and Q4, and hence result in 2018 margin improvement versus 2017?

Is that fair or did I muddle up the thought process?

So if you look in our outlook and the 10-Q, we give you a view on what the underlying underwriting margin and underlying combined ratio is going to do compared to the same period in 2017.

So you can take a look at that.

A significant part of that is, we are assuming that putting this quarter aside what we saw in the corresponding periods in 2017, where we had adverse small weather and we had adverse large loss activity, where our view going forward for the second half of '18 is that those things will return to more normalized levels.

Fair enough.

Abbe will talk to you about that off-line and make sure that...

Yes.

That's fine.

Yes.

I think I follow what you're saying.

I think maybe there's an extreme focus on parsings for the language of the Q and then comparing it to the reported results.

But what I understand from your answer is, there's always this additional volatility which one should factor in when looking into underlying trends.

Is that fair?

Yes.

We're giving you our best view of outlook in the 10-Q, but there are things that are going to change in small ways.

And there is a number of those things this quarter, and they net -- you can do the math, take out the large losses, but they net to what they net to.

Your next question comes from the line of Greg Peters from Raymond James.

I appreciate some of the answers you have provided before, Jay, and I don't want to get too hung up on 1 quarter's results.

But the trend in PYD has been down for a couple of years.

And now that earned pricing is essentially covering trend in BI, is it fair to extrapolate that going forward maybe PYD stabilizes at these levels?

Or perhaps you can use this opportunity to update us on your views of PYD?

I think my response is going to be something you're not going to feel all that great about.

In the sense that we don't have a view and have never had a view of future PYD.

Every quarter, we do a very, very thorough and diligent review of all of our reserves to make sure that whatever trends we're seeing, whatever changes in those strands we're seeing, whatever data we're seeing is not only factored into the reserve estimates but factored into our pricing on a very current basis.

So we make no predictions whatsoever as to the future of PYD.

Now looking back at history, you can see that we have had considerable amounts of PYD in some periods and less PYD in other periods.

So it has fluctuated.

There is no change in the way we think about the business.

There's no change in the way we think about reserving.

So I think you can draw your own conclusions as to what might happen in the future, but we make no predictions.

Our view is we need to get the reserves right.

They represent management's best estimates.

And when new information comes up in the future, we adjust them.

Okay.

I wanted to give it a shot.

My second question, in the past, you've talked about organic innovation.

And also I realize there is this tremendous loyalty to the independent agency channel.

But -- and I've asked you guys about this before, but insurtech seems -- there is a lot of insurtech noise in the marketplace.

A lot of it seems focused around disintermediation of the agents.

And I was just wondering if you can update us on your perspectives around direct operation that's not only in personal lines but business?

Sure, Greg.

We -- you are exactly right.

We have a very active innovation agenda going on.

We do that a very important part of our -- the strategic work that we're doing, and we think it's important for us and for this industry.

So we're hard at work.

Yes, we have, for a long time, been an independent agent broker company and I think we will for the foreseeable future.

I don't think that distribution channel is going anywhere.

And I would say that as we think about insurtech, it goes way beyond distribution.

There are opportunities to invest and think about opportunities whether it's in collection and leveraging new data sources or whether it's in leveraging artificial intelligence, whether it's in using drone technology to reduce claim cost.

We think very aggressively about innovation across the entire value chain.

And so from marketing to claim settlement and everything in between, we do think that distribution is obviously an important part of that, and we're highly focused on it.

We brought Simply Business last year because we knew that advancing our capabilities as it related to engaging with Commercial Accounts was going to be important, and we needed to advance our capabilities.

At the moment where you see distribution on a direct basis in commercial insurance, it's micro.

And that's not just us, that's the industry.

It's really, really small stuff, like, hundreds of dollars of premium, but we do see something of a trend there.

And we are not going to miss that.

We'll see where it goes, and we'll see how fast it goes, but we're on top of it.

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

First, maybe a clarification to your answer to Elyse's question on the inflation commentary in the Q, and I apologize if I missed it in your answer.

But if I look at the commentary around inflation maybe being a negative for next year and then I look at the commentary around business unit margins being stable or consistent with this year or possibly even better in Business Insurance.

Does that -- there's a commentary around the business unit margins incorporate the risk of inflation going up?

The answer is yes.

I mean, clearly, our view of margins does incorporate a view of loss trend.

That has a view of severity which in turn has a view of inflation in it.

I'll just remind you that our results are most heavily leveraged to medical wage and toward inflation.

So we're not as highly leveraged to the typical sort of CPI kind of inflation that impact short-term lines and that we think we can react to.

So -- but whether it's CPI-type stuff then the potential impact of tariffs or whether it's wages or whether it's medical or social inflation, whatever it is, our view of it is reflected in our outlook and margins.

We could be surprised, it could be worse than we think, of course, but we do have a view of it.

And we look at it on a very, very granular basis aligning to the individual components of our loss cost.

So we have a pretty sophisticated process of looking at it, and it's incorporated in our view.

Got it.

That's helpful.

And if I could turn to net investment income, again going back to the Q for a second, I think you're guiding to $20 million to $25 million increases in 2019 per quarter.

Greg, maybe you can help me understand what was the duration of over 4 years?

And with premium growth accelerating, why wouldn't that number be higher a year out relative to where it was in -- sorry, go ahead?

Sorry.

It's a good question, but you have to realize that it's really being driven by the actual securities that are maturing in a particular period of time and assumptions associated with the reinvestment rate that's going to be available at that time.

And it will be determined by the actual investment opportunities then and what the yield curve looks like and where interest rates are in general.

But at this stage, that's our best view as to what's maturing and what the interest rates might be.

Your next question comes from the line of Sarah DeWitt from JPMorgan.

On the weather-related losses, and that includes the cats and non-cat weather, you've seen elevated weather losses for the last several quarters that you pointed about.

So at what point do you call this the new normal and start pricing for us?

Yes.

So the question is, over what cycle do you look at changes in weather patterns?

And so just to take you back on a slightly longer-term view, if you go back to 2009, whether it was better than we expected, 2010 to 2012, few year period, that was worse than we expected.

Then 2013 to 2015, I think, it was better than we expected.

2016, it was about on plans.

So my point is, you look back over time and you have some good periods and you have some bad periods.

And so we look at what's now been a 4- or 6-quarter period, and we're not going to overreact to it, but we will react to it.

So the first thing we'll do is we will update our actuarial and weather models for the actual activity.

So just that alone rolling through our models will change our view of the world if we did nothing else.

As we are always looking at recent weather activity and trying to decide how to wait more recent periods as compared to longer periods, that's an ongoing process here and part of our thought process around weather.

Again, we wouldn't look at 4 to 6 quarters and want to overreact to that.

And then we always look at lessons learned from weather activities.

And there are always lessons learned anytime you have catastrophes like this.

So we look at it, we take it in stride.

We take it into account.

We put into our models.

But we would not look at a year or so and say that we've seen a fundamental change because we've seen this before.

Okay.

Great.

And just following up on the Business Insurance pricing.

Now that earned prices are in line with loss trends, will you still push for higher rates versus the 2% you achieved in the quarter?

Or is that now sufficient to achieve your targeted returns all else being equal?

We will push for rate in those lines and on those accounts where we need it, and that's the way we always execute.

We got lots of accounts in our book of business that are rate-adequate, and we're going to try really hard to keep those accounts.

As always, there is some portion of our book where we look at an account and say, "That needs rate", and we will continue to try to get rate there, and the aggregate of those individual account decisions will aggregate up to a number.

I would say there's sort of a lot of things out there that are working together.

So pricing is better, interest rates are ticking up.

That's on a lag basis will work its way into net investment income.

You've got the benefit of tax reform.

You've got all the other levers that impact profitability, things like selection, mix, claims handling, expense initiatives.

So all of that goes into the hopper, and we're always managing all of those levers to make sure that we're executing towards our return objectives.

Your next question comes from the line of Josh Shanker from Deutsche Bank.

I will give you both of my questions upfront, so you don't think there's a long one coming in the back.

So I just wondered -- no, I'm not going to get more granular I don't think than your guidance in the 10-Q, but when you say improving margins, I'm wondering now that we have the pricing versus loss cost trend out of the way, how much of that is the cessation of the non-cat weather and the fire losses and what not?

And how much of the innovations you were talking about in expense saves -- you've been talking about for a year -- I'm wondering if you can break it between those 2 things.

And the second question, I just want to say one thing about that loss cost trend versus rate.

If net written premium trend in BI is 2.1%, net premium earned trend is probably somewhere less than that I would guess.

Maybe I'm wrong about that.

How -- is that conservative to say that's covering loss cost trend?

I'm just trying to understand that better.

Yes.

So let me take the second one first.

So we don't think -- we think it's appropriate it's [covering] loss cost tranches.

The math we have -- and maybe what you're leaving out, Josh, is the economic impact of the earned exposure, a significant portion of which behaves like rate.

And then, again, we all -- and by that I mean you all get very, very focused on this very narrow rate versus loss trend.

And again, that ignores the economic component of exposure that behaves like rate and it ignores all the other things that go into improving margin.

And again, I just mentioned the list that things like selection and mix and claims handling and expense initiatives, that's all in there.

So you can't look at that 2.1% in isolation without thinking about all the other factors that impact margin.

In terms of the outlook, a significant part of the improvement is what was adverse small weather and large losses last year returning to more normal levels, but everything is in the number, right?

It reflects everything we know in terms of the outlook, but as we identify in the 10-Q, that is very significantly a return to normal level of small weather and large losses.

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

Two quick ones here.

Michael, I'm just curious could you give us some thoughts on a competitive environment write-down in personal auto insurance?

I think we've seen some major companies reduce rates.

Are you seeing that in the [agency] system?

Thanks, Brian.

Yes, I mean, I would say, certainly, the environment in all of our markets is consistently competitive.

We have certainly seen the announcements that you're probably referring to of one carrier in particular taking rate reductions.

At least at this point, that's one carrier taking rate reductions.

The industry rate remains positive.

When we pull filings and do our analysis that we've talked to you about in the past around what we're seeing, we do see most carriers continuing to file for increases.

Those increases so far this year are not quite as high as they were in 2017.

I think that's all very consistent with the industry data around loss trend and the more benign loss environment that I described earlier.

I think that all aligns up from our perspective.

Great.

And then just one quick follow-up.

Alan, you did mention little bit about interest rates moving up and that definitely can have some impact, I think, on pricing.

Is that flowing through your models at this point when you think about the higher interest-rate environment right now and your kind of thoughts on pricing and profitability by account?

Or is it still too early for that to kind of have an effect?

Yes, we update our models on a pretty regular basis, and they reflect everything that we know we're aware of or think about on an outlook basis.

It's hard to look at any one element in our pricing model and isolate it, but we do have a view of inflation and we put it in there.

You're talking about interest rates, you think about net investment income, in particular.

I mean, I would say, you need to think about the fact that, that will flow through our fixed income portfolio on a lag basis because as Jay Benet just explained, the portfolio has to turn over.

So the impact of -- the favorable impact of interest rates will be muted by that fact.

Right.

But I'm talking more in pricing decisions than commercial line.

And is the fact that you typically look -- typically some of your longer developed lines?

Say that again.

I'm thinking more on pricing kind of decisions with respect to commercial lines in which your rate activity looks like -- as far as the model and putting in interest rate assumptions, are we at a point where that maybe having an impact bought on pricing?

Brian, this is Jay.

Within our pricing model, what we said in the past, we take into account not an embedded rate in the portfolio when it comes to looking at the product cash flows.

We look at the product cash flows associated with a current interest-rate environment and a view towards the future, with regard to how this is -- how things are being invested along the yield curve.

So the pricing models themselves would be updated based on current views of interest rates.

However, having said that, it's not done in isolation as Alan is talking about, and one of the things that you have to look at is, well, what are your return expectations, the written expectations for products change when the 10-year treasury is at 2% versus when the 10-year treasury is a 3%.

So it's not just that it goes in and immediately has a point-for-point, dollar-for-dollar impact on the pricing.

It's a very thoughtful approach to it.

The last question comes from the line of Jay Cohen from Bank of America Merrill Lynch.

Just one final question.

The non-fixed income yield, do you view that as better than you normally expect, a little worse or in line for the quarter?

No, it was actually -- when we think about what the -- how that portfolio is doing, we like its performance.

And on a long-term basis or a planned basis, we think it's performing well.

It's just not performing as well as it did in the prior year quarter.

But if you look at it on a quarter-to-quarter basis, I think you'll see the second quarter is very much in line with some recent quarters.

And as you know, Scott, combination mostly a private equity and real estate partnerships and real estate owned.

So there's going to be some fluctuation in it, but we like the performance of the portfolio.

But not far often what you would normally expect?

Actually, it's probably a little better than what we would normally expect.

Great.

Well, thank you all for joining us.

This concludes today's conference.

You may now disconnect.