Allstate Corp (ALL) 2016 Q2 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Allstate second-quarter 2016 earnings conference call.

(Operator Instructions) As a reminder, today's program is being recorded.

I would now like to introduce your host for today's program, Pat Macellaro.

Please go ahead.

Thank you, Jonathan.

Good morning and thanks, everyone, for joining us today for Allstate's second-quarter 2016 earnings conference call.

After prepared remarks by our Chairman and CEO Tom Wilson, Chief Financial Officer Steve Shebik, and myself, we will have a question-and-answer session.

Yesterday following the close of the market, we issued our news release and investor supplement, filed our 10-Q for the second quarter, and posted the results presentation we'll use this morning along with an update to our 2016 countrywide reinsurance program to reflect the replacement of our Florida program.

These documents are all available on our website at allstateinvestors.com.

As noted on the first slide, our discussion today will contain forward-looking statements regarding Allstate's operations.

Allstate's results may differ materially from these statements, so please refer to our 10-K for 2015, the slides, and our most recent news release for information on potential risks.

Also, this discussion will contain some non-GAAP measures for which there are reconciliations in our news release and our investor supplement.

As many of you know, this will be my final earnings call as the leader of our Investor Relations team, as I have transitioned to leading our Encompass team, leaving investor relations in the capable hands of John Griek, who will be a great partner for all of you going forward.

John and other members of our senior leadership team will be available to answer any follow-up questions you may have after the call.

Now I'll turn it over to Tom.

Good morning.

Thank you for investing your time to keep up with Allstate.

I'll provide an overview of results and then you will hear from Pat and Steve.

We also have Matt Winter, our President, here with us; Don Civgin, the President of Emerging Businesses; Mary Jane Fortin, who is President of our Allstate Life and Retirement; and Sam Pilch, our Corporate Controller.

Let's begin on slide 2. The second-quarter results highlight really how our proactive approach to external conditions helps us achieve our objectives and create long-term shareholder value.

We earned $242 million, despite seasonally high second-quarter catastrophe losses, a record hailstorm in Texas, continued increases in the frequency of auto accidents, and the impact of Brexit on the investment markets.

And while we did not predict all of these would happen, we considered the possibilities when we executed our business plan.

Operating income was $0.62 per share for the quarter, reflecting a recorded combined ratio of 100.8.

And then underlying combined ratio of 88.6 for the quarter and 87.9 for the first 6 months, which is in comparison to our full-year forecast of 88 to 90.

Allstate Financial's strategic repositioning is working, with operating income up slightly to $120 million.

Investment returns were strong, at 1.9% for the quarter, about half of which is from current income and the other half is from appreciation of the bond portfolio.

Let's go to slide 3, which provides an overview of the second-quarter operating results for our four property liability customer segments.

As you know, we have a consumer-focused strategy and have over 40 million policies outstanding, of which 34 million are for property-liability protection.

The results of each of the four segments of the property-liability market are shown at the diagram on the bottom of the page.

Allstate agencies provide local advice in a broad range of branded products.

It represents about 90% of total property-liability premiums.

Total Allstate-brand policies in force declined from the second quarter of last year by 1% as we intentionally reduced new business levels until we improve returns on capital for auto insurance and the impact of these programs on customer retention.

While auto policies declined, net written premium increased by 3.9%, which was essentially offset the impact of continued increases in frequency and severity.

The recorded combined ratio was 101.2.

The recorded combined ratio for homeowners was 97, despite $645 million of catastrophe losses in the quarter.

The underlying combined ratio was 87.5 for the Allstate-branded business for the quarter.

Moving across to the lower right is Esurance.

Policies in force declined as we reduced new auto business in this segment, but net written premium increased by 5.7% over the prior-year quarter.

The recorded and underlying combined ratios are 108.9 and 104.8, which were 1.3 points and 1.9 points better than last year's second-quarter results.

The underlying loss ratio was consistent with the prior-year quarter.

In the upper right is Answer Financial.

That's our self-serve aggregator that sells products from more than 20 different companies.

Premiums grew by 3.4% in the second quarter, lower than the prior years, which reflects fewer leads from Esurance as we?ve reduced growth at Esurance.

Independent agencies serve customers who want local advice, but do not have a high affinity for branded products.

Encompass in the upper left serves this segment and has taken significant actions that have reduced the size of the business while simultaneously improving returns.

Policies in force were down 11.4% -- you can see in the red box there at the top.

Net written premiums declined by 6.8% as higher prices partially offset the decline in policies.

The recorded and underlying combined ratios both improved in the quarter from the prior-year quarter.

As you know, we establish and communicate our operating priorities each year.

And they range from doing a better job for customers to driving long-term growth, which are shown on slide 4.

The first three priorities are all interrelated.

We are of course always working to provide customers even better value and service.

This ranges from the rapid response to catastrophe that we've become known for as well as our agency owners being trusted advisors in helping customers prepare and protect themselves from life's uncertainties.

We've had over 215,000 catastrophe claims through the end of June and have closed approximately 95% of those.

We are continually focusing on improving service from Allstate agencies and have made great progress in initiating relationships with customers.

Achieving target returns on capital, however, has required us to raise prices on auto insurance, given a greater frequency of accidents and increased severity of claim costs.

So we've responded with a comprehensive profit improvement program which prioritizes existing customers ahead of new customers, leading to a significant reduction in new business volumes.

It has also had a negative impact on customer retention levels.

This has resulted in a 1.4% decline in the number of property-liability policies in force.

Allstate Benefits, on the other hand, added almost 0.5 million new customers over the last year, offsetting the property-liability decline.

But all of our businesses have the potential for growth over the long term.

As you know well, the investment markets this year have been highly volatile from the early decline in energy prices to a rebound and then down again, Brexit, you have record low interest rates, and global equity markets that have been moving up, down, and sideways.

Our investment strategy and execution have served us well in this environment with a total return of 3.9% for the first 6 months of the year.

We continue to build out our performance-based asset team and have made progress in increasing the amount of these assets backing the $12 billion payout annuity block.

As we discussed on last quarter's call, this is economically the right risk and return trade-off for shareholders, given the long duration of these liabilities.

We've also made progress on building long-term growth platforms in a number of areas, from Allstate Benefits to roadside services to telematics.

For the second quarter, we wanted to highlight where we stand on telematics, which is shown on slide 5.

Allstate began investing in the telematics space over six years ago as a way to improve our business model by serving customers in new and different ways.

And we've made a tremendous amount of progress over this time period.

First, telematics is a very powerful pricing tool that enables us to give customers the most accurate price.

Secondly, we?ve found ways to use a continuous connection to broaden the value they get from Allstate and provide additional services to them.

For example, we provide customers with tips on how to better protect themselves by changing their driving habits.

We've also expanded the value they get from being with Allstate, such as merchandise discounts based on how safe they drive.

The Safe Driver Rewards are very popular and expand that customer value proposition.

We've built up a wide range of capabilities, have made significant investments, and today we have over 1 million connected customers.

We've also concluded that this could be a strategic platform for Allstate, the concepts of which are laid out in the annual report to the shareholders.

As a result, we established a connected car entity, Arity, outside of the insurance company, which you can see in the graphic on the lower left of this slide.

This provides us with the strategic and operating flexibility to capture additional value from our growing connected customer base by providing other companies the ability to offer services to our customers.

It also enables other companies to connect with their customers by using this platform and Arity's capabilities.

We believe that the transformation of the personal transportation system will be one of the largest economic benefits for individual households in the future.

And this structure enables us to participate in that value creation.

Now let me turn it over to Pat.

Thanks, Tom.

I'll start by reviewing the property-liability P&L at the top slide 6. Property-liability earned premium of $7.8 billion in the second quarter of 2016 was 3.5% higher than the same period last year.

Through the first six months of 2016, earned premium grew by 3.8%.

Catastrophe losses through the first six months of 2016 meaningfully impacted underwriting income.

Second-quarter catastrophe losses of $961 million were 20.6% higher than the prior-year quarter, while catastrophe losses of $1.8 billion for the first half of 2016 were almost $700 million higher than the first 6 months of last year.

These higher catastrophe losses drove recorded combined ratios of 100.8 in the second quarter of 2016 and 99.6 for the first half of 2016.

When we exclude catastrophes and prior-year reserve re-estimates, the underlying combined ratio of 88.6 in the second quarter and 87.9 in the first 6 months of 2016 were both below their respective levels in 2015.

The June year-to-date result is slightly below our annual outlook range of 88 to 90.

Property-liability net investment income increased 8.2% to $316 million for the second quarter of 2016, driven primarily by higher performance-based investment income.

As a result, property-liability operating income of $186 million in the second quarter of 2016 was 6.1% below the prior-year result, while the $477 million of operating income through the first 6 months of 2016 was 36.7% below the first 6 months of 2015.

The bottom of this slide contains growth trend information as well as a view of property-liability recorded and underlying combined ratio trends.

In the chart on the bottom left, the blue line represents net written premium growth, while the red line shows our policy in force trend.

Property-liability policies in force declined by 1.4% or $471,000 in the second quarter of 2016 compared to the second quarter of 2015, while net written premium increased by 2.2% in the same time period.

These trends have been heavily influenced by our auto profit improvement actions across underwriting brands.

The widening gap between these two trends reflects increases in average premium per policy, given ongoing rate increases.

The exhibit on the bottom right shows the property-liability recorded and underlying combined ratio along with some history.

As you can see in the red line, recorded results in the first two quarters of 2016 have been impacted by higher levels of catastrophe losses.

Taking a longer period of time into account, the recorded combined ratio on a 12-month moving basis is 96.2.

Slide 7 provides a more detailed view of our Allstate-brand auto margin results.

The chart on the top left of this page provides a view of the quarterly recorded and underlying combined ratios for Allstate-brand auto.

The underlying combined ratio of 97.8 in the second quarter of 2016 was unchanged compared to the second quarter of 2015.

A lower expense ratio offset an increase in the underlying loss ratio in the quarter.

Our early recognition of increased frequency and severity along with the aggressive actions we continue to take have enabled us to keep auto margins stable, despite a continued challenging auto loss cost environment.

The chart on the top right highlights the drivers of the Allstate brand auto underlying combined ratio.

Annualized average earned premium per policy, shown by the blue line, continued to increase as approved rates have resulted in a 5.9% increase in the second quarter of 2016 compared to the second quarter a year ago.

Average underlying losses and expenses per policy in the second quarter of 2016 increased 5.8% compared with the second quarter of 2015.

The positive gap between these two trends narrowed in the second quarter based on ongoing higher frequency, but was consistent with the level we saw during the second quarter 2015.

The bottom two charts on this page provide 20 years of history for Allstate-brand auto gross and paid property damage frequency.

As we've discussed in prior quarters, we evaluate frequency on a gross and paid basis for a variety of reasons, such as managing claim staffing, evaluating cost trends, and estimating our ultimate losses.

We watch both metrics to ensure we can evaluate and react to changes in our results as quickly as possible.

Gross frequency is a lead indicator of future loss trends, while paid frequency helps us understand changes in the proportion of claims we close with a payment.

The relationship between these two measures will fluctuate over time, given environmental impacts and claim department process changes.

As you can see in the charts on this page, both measures are up substantially from where they had been performing in recent history.

For the first 12 of the past 20 years, you can see a fairly steady decline in frequency as the safety of cars was enhanced.

As the impacts from safety improvements fully worked their way into the fleet, we saw flattening trend for approximately five years.

Now, the results we've seen in the past 18 months have taken us back to levels not experienced since 2003 for gross frequency and 2004 and 2010 for paid frequency.

This most recent period reflects just how challenging an auto loss cost environment we continue to operate in.

We first identified the uptick in gross frequency during the fourth quarter of 2014 and our analysis indicated it was being driven mainly by environmental factors unrelated to our pricing and underwriting.

We continue to believe that our early identification of the issue along with our proactive and aggressive response will position us well to accelerate profitable growth as loss trends stabilize.

We continue to implement our profit improvement plan, which is summarized on slide 8. Given ongoing auto loss pressure, we continue to seek approval for higher auto prices.

In the second quarter, we received approval to increase rates by $628 million annually, bringing the total for the first 6 months of 2016 to $963 million, as you can see from the bar chart on the lower left.

Rate increases in the second quarter of 2016 were approved in 35 states and Canadian provinces and were on average 6.2%.

The amount of rates approved for the second quarter of 2016 includes a significant amount of rate increases in large states, which drove the total to be much higher than our previous run rate.

The impact of these rate approvals on average premiums for Allstate-brand auto is shown on the lower right.

Average gross written premium per policy increased by 5.7% in the second quarter of 2016 compared to the second quarter of 2015.

Average net earned premium per policy, which lags written, increased by 4.7%.

Allstate-brand auto rate changes take six months to be fully recognized in average gross written premium, while they take at least 12 months to be fully earned into the P&L.

The significant amount of premium we've generated by seeking approval for auto price increases has served as well so far.

So if we had not moved early, our auto returns would be significantly lower and we would be playing catch-up until well after the loss pressure we and others are experiencing subsides.

We tightened underwriting guidelines in 2015 to reduce new business in underperforming segments and reduce the new business penalty.

These guidelines are being modified for specific segments of business within each state and local market where we feel comfortable that we've achieved rate adequacy.

Our claims team continues to address physical damage severity trends, which are being unfavorably impacted by stress to the auto repair industry from rising industry auto frequency, higher costs associated with repairing newer, more sophisticated vehicles, and greater total loss volume on older model year cars.

Property damage paid severity in the second quarter of 2016 remained elevated at 5.3%, but the trend improved relative to the first quarter of this year.

Property-liability expense ratio decreased by 0.8 of a point in the second quarter of 2016 compared to the second quarter of 2015, primarily reflecting reductions in professional services and advertising costs as well as lower accruals for compensation incentives.

We continue to evaluate investments in growth and would expect to accelerate these investments as loss trends stabilize.

Allstate-brand homeowners results are shown on slide 9. In the chart on the left, you can see the impact catastrophes have had in the first two quarters of 2016, given the gap between the blue columns and the red line.

The recorded combined ratio on a 12-month moving basis was 83.5 as of the second quarter of 2016.

On an underlying basis, continued favorable non-catastrophe losses and lower expenses resulted in a 58.6 underlying combined ratio in the quarter, which was 2.1 points lower than the prior-year quarter.

The components of the second-quarter homeowners underlying combined ratio are in the chart on the bright.

Average earned premium per policy increased by 2.5% over the prior-year quarter, while underlying loss and expense per policy declined by 1.1%.

Slide 10 provides a view of top- and bottom-line trends for Esurance.

I will begin on the left with a summary of the combined ratio.

Esurance's recorded combined ratio of 108.9 in the second quarter of 2016 was 1.3 points better than the same period in 2015, reflecting lower operating expenses which more than offset higher catastrophes and unfavorable auto claim frequency.

As Tom mentioned earlier, the underlying loss ratio of 74.5 remains higher than where we would like it to perform in the long term.

On the right, you can see Esurance's premium and policy in force trends.

Growth in Esurance has been impacted by ongoing profit improvement actions, including rate increases, underwriting guideline adjustments, and decreased marketing in select geographies.

Given these actions, policies, which are represented by the gray line, declined by 1.4% compared to the second quarter of 2015, while net written premium in the second quarter of 2016 grew by 5.7% compared to the same quarter a year ago, driven by higher average premiums per policy.

Encompass's results are highlighted on page 11.

The left-hand chart summarizes our combined ratio trends.

Encompass's recorded combined ratio of 104.9 in the second quarter of 2016 was 10.8 points below the prior-year quarter, driven by a lower level of catastrophes, a reduced expense ratio, and a 2.4-point improvement in the loss ratio, excluding catastrophes.

The underlying combined ratio of Encompass of 92.8 was 3.7 points better than the second quarter a year ago, the result of ongoing pricing and underwriting actions to achieve target margins.

The chart on the right of this page shows how the size of the business has been impacted by profit improvement actions.

Net written premium, as shown by the blue line, declined by 6.8% in the second quarter of 2016 compared to the second quarter of 2015, driven by an 11.4% decline in policies in force, which more than offset higher average premiums from increased rates.

Encompass has continued to take actions to achieve targeted returns by enhancing its pricing, contract coverage, and underwriting sophistication.

Now I'll turn it over to Steve.

Thank you, Pat.

Slide 12 provides an overview of Allstate Financial's results for the second quarter of 2016.

We refocused Allstate Financial over the past several years, primarily on business written to the Allstate agencies and on voluntary workplace products for customers of Allstate Benefits.

The annuity product line is closed to new business and is effectively in run off.

Premium and contract charges totaled $564 million in the second quarter of 2016, an increase of 5.2% in the quarter versus the prior year.

The solid growth in premium and contract charges was driven by Allstate Benefits, which grew 9.6% in the second quarter with an increase of 468,000 policies over the prior 12 months.

Allstate Financial operating income decreased to $120 million in the quarter from $139 million in the prior year.

Across the top of the slide, we show net and operating income for each business.

Life business operating income of $64 million in the second quarter increased $9 million compared to last year, driven by a favorable mortality experience and premium growth.

Allstate Benefits' operating income of $29 million for the second quarter was consistent with the second quarter of last year.

The annuity business generated operating income of $27 million, down $29 million from the second quarter of 2015 due to 2015's portfolio repositioning and worse mortality.

The chart on the bottom right shows the liquidity need by product.

For immediate annuities in the gray bar, benefits will be paid to annuitants over many years, lowering required liquidity.

As a result, we repositioned this portfolio from longer-duration fixed-income securities and are shifting to performance-based investments, which offer higher risk-adjusted returns.

Now let's go on to slide 13.

We've been proactively managing the investment portfolio in response to changes in our liability profile in a low-interest rate environment while reducing the portfolio's sensitivity to an eventual rise in interest rates.

As we've discussed in prior quarters, we have been shifting the risk posture of our portfolio over time, which is shown in the chart on the upper left.

Several years ago, we began to reposition the portfolio by moving away from lending and toward ownership.

This is reflected in the decreased allocation to investment-grade fixed income and increasing performance-based investments, such as private equity, real estate, timber, and agriculture.

Our performance-based strategy is expected to generate higher, but more variable returns over time.

We created capacity for the incremental risk by strengthening our capital position through issuing preferred securities, reducing debt, decreasing exposure to natural catastrophes, and shrinking our annuity business.

Total return for the quarter in the upper right remains strong, at 1.9%, as the valuations continue to be positive, reflecting lower market yields.

The unrealized gain in the portfolio increased to $2.7 billion.

Gross investment income is provided at the bottom left of the slide.

Durability and income largely results from our performance-based investments shown in gray.

Investment yield by business segment are provided in the bottom right.

We shortened the direction of the property-liability portfolio in 2012 and 2013, resulting in an interest-bearing portfolio yield closer to current market yields, which will be more responsive to rising interest rates, as result of the shorter maturity profile.

The Allstate Financial yield trend reflects the impact of last year's portfolio repositioning when we sold longer-duration fixed-income securities in the immediate annuity portfolio.

Except for the impact of these sales, the interest-bearing portfolio yield at Allstate Financial has been relatively stable, despite the rate environment as investment cash flows have been used largely to fund liability outflows.

The yields for both segments continue to be pressured to the extent rates remain below the portfolio averages.

Slide 14 provides an overview of our capital strength and financial flexibility.

We finished the second quarter of 2016 with $20.6 billion of shareholders equity, a debt to capital ratio of 19.9%, and deployable Holding Company assets of $2.5 billion.

Book value per common share for the second quarter of 2016 was $50.05, up 4.4% from 2015, reflecting higher unrealized net capital gains.

We returned over $1 billion in cash to common shareholders through the first 6 months of the year.

We paid $240 million in common share dividends and repurchased $829 million of common shares.

On June 1, we entered into an accelerated share repurchase agreement to purchase $350 million of our outstanding common stock.

The completion of the ASR agreement will be on or before September 23 of this year.

As of June 30, $1.2 billion remained on the $1.5 billion common share repurchase authorization.

With that, let me ask Jonathan to open the line for your questions.

(Operator Instructions) Greg Peters, Raymond James.

Good morning, everyone.

And congratulations, Pat, on the promotion.

I wanted to focus on two areas: one, on the rate change in your auto, and then secondly on distribution platform.

In addition to the comments you made on rate change, I was looking at slide 14 in your supplement.

And the auto for Allstate brand was up 3.2% in the second quarter.

That's a noticeable change from previous quarters, and I'm curious if there was some geographic concentration with the rate change or is it broad-based?

I think the annualized pace of that is in excess of 12%, so I'm also curious about any regulatory pushback or competitive issues there, too.

Greg, do you want to give us your distribution question, too, so we can -- maybe they link together, I'm not sure.

But we can handle of them both that way?

Sure.

On the distribution side, I noticed that you -- the total number of Allstate agencies, licensed sales professionals, and independent agents all increased on a year-over-year basis.

And I'm just curious if you can update us on what you're doing there and how you are able to grow that business in a challenging marketplace?

Okay.

Well, Matt can deal with both of those.

Let me just give him a little bit of air cover, which is we always encourage you not to take our quarterly results and multiply them by four.

(laughter)

Good morning, Greg.

It's Matt.

Thanks for the questions.

I think as Pat said during his opening remarks, some of the quarterly rate change was driven by the fact that we took rate this past quarter in some extremely large states.

And that impacted it disproportionately when you look at it on the countrywide impact.

So you are correct: it is disproportionately high to what you have been seeing.

And I certainly wouldn't extrapolate out that number for the rest of the quarter -- for the rest of the year.

However, I would be remiss if I didn't say we are going to continue to monitor rate need and rate indication and take appropriate levels as we have.

When it is primarily in smaller states, it has a smaller impact on the countrywide average, but when it's in larger states, it will have a larger impact.

And so it will fluctuate because it's based upon need and indication in particular geographic areas as opposed -- which has a countrywide impact.

But you can't think of it as a countrywide rate increase.

The second question on distribution.

It's been a tough period, yet we've been able to hold fast and slightly grow some of our distribution points of presence.

It's hard to bring on new agency owners in this environment when you are taking a lot of rate when there's a great deal of activity going on that is requiring them to respond and do things other than merely attempt to get new customers.

Our termination rate and turnover rate has not changed.

We have added fewer in the recent past, but we expect as this activity dissipates and lessens, we'll be able to add additional agency owners, additional LSPs, and, I would add, additional financial specialists.

Because we think the fact that we entered this phase early, took action quickly and aggressively, will allow us to emerge quickly and position us for growth and continued growth in points of presence as well.

I hope that answers your question.

It does.

Thank you very much for the answers.

Greg, let me add a longer-term view to that as well.

So we have 10% of the overall auto market.

We have a bigger share of the lower left, but not such a big share that we don't think there's plenty of room to grow.

Got it.

Thank you.

Elyse Greenspan, Wells Fargo.

I was just hoping to talk a little bit about the expense level.

It did rise a little bit in the quarter.

And I know you can look at it year over year, but it rose from how it had been trending the past few quarters on a sequential basis.

Can you just comment on what's driving that?

And I know last quarter, you guys had spoken to some advertising programs rolling out in the later stages of this year.

Is that still the case as we continue to see the elevated-frequency trend?

As you noted, we did decide we would reduce those expenses that were related to growth when we were not in the market for new customers.

So Matt just mentioned, for example, not adding a bunch of new agency owners.

That costs money to do that because we have to help them get started and get supported.

We did launch our new advertising program already.

And so you've started to see that uptick in that portion of expenses.

So the expense reductions are not as great as they were in prior quarters in the year-over-year comparison.

We manage it both on a short-term and long-term basis.

We try to control our expenses in total to make sure we're achieving our target returns.

On the other hand, we're not willing to give up our long-term growth.

So I would just say it's balanced.

It's hard to predict where your expenses and your frequency and severity will be in a quarter, obviously, on a forward-looking basis.

So we manage it on sort of a rolling 12-months basis.

Okay.

And then on slide 7, where you guys kind of go through the frequency trends, I was just trying to tie together some of the trends on both the gross and the paid side.

So the paid trends improved in the quarter, while the growth trends continue to remain elevated.

I'm just trying to kind of tie together why the improvement on the paid side wouldn't necessarily be an indicator that incurred-frequency trends are improving.

And then also within the auto book, was there any kind of seasonality within the underlying numbers in the quarter as we think about the margins in the back half of the year?

Elyse, it's Matt.

I'll try to answer both of those.

I think the graph -- the side-by-side of gross and paid is informative because we've tracked both and reported on both.

They serve different purposes, as Pat described, and they show up differently.

Clearly in gross, it's a leading indicator.

So you are reporting on and showing everything that's coming in that might be a claim.

And in fact, this is subject to every single change in potential opening practice in claims, how we think about it, how we capture information.

And clearly, there's some segment to this that is closed without payment.

But for these, and especially in the most recent difficult frequency environment, we're trying to capture as much information as possible with as much granularity as possible and predict claims need, claim staffing need, so that we can provide our customers with the absolute best service in the time that they need it.

So it tends to lead and it tends to overcapture, obviously, what is going to be on the paid side.

And you have this gap -- it?s especially true on BI because of the long tail and that's a timing issue.

But you have a natural gap in just claims closure.

You also see some small areas where it seems out of pattern, Elyse, but it's really a question of catching up.

It's a question of closing out and paying a bunch of claims.

It's a question of figuring out when demand is.

We see this not just here, but you see it on subro as well.

Sometimes we'll get hit with a big lump of subro claims from third-party carriers because they just backed up, and all of the sudden, they come in.

And so that kind of -- I refer to it as peristalsis in our claim system -- sometimes there's a catch-up as you had a backlog, things slowed down, and then it's pushed through.

You manage to clear a lot of claims.

And so it will never mirror it exactly, but it's eerily similar if pushed over a little by time and dampened a little in terms of its volatility and variability.

Okay.

Thank you.

And then just lastly, was there any kind of seasonality within that auto underlying numbers in the quarter as you kind of think about the back half in the Q3 and the Q4 of 2016?

Yes.

Well, I'd encourage you to look back historically on quarter-by-quarter underlying.

I don't want to draw a bunch of general conclusions for you, but I think you will see there is seasonality not only in the recorded, but there is always some seasonality in the underlying based upon weather and other activities.

And so I'd encourage you to look, and we do have the historical graphs, and that's apparent from taking a look at that.

We've not changed our full-year outlook, which is the way we like to look at it.

Because as Matt points out, it bounces around by a quarter.

Our full-year outlook is still 88 to 90 for underlying combined ratio.

Okay.

Thank you very much.

Ryan Tunis, Credit Suisse.

Thanks.

I guess the first one, just following up on some of Elyse's questions on the difference between paid and gross.

Some of that new disclosure -- I think it's a footnote in the supplement -- seems to indicate that one of the biggest differences is vehicle-on-non-vehicle collisions.

And I guess just trying to drill into that a little bit more, anything you can give us on either what percentage of your claims base tends to be vehicle-on-non-vehicle, how the average severity of those non-vehicle crashes kind of compares to the vehicle-on-vehicle, or just how to think about -- I realize a lot of those don't close with a claim.

But how do we think about the percentage of those that close with a claim versus the percentage that close with a claim on just a vehicle-on-vehicle?

Okay.

Thanks for the question, Ryan.

It's Matt.

So I just want to point out and restate what I said in answering Elyse's question.

So we made a decision during the second half of 2015 to try to capture additional information on every claim.

And as we did that, we did that for the purposes of really digging at a very granular level into what was happening in our claims pattern and make sure that we were ahead of the game in both an ability to staff it and an ability to predict to the extent you can emerging trends.

So as we began capturing additional information, we began capturing additional information on what we call hit-fixed-object or non-vehicle accident.

And the fixed object that you hit can vary depending upon the season.

In January and February, that fixed object could be a snow bank.

It's less likely to be a snow bank in July and August.

Sometimes, it's something that can result in a claim, like a mailbox or somebody's tree or a garage door or something like that.

And sometimes, it's something that would not result in a claim.

So we wanted to capture additional information.

We did, and as a result, we were probably capturing and continue to capture some incidents that when you look at it, you are less likely to think it's going to result in a claim.

But to be complete, we wanted to capture everything.

It allowed us to make a more informed assessment of liability earlier in the process.

As we disclosed in the supplement, as a result of that change, there's been a gross frequency for PD increase.

And we quantify that.

It's about 1.5 points in the third quarter of 2015, it's about 2 points in the fourth quarter, and it's about 3 points in the second quarter of 2016.

And so you'll see adjusted numbers in the supplement as well as the quote gross numbers' unadjusted numbers.

Most of those additional claims that we're capturing are closed without a payment, so they have no impact to the income statement.

And they are not impacting our paid, but we wanted to capture it to see trends.

And when you are as large as we are, every little change in what you capture and in claims opening practices can influence the gross number.

Ryan, this is Tom.

When Matt said this, I want to go at a level below that, which is in doing our net income in our reserving processes, we use actual claim counts and severity by type of loss.

As you know, some longer-tail businesses, where they don't have the data that we do on a real-time basis, they book to estimated loss ratios.

When Matt is talking about what we are doing with claim counts, that's all factored in.

So we have greater precision in the way we estimate our current income levels.

Okay.

That's very helpful.

And I guess just as a follow-up, we've heard some of the life companies this quarter talk about the impact of the pullback in interest rates, both on the income statement and also some charges.

I guess just with Allstate, maybe looking out over the next 12, I don't know, maybe even 24 months, how should we be thinking about the income statement of low interest rates?

And I guess also the impact that it could have in terms of GAAP charges or stat charges in the runoff life block?

Thanks.

Ryan, this is Steve.

Let me talk a little bit about our investment portfolio, which I think is what you are referring to.

I made a comment in the prepared remarks about the difference between the portfolio yield and the current yield you might be able to reinvest in.

And for property-liability, it's fairly close.

For life, there's a difference.

Obviously, we have longer-duration assets.

We are getting good yields on those today, but they slowly run off over time.

The good news is for us, unlike many other companies, is we've been paying out our runoff annuity liabilities with a lot of that cash flow that is coming from the portfolio.

So we haven't had to reinvest in the lower interest rate environment.

Today, that's slowing down a little bit, so we have some cash flow beyond that, which we are investing primarily in performance-based investments, as we said, which really is designed to match up to our liability profile, which is longer duration than many of our competitors in terms of -- because we're running it off.

So if you look at that, and performance-based investments we believe would earn substantially higher, albeit variable returns over a period of time than you would in fixed income.

So we feel our income for Allstate Financial should hold on an investment -- should hold reasonably well.

Probably be a slow downturn, as you can imagine, given some cash flow and some of the market environments and what we do in managing the portfolio on a routine basis.

In terms of charges, probably I think you are thinking of maybe a premium deficiency charge, where we aggregate all of our life and annuity businesses.

And we disclosed this in the footnotes in the financial statements.

Right now, we have sufficiency in that.

We continue to look at that.

And once again, they have, being a performance-based investing, will help us significantly in terms of increasing and holding the returns we have in those long-dated annuities.

I hope that helps and answers your question.

Yes.

And I guess how about on a stat basis, like an AAT or anything like that at year end?

On a stat basis, we do look at that year end, as you say.

So I don't have any really current comments on that.

But once again, it really is based upon where your investment income is coming out.

And moving more towards our performance-based investing should help us on that note.

And last year, we did take a charge on stat of $259.

Thanks so much for your answers, guys.

Josh Stirling, Sanford Bernstein.

Thanks for taking the call and congratulations on the quarter.

So I was hoping you guys can -- we're all sitting here trying to figure out how to model your earnings going out over the next couple of years and balancing, basically, the frequency trend and severity trends you are talking about as well as the very active pricing campaign you're managing.

And I was wondering if you could help us -- I'm not asking you for guidance, per se.

But help us understand -- remind us a bit, like at the business segment level, what the underlying combined ratio targets you guys are trying to get to are.

Because we can sort of do our own math about judging how long you will get there, but I think it's been awhile since we've had that conversation.

And as you've made a lot of changes in Encompass and Esurance in particular, I don't think that we actually really know for certain what we should be looking to as the long-term goals for those businesses.

Good morning, Josh.

Let me go way up for a minute and I'll come down.

So we expect all of our businesses to earn a current return on the business they write over our cost of capital.

So we start there.

And that's true with all of our businesses today.

But then what we do is we adjust it for what our strategies and the volatility of earnings are.

So when you look at the underlying combined ratio for homeowners, for example, it's lower than you would see for the other businesses because it utilizes more capital because it's more volatile.

We have a very sophisticated process of allocating economic capital, not just to those lines of businesses, but then Matt and his team push it down by state.

So we've always talked about the system we have.

That system includes taking economic capital for homeowners down to places like Mississippi on the coast is going to be different than Wisconsin in the woods.

So by homeowners will always have a lower underlying combined ratio and lower underlying loss ratio than you would see from the auto business.

The auto business, we don't give a target like some of our competitors do, but if you looked at sort of where we've operated, it's kind of been where Geico and Progressive is, which call it the mid-90s.

That's a place where our customers have been willing to give us a good return for providing the services we do.

That's kind of where we look to be at is in that zone.

But we don't have a specific number that we give out either by quarter or year.

If you look at the other property-liability businesses, they look more like auto insurance than they do homeowner insurance.

If you go over to Esurance, we look at it on a longer-term basis than just the underlying combined ratio because of the way the accounting works on getting a new customer.

So you spend all your money upfront getting new customers and it gets expensed right away, as opposed to in the Allstate channel, we are providing service all along.

So the commission we paid our customers provide that service that'll get amortized [long].

So you should expect to see a higher recorded combined ratio for Esurance than you would for Allstate because of that accounting.

But also because of its size and the growth potential we have in that segment.

So we've been -- as long as Don's team is creating economic value, we will continue to invest because we believe we are creating long-term shareholder value, even if it hurts current earnings.

So if you look at the underwriting loss, we've been running underwriting loss since we bought Esurance, but it's doubled its size.

And it's doubled its size and we believe we've created a lot of economic value.

Encompass looks more like the Allstate channel and customer segment than the other one.

So obviously we have similar conversations for the life business and Allstate Benefits.

Is that helpful?

Yes, that makes sense.

If I could just -- I'll use my one last follow-up on this sort of similar points.

If I try to keep it at 50,000 feet, it feels like your guys' auto businesses are running a couple of points below -- or above where you would like them to be.

If we assume that you can fix those over the next couple of years, that would be fantastic for margins.

One, is that sort of the right way to think about it, if you guys can get ahead with pricing and/or you get a tailwind because frequency trends stabilize or even reverse.

And then I guess the other thing that's sort of implicit in that assumption is that you will be able to maintain the margins you have currently been getting in homeowners.

And we haven't been talking about it quite as much just because auto has been a focus.

But if you're sort of thinking about this big picture and staying at that level, are those sort of the right way we should be thinking about this?

Yes, it is.

I would say if you look at where we operated the auto insurance business with flat severity.

As Matt pointed out, for a long period of time, we've had flat severity.

If you look at -- sorry, flat frequency -- thank you, Pat -- we were able to operate that business at a combined ratio below where it is today.

We see no change in our capabilities relative to our competitors.

No change in the overall competitive environment that tells us we can't be back in that space.

It will take us in time, given that those frequency numbers have been headed up and we price on a lag basis.

But we see no reason why we can't get there.

The homeowners business has been successfully repositioned.

And we're sort of zeroing in on around four years of the new environment, and we see no reason why we can't continue to maintain that profitability where it is today.

Great.

Thank you.

Good luck.

Amit Kumar, Macquarie.

Thanks and good morning.

Thanks for fitting me in.

Just two quick questions, if I may.

First of all, again, going to slide 7, could you perhaps give us more color as to what exactly are these type of claims that are being closed without payment?

And has that percentage increased?

Well, the percentage, it bounces around a lot.

For example, let's say somebody called us and they had a claim and their deductible is $1,000 and the damage was $700.

We close that claim without payment.

There's lots of reasons.

It could be they called us and the other person was at fault.

And so there's lots of things that happen where we don't have to actually pay money.

But we want to know that somebody has a claim so that we can respond proactively and do what we get paid for, which is help them at that time of trouble.

On a macro level, it fluctuates a lot, but the trend hasn't changed.

Got it.

That's actually quite helpful.

The other question which we were getting was on advertising, and I know you talked about this.

Going back, I thought the thought process was that the advertising piece will come back once the book is fixed and under control.

And I think some people were surprised on the tick-up in the advertising level.

Maybe just talk about that.

Did we misunderstand you at that point, or you have greater comfort where we are right now.

And hence, that's why it's ticking up?

I can't speak for what people thought we said, but what I can do is say we're comfortable doing more advertising.

But I would also point out advertise -- it is really marketing.

Because we have made a huge shift in our allocations away from TV to digital.

And so -- over the last couple of years.

So you might not see as many ads on TV, but that doesn't mean we are not out banging away trying to find new customers.

But we're comfortable that from both an ongoing standpoint we need to invest to make sure our brand is out there and relevant.

And secondly, that we're advertising for the types of customers that we have the ability to take in.

And then does that run rate sort of go up from here or remain stable?

You know, it varies by quarter.

So if you are trying to do a model, I wouldn't get too hung up on it.

I would think about expenses overall.

We think about expenses as a way to provide both an opportunity for us to grow and create service for our customers.

Right now, our expense levels are down.

If we had a lower loss ratio and we felt like we could expand, we would then continue to invest until we thought the average acquisition cost was too high.

Got it.

That's very helpful.

Thanks for the answers and good luck for the future.

Kai Pan, Morgan Stanley.

Thank you and good morning.

The first question is on the sort of inflection point on your core loss ratio in your auto book.

You've been starting raising price actively the second quarter of last year.

Given the short-tail nature of your business, I was wondering when do you think that will eventually earns through above your loss cost trend.

And also, your second quarter of this year, a big increase in term approved rate increases.

Do we have to wait until this chunk earn in to see that inflection point?

The answer to the first one is it all depends on what you think is going to happen to frequency.

And we can't predict the bottom part of that chart that Pat and Matt talked about.

We can't predict that.

So as Matt indicated, as long as it is going up, we will continue to adjust our pricing to reflect the cost that we have to cover for our customers.

When it will go through will be -- and you'll see it in the P&L -- is when it exceeds the growth in frequency and severity.

It didn't happen this quarter versus the second quarter of next year, but we don't price just to hold even.

So at some point, we would expect to get back to that loss ratio that we talked about earlier.

And I would add, Kai, I think you were asking about the timing.

It does, since they are 6-month policies, it does take 12 full months to fully earn in and it's earned in in the 13th month.

So you can't accelerate that; that's just a function of how the rates work in the system.

And so you can look at what our effective rate is.

And we talk about the effective rate increase and how much is likely to burn in as result when it's fully developed.

And some of that will actually burn in in 2017 of what we took in second-quarter 2016.

Okay.

That's great.

A second question is on the Arity.

Could you talk maybe more about your business model there and what it would over time what value it could provide to Allstate, both strategic as well as economic value to your franchise.

And how much investment you need to make in that unit?

Okay, Kai.

Let me answer it first with a two-component answer.

One is the overall environment and second is Allstate specifically.

So if you look at the personal transportation system, we think it offers one of the biggest economic opportunities facing America -- the restructuring of it.

Just a few statistics: there's 240 million cars in the United States that are worth about $4 trillion.

Direct cost to maintain and run those cars is about $2 trillion.

You can add another trillion of indirect costs.

And it?s very inefficient capacity utilization is about 4% in total.

It's about 33% at peak hours.

And then you see one person in the car most of the time rather than multiple people.

So everywhere I look, I see idle cars and if they are moving they only have one passenger.

At the same time, there's about 30,000 people killed in that system and there's billions of hours spent either sitting in traffic or waiting looking for a parking spot.

So we can do better.

And if it was your system, you would shut it down and rebuild it.

That will be done over time.

We believe it has the opportunity to increase personal household consumption by up to 5% on an annual basis.

So there's big money here.

The train's on the track, it's moving, and it is picking up speed.

So what is Allstate going to do about that?

Is first, we can use the telematics data to improve the accuracy of our pricing.

Today, most companies estimate the chances of people having an accident by their history and where they live.

With Drivewise, we can look contextually at everything you do about driving and price it with real driving behavior.

That gives us a more accurate price.

We are also testing pricing plans so that you can continue to pay us like we do today over six months, as Matt talk about, or you can pay us per mile if you want.

So there's lots of things that will happen in this personal transportation industry that we believe creates great opportunity for Allstate.

At the same time, we're broadening that value proposition to customers beyond just pricing so they get more than -- from Allstate than just advice on how to protect themselves and more than help in fixing their car when they get in an accident.

So today, for example, we give them safe driving tips as to how to do better driving.

We connect them with roadside so they be -- contextually we can get there faster.

We're also doing a number of things in terms of giving them rewards.

So we are both doing a better job on what we currently offer and we're expanding the offering we give them.

Now there's many companies that want to get in this space and we think we have a natural way in because it's direct and it's low friction in terms of giving customers direct benefit from the connection.

So it's our belief that we can use that connection then for multiple purposes, such as road usage, providing real-time safety advice, consumer benefits that are determined by a customer's interest and the time and place specific location they are at.

So as a result of that, we?ve created Arity as a way of bundling those services together on one platform.

So the strategy is strengthen our existing businesses and find new revenue sources by leveraging those connections with customers.

Actually, we don't disclose the amount of money.

We spent -- we don't get to 1 million customers without spending some money on it, but we don't break that number out.

Thank you very much.

Sarah DeWitt, JPMorgan.

Hi, good morning, and thanks for squeezing me in.

I would be interested to get your perspective on loss trends.

Why do you think they haven't stabilized yet, particularly given we've lapped on lower gas prices and the economy isn't booming?

Well, I think you answered part of the question there, which is people do three things in their cars: they go to work, they run errands, and they take trips.

It's about a third each of those, roughly speaking.

So as people -- more people are working, they drive to work more.

That creates more economic activity, which also makes the roads more crowded.

The trips tend -- the errands tend not move around that much.

And then you have trips.

And so lower gas prices help.

Matt, anything you want to --?

Yes, I would add, Sarah, in addition to the miles driven, which as Tom said, miles driven is really a factor of gas prices, employment rate, because that drives those behaviors.

The other component that is a little more difficult to quantify, but we know is out there, is the level of distracted driving that's taking place today.

And that distracted driving -- the use of cell phones while driving, the texting, Facebook surfing -- we've all watched it.

Some of us have been victims of it on the road.

We all watch people sliding around in the lanes.

And the combination of increased number of cars on the road due to the first thing -- due to the miles driven -- means you have a greater density of cars on the road with less margin of error.

So there's less space in between those cars.

When you add distracted driving to that and you have people swerving and not paying attention and going to hard stops, it's an increased likelihood that they are going to hit somebody else.

So it's a perfect storm to have greater number of cars on the road, driving faster -- and we have lots of evidence that speed has increased as well.

Our Drivewise data is showing us greater number of trips, greater length of those trips, and greater speed during those trips.

So it's a bad combination when you add to it people looking at their cell phone instead of being focused on their driving.

And then as we mentioned as well, the loss cost and the same trends Matt and Pat have talked about, which is cars are more expensive.

Knock off your mirror, it's $1,000, not $150 because it's got all those sensors and stuff on it.

So let me just close with we take a proactive approach to the current environment and making sure we invest for the long-term strategic growth opportunity.

So we are adapting to the higher costs associated with auto insurance, as most of you have asked us about.

But I want to be clear: it's comprehensive and a multifaceted approach.

It's not just about raising price or taking a more segmented approach to that.

We are managing risk-adjusted returns to create shareholder value.

That's like what we did in homeowners, where we got smaller to get better.

Or as we're doing today in payout annuities.

We are focused on growth, whether that's Allstate's Benefits, Allstate agencies and their relationships with the 16 million households, Esurance, or our telematics offering.

We are investing for long-term growth and we will continue to do so.

So we have a strategy, resources, and a team to continue to create their value for our shareholders.

So thank you again for spending time with us this quarter and we'll talk to you next quarter.

Thank you, ladies and gentlemen, for your participation in today's conference.

This does conclude the program.

You may now disconnect.