Good day, ladies and gentlemen, and welcome to the Allstate second quarter 2018 earnings conference call. At this time, all participants are in a listen only mode. Later, we will conduct a question and answer session, and instructions will follow at that time. If anyone should require assistance during the conference, please press star then zero on your touchtone telephone. As a reminder, this conference call is being recorded. I would now like to introduce your host for today's conference, Mr. John Griek, Head of Investor Relations. Sir, you may begin.
Well, thank you, Daniel, and good morning and welcome everyone to Allstate's second quarter 2018 earnings conference call. After prepared remarks, we will have a question and answer session. Here today are Tom Wilson, CEO, Steve Shebik, Vice Chair, Mario Rizzo, CFO, Glenn Shapiro, President of Allstate Personal Lines, Don Civgin, President of Service Businesses, John Dugenske, Chief Investment and Corporate Strategy Officer, Mary Jane Fortin, President of our Life, Retirement and Benefit Businesses, and Eric Ferren, Controller and Chief Accounting Officer. Yesterday, following the close of the market, we filed the 10-Q for the second quarter and posted the press release, investor supplement, and today's presentation on our website at allstateinvestors.com. As noted on the first slide of the presentation, our discussion today will contain forward-looking statements about Allstate's operations.
Allstate's results may differ materially from these statements, please refer to our 10-K for 2017 and other public documents for information on potential risks. This discussion will contain some non-GAAP measures for which there are reconciliations in the news release and investor supplement. Now I'll turn it over to Tom.
Well, good morning. Thank you for investing your time to keep up on our progress at Allstate. Allstate continued to deliver excellent operating results, and we're on pace to achieve our 2018 operating priorities. The strategy to deliver differentiated products to consumers is laid out on slide two, and is working. Starting with our property liability businesses on the left, we have four market-facing businesses, each with a different customer value proposition and go-to-market business model. The Allstate brand in the lower left competes in the local advice and branded segment, and growth accelerated in the second quarter. Esurance in the lower right focuses on self-serve customers who prefer a branded product, and its premium increased 12.5% over the prior year.
We leverage our pricing, analytical, and operating expertise across these businesses, which comprise 86% of revenues and 91% of adjusted net income, and create substantial value for our shareholders. Our focus on value creation also includes the four areas in the right panel. We're enhancing customer value propositions through increased connectivity. A good example of this is the use of telematics in the Allstate agency and Esurance businesses. We use connectivity to give customers a price for auto insurance that reflects their individual driving behavior and give them the opportunity to buy it by the mile. For those of you who are in New Jersey, you've likely seen our wraps on the ferries and in the advertising in the subway. We're also working to improve our customers' driving experience by maintaining this connection beyond what's needed to create an insurance risk score.
QuickFoto Claim enables us to pay auto insurance claims in hours versus days, and we believe we lead the industry in implementing this technology, which is highlighted in our advertising. We also create value through growth of a number of large significant businesses like Allstate Benefits, Allstate Life, and SquareTrade. The value of these businesses is substantial but can be overshadowed by the sheer size of the property and liability business. As Mario will discuss, we also proactively manage capital to create shareholder value. Turning to slide three, Allstate's growth increased, and returns were excellent this quarter. Revenues exceeded $10 billion with increases in average premiums and policies in force.
As you can see from the green box on the far right of the table, Allstate Protection policies in force grew 0.9% in the second quarter of 2018, which is due to growth in the Allstate and Esurance brands. Service business group policies in force nearly 36% due to SquareTrade. If you move back to the left, you'll see that net income was $637 million, or $1.80 per share in the second quarter of 2018, with adjusted net income of $675 million or $1.90 a share. In the middle at the bottom, net income return on equity was 17% and was 15.8% on adjusted net income. The property & liability underlying combined ratio for the first half of 2018 was better than our annual outlook of 86%-88%, primarily driven by lower auto accident frequency.
Given this positive result, the underlying combined ratio outlook improved to 85%-87% for the full year 2018. If you turn to slide four, we also made good progress on our five 2018 operating priorities. Those first three priorities, better serve customers, achieve target economic returns on capital, and grow the customer base, are intertwined to ensure profitable long-term growth. We made progress on better serving customers, as the Net Promoter Score improved across most of our businesses. Customer retention improved for both Allstate and the Esurance property & liability businesses, which is a key driver of growth. Returns were good in total, as we just discussed. The property & liability reported combined ratio of 94.9% generated $416 million in underwriting income for the quarter. Allstate Life and Allstate Benefits also generated attractive returns. Growth increased.
Allstate and Esurance brands grew policies in force from higher retention and increased new business. Allstate continued its long track record of growth with policies in force increasing 5.4% over the prior year quarter. SquareTrade added 13.2 million policies over the last 12 months. The fourth priority, to proactively manage investments, is imperative in a low interest rate but accelerating growth environment. The $83 billion investment portfolio generated $824 million in net investment income in the second quarter. Total return was 0.5% as the contribution from investment income was partially offset by lower fixed income valuations. We're also committed to building new long-term growth platforms. SquareTrade and Arity are two good examples of progress on this priority. John will now go through our property liability results in more detail.
Thanks, Tom. Slide five shows an overview of our property liability results. Net written premium growth of 6.4% in the second quarter was driven by accelerated growth in the Allstate and Esurance brands. The recorded combined ratio of 94.9 was 1.7 points better than the second quarter of 2017 due to increased premiums earned, lower catastrophe losses, higher favorable non-catastrophe prior year reserve re-estimates, and lower auto insurance accident frequency. This was partially offset by an increase in the expense ratio due to higher agent and employee-related compensation costs. The underlying combined ratio, which excludes catastrophes and prior year reserve re-estimates, was 85.5 for the second quarter of 2018 and 84.8 for the first six months of the year. As Tom mentioned, we are improving the guidance range by a point to 85 to 87 for the full year 2018.
This revised range takes into account more moderate frequency assumptions than originally planned, the seasonal nature of loss performance in the second half of the year, investments in growth initiatives, and the deployment of tax savings. Let's spend a few minutes discussing each brand in some detail. Slide six covers operating results for Allstate brand auto insurance. Starting with the bottom left chart, policies in force grew by 262,000, or 1.3%, in the second quarter of 2018. The renewal ratio of 88.5 was an improvement of 1.1 points from the prior year quarter, benefiting from our focus on the customer experience and a stable rate environment. New issued applications grew year over year for the sixth consecutive quarter, increasing 18% compared to the second quarter of 2017.
On the right-hand box, the recorded combined ratio for the second quarter was 93.0, 2.6 points better than the prior year quarter, and generated $358 million in underwriting income. The primary drivers of profitability improvement were increased average earned premium, higher favorable prior year reserve re-estimates, lower catastrophe losses, and a broad-based decline in accident frequency. The underlying combined ratio of 92.8 in the second quarter of 2018 included an underlying loss ratio of 66.8, an improvement of 0.7 points compared to the prior year quarter. This was more than offset by a higher expense ratio, driven by higher agent and employee-related compensation costs. Let's go to slide seven to cover Allstate brand homeowners insurance results. Starting in the bottom left, policies in force grew 0.8% compared to the prior year, as both the renewal ratio and new issued applications increased. The bottom right chart provides detail on profitability.
Homeowners insurance recorded combined ratio was 98.3 in the second quarter. Performance for this line is better evaluated over a 12-month period, given weather seasonality and variability. Allstate brand homeowners insurance generated $932 million of underwriting income, with a recorded combined ratio of 86.5 over the last 12 months. Slide eight provides financial highlights for Esurance and Encompass. Esurance had strong growth and improved underlying profitability in the quarter. Esurance net written premium grew 12.5% compared to the prior quarter, reflecting increased average premium in auto and homeowners insurance, and a 4.1% increase in total policies in force. This quarter's policy growth benefited from a 3.2% increase in auto insurance policies in force and the expansion into homeowners insurance.
The recorded combined ratio of 101.9 in the second quarter was 4.2 points below the prior year quarter, as shown on the upper right, due to the improvement in both the loss and expense ratios. The underlying combined ratio of 95.9 was 4.6 points better than the prior year quarter, as both auto and homeowners insurance results improved. Encompass continues to execute its profit improvement plan and generated underwriting income in the quarter. There was a 6.6% decline in net written premium over the last 12 months, and policies in force were 11.1% lower in 2018. Encompass' recorded combined ratio of 98.4 in the second quarter of 2018 was six points lower than the prior year quarter, as the improvement in the underlying loss ratio more than offset an expense ratio.
The underlying combined ratio of 95.5 for the second quarter was 2.1 points better than the prior year period, due to increased average premium and in auto insurance frequency. Now I'll turn it over to Mario.
Thanks, John. Let's go to slide nine, which provides detail on our service businesses. In the second quarter, revenue grew to $320 million, and policies in force reached 9.1 million. Adjusted net income was $1 million
In the quarter, a $9 million improvement over the prior year quarter due to improved loss experience at SquareTrade and Allstate Dealer Services. SquareTrade revenues of $122 million were $52 million above the prior year, with half the increase driven by organic revenue growth and the remainder due to the adoption of a new accounting standard. SquareTrade made progress on the three objectives supporting its acquisition as domestic policies in force increased, loss experience improved, and the international business expanded. Turning to slide 10, let's review our Allstate Life benefits and annuities results. Allstate Life generated attractive returns on capital with adjusted net income of $78 million in the second quarter, shown in the bottom left chart. This was due to a lower effective tax rate, higher premiums, and increased net investment income.
Benefits adjusted net income, shown in the top middle chart on the page, was $34 million in the second quarter. The $9 million increase from the prior year quarter was primarily driven by increased premiums, improved benefit ratio on selected products, and a lower effective tax rate. Allstate annuities on the far right had adjusted net income of $44 million in the quarter, which was $21 million lower than the prior year quarter, primarily due to lower performance-based investment results compared to a very strong prior year quarter. Returns remained low due to the relatively high regulatory capital requirements. Slide 11 highlights our investment results. We proactively manage our investment portfolio based on our long-term strategic risk profile, relevant market conditions, and corporate risk appetite. Net investment income was $824 million, 8.1% lower than the prior year quarter due to exceptional performance-based results in 2017.
The chart at the bottom left of the slide shows net investment income split between the market and performance-based portfolios. Market-based investment income, shown in blue, increased and reflects higher portfolio yields. Performance-based investment income, shown in gray, was the primary driver of the decrease over 2017. While performance-based income in the quarter was lower than the last four quarters, this reflected very strong returns last year due to favorable equity markets. The annualized yield on the performance-based portfolio was 9% this quarter in comparison to 16.8% last year. Total return in the quarter is shown on the lower right. The positive return in the quarter was supported by a stable contribution from income, shown in blue, but was dampened by lower fixed income valuations due to increased market yields, which is shown in gray.
Total return was breakeven for the first six months of 2018, as $1.6 billion of investment income was offset by a decline in the value of the portfolio, reflecting higher interest rates and wider credit spreads. Slide 12 provides an overview of our capital strength. Adjusted net income return on equity, shown in the bottom left chart, was 15.8% for the 12 months ended June 30th, 2018, an increase of 2.3 points compared to the prior year period. Book value per share increased to $59.16, or 9.9% since the second quarter of 2017, as higher earnings and a 4.1% reduction in shares outstanding offset the impact of dividends, share repurchases, and a decline in fixed income unrealized gains and losses. Our capital position is excellent, and we returned $722 million to common shareholders in the second quarter.
Since the end of 2012, we have reduced the number of shares outstanding by 132 million or 28%, as shown in the bottom right chart. We'll open up the line for questions.
Thank you. Ladies and gentlemen, if you have a question at this time, please press the star then the number one key on your touchtone telephone. If your question has been answered or you wish to remove yourself from the queue, please press the pound key. Again, that's star then one to ask a question. To prevent any background noise, we ask that you please place your line on mute once your question has been stated. In the interest of time, we ask that you please limit yourself to one question and one follow-up. Our first question comes from Sarah DeWitt with J.P. Morgan. Your line is now open.
Hi, good morning and congrats on a good quarter. I just wanted to get your thoughts on the competitive environment. We're hearing that some competitors are starting to cut price, I just wanted to get your thoughts. Do you still think you can accelerate growth and maintain underlying margins in that environment over the next few years?
Morning, Sarah. This is Tom. I'll make a comment, then Glenn might want to jump in here as well. First, I assume you're talking about just auto insurance as opposed to home insurance or any of the other things we do.
Yes. Sorry about that.
In the auto insurance business, if you go back to 2015, 2016, we were early on increasing prices. We saw that coming perhaps slightly earlier than some people. You saw a number of companies start to increase prices after that period of time. You see that moderating with many competitors now, but there are some, particularly one direct company, who's still taking some pretty big price increases. State Farm did recently reduce their price in a number of states. I think you have to, rather than look at it on a percentage basis, think about what the absolute price is because
If you come down 2% or 3%, but you're 10% higher than us, it doesn't really matter that much. It's all on what your absolute price is. We feel very comfortable with where our absolute price is today when we look at our close rates relative to the number of quotes we get. We feel good about where we're positioned in the industry, and don't think there's any big price war coming, if that's what's in your mind. Glenn, maybe you want to add something to that.
Yeah, thanks, Tom. Just adding on, we feel good about where we are competitively. We look at these rates on a state-by-state basis, and we monitor the filings of all of our competitors. What we see is, yes, there's some slowdown, Tom talked about one carrier that took some reductions as part of a change in their pricing. We still see rate coming through the system. If you look at the CPI, in the first quarter, it was really high on a year-over-year basis. It was like 9% in auto insurance. Year to date, we look at it most recently, it's still around 8%. A lot of that are the rates burning in that were taken in the last year, and we don't see as much filing activity right now. There's still some.
It's still a, I would call it, a positive rate environment, and we're playing well on the fact that we've taken those rates in a good pricing position to be stable for customers.
Okay, great. Thank you. Secondly, you talked a lot about frequency declining. To what extent do you think that persists, and is that factored in in your prices?
Well, Glenn will talk about what we see to date that's been realized, and then we'll come back to forecasting.
Yeah. Frequency, as Tom always says, there's a level of unpredictability to it. There are components that we can have both some facts on and some opinions on. I'll talk internal and external. Internally, we like the quality of our book. If you look back over the past couple of years when we took some rates, those rates tended to shift the quality of our book because some of the less tenured business walked away from us as opposed to the more tenured business, and higher-risk business walked away more quickly than lower-risk business. Our quality improved. You look at the business we're bringing on, and we like the growth that we're getting because about 70% of our growth is actually coming from improved retention, which is a nice balanced way to grow.
The other 30% that's coming from new business, we monitor the quality of that business, and we feel good about the quality coming in as we look at the risk profile, the percentage full coverage, the percentage bundled, all that gives us confidence that we have a good quality coming into the book that's favorably impacting frequency. From an external standpoint, there's more of a mix of backed-in hypothesis. You've got the miles driven trends we look at, and a number of other external trends. I would say automobile improvements that are driving less frequency are starting to burn into the overall car park, the car community, that there's more lane departure warning and blind spot warning. Some of those things have a favorable impact.
All that said, there is a level of unpredictability, as we've seen over the last few years, that things can change in sort of macroeconomic environment that move frequency.
Thank you. Our next question comes from Gary Ransom with Dowling & Partners. Your line is now open.
Yes, good morning. I was wondering if you could compare and contrast the Milewise versus Drivewise, some things are obvious. The pricing is different. The level of connectivity and the level of customer interaction, I wonder if there's any differences there. I'm also interested in your opinion of whether pay-by-the-mile insurance is likely to attract a large share of the auto insurance market over time.
Both important questions, Gary. First, we're using increased connectivity for two purposes with our customers. One, to give them a more specific price relative to their individual driving behavior. Two, to improve their driving experience. The first piece is what many people in telematics are doing, which is you can look at not just where somebody garages their car or how they used to drive based on the motor vehicle records and claim behavior, but you can look at actually where, when, and how they drive specifically that day. That gives you a lot of options, including Milewise. Glenn can talk about Milewise, what we're in 3 states now, and particularly why we went to New Jersey. That'll give you some sense for how ubiquitous Milewise might end up being as a payment and pricing setup.
The second thing we're doing is working hard on using that connection to improve the driving experience, to improve customer loyalty, and drive retention. Not everybody has taken this approach, but we maintain the connection, whether it's through your phone or your phone connected to an OBD port device, to give you information about what the warning light is on your car. John, a couple of months ago, came in, had a warning light on his car. By the time he got to his desk, it said, "Here's what's wrong with it. Here's how serious it is. Here's about what it'll cost, and here's three people you can connect to help you fix your car." You can also do things like finally resolve which of you or your significant others are really the better driver.
There's a lot of things we're trying to do to increase and take what is oftentimes a low engagement product into a high engagement product. Glenn, do you want to talk about Milewise and what you see there?
Yeah. Milewise, the early read, I would say, very positive. I think there's a lot still to be determined. It's a good question about what percentage of the market it'll take. I can tell you the experience in expanding into New Jersey a few months ago has been very favorably received, and I'll share an anecdote or two that tell you how it works for folks. People intuitively don't feel good about paying full freight if their car sits parked most of the time because they use public transportation, they're in a metropolitan area. It doesn't make sense to them to pay those type of prices for full auto insurance when they don't use their car very much.
We've had specific situations and really, even in the first week after we launched, I got some notes from agents that said, "I had a customer that was about to walk out the door with their business because they just felt like the rate was too much for the amount they used their car, and they wanted to get less coverage somewhere else. I was able to save the business by saying, 'Hey, this might make sense for you.'" We're seeing really nice uptake. The early signs are very positive. It is a niche within our market, so the percentage remains to be seen. I think it's an important niche and potentially a growing niche over time with people that don't use their car primarily and use other sources.
Thank you. Our next question comes from Greg Peters with Raymond James. Your line is now open.
Hi, good morning. This is Marcos calling in for Greg. My first question has to do with processes to control costs in the context of InsurTech. It seems like the next wave of operational efficiency for carriers following online quoting will be on the claims side. Do you guys agree with that? Can you perhaps talk about how the adoption of QuickFoto Claim is developing, and what are some of the challenges there?
Marcos, I'll give you a perspective on what we would call Integrated Digital Enterprise, which we've been at for a number of years now. Then Glenn can give you an update on QuickFoto Claim without getting into any specific numbers because it's competitive information on what we're doing. The combination of digital recognition technology, that's the ability of a computer to look at a picture and tell something from it, is developing quite rapidly. When you combine that with the ubiquitous amount of cameras out there, communication costs, and the ability to build new and more sophisticated algorithms which take that digital information and use it for decision making, it's going to dramatically change the business. Claims is clearly one of those places where it can change the business. We're using that in QuickFoto Claim.
It can be used in many other parts of the business. InsurTech goes well beyond just experiences. For example, we're just launching the ability to sell a homeowners policy by verifying three pieces of information rather than have our agency owners and licensed sales professionals be in the data collection business and ask you a whole bunch of questions of which maybe you know, maybe you don't know, but we certainly can buy the information digitally. You're correct in saying that. We have this effort, it's called Integrated Digital Enterprise, which is to do that throughout our entire company. Glenn, do you want to give them an update on QuickFoto Claims?
Yeah. Very happy with what's gone on in the adoption of QuickFoto Claims, it's one of those change management exercises that at first, it feels very big. It's a big shift for people. A year and a half or so into a bigger launch of it's just the way we do business right now, customers really like it, our agents like it, the employees like working through it. I'll break down the parts of work. If you just think about segments of work. Historically, about half of the units of work were getting the human to the car or the car to the human, that was a customer effort. That was our effort for employees. In some way, we had to get the person that's going to write an estimate in front of that vehicle physically.
The other half of the work was actually writing the estimate. What Tom talked about is right on with photographs potentially doing that work through artificial intelligence on that piece. The QuickFoto was essentially eliminating that first half of the work. We didn't have to get the car to the human or the human to the car. The photos flowed through, we're able to, as Tom said in the opening, in hours, settle claims that would take days previously. Very positive adoption, very positive customer experience.
Thank you. I appreciate that color. My follow-up would be on SquareTrade. There was a nice uptick in PIF in the quarter, but revenues came flat versus Q1, following several sequential quarters of improvement. I guess just some color there?
Marcos, it's Don. First, SquareTrade is off to a really great start with Allstate. When we did the acquisition, we set three goals. The first was to increase our business with domestic retailers, both with additional retailers as well as the business we do with the existing ones. Second was to improve the profitability, you can see the improvement this quarter, partially as a result of us taking the underwriting risk, but also the loss and the expenses have been trending very nicely as well. Third was to build
An additional platform for them for future growth. Europe has done quite well for them. This quarter, we're seeing roughly double the volume. It's still fairly small, but double the volume we had last year. Specifically to your question, you have to look a little bit beyond quarter to quarter because it can be a little bit lumpy. You've got some seasonality in there. It depends on when you pick up additional retailers and when they begin to ramp up their program. I think the better trend to look at is year-over-year. Now, the numbers look a little exaggerated because of the revenue recognition, but roughly half of the increase is organic. Part of that is picking up additional retailers, but part of it is just doing a better job for the retailers we have and their customers.
Very happy with their progress thus far.
Thank you. Our next question comes from Jay Gelb with Barclays. Your line is now open.
Thank you. My first question is on the strong growth in Allstate brand. That continues to accelerate each quarter. Just trying to get a perspective on if you would expect that trend to continue in the back half.
Jay, this is Tom. As you know, we don't give guidance on either the top or the bottom line. When you look at the growth that it's the first time where it's been year-over-year up. If you look over the last three quarters, it's been headed in that direction. We would expect to continue to head in that direction. A key point I would note is that a significant portion of that growth is driven by retention. That's the gift that keeps on giving.
Right. Okay. Can you also discuss the strong growth in new applications in the Allstate brand? What you view as the drivers there?
Yeah, this is Glenn. I'll take that, Jay. I'd really look across the whole system at what's gone on to drive the growth. We've talked about going back over the past few years, there's been a plan, and Matt had talked about the plans of how we're going to grow the distribution system and make it more effective. We've increased our footprint. We're up, if you look year-over-year, about 1,600 points of presence between agencies and licensed sales professionals. We have them more engaged because they're winning in the market right now. They're marketing locally, and they're drawing those quotes in. They're hiring more staff. We've improved in meeting our customer needs because we have a higher percentage of full coverage and a higher percentage of bundling going into.
As Tom Wilson said, retention is up, driving a lot of the growth, and the more quoting activity is really those agents hunting within their markets and being effective marketers in their communities.
Thank you. Our next question comes from Michael Zaremski with . Your line is now open.
Hey, thanks. I had a follow-up on auto frequencies. I appreciate you guys used the terms, there's a level of unpredictability, and there's hypothesis surrounding what drives them. I was curious if The Allstate Corporation has a view on whether the level of unpredictability or the volatility of frequency has changed on a go-forward basis versus what it's been historically. Basically, do you think there's consumer factors out there that should cause us to think it might be more or less volatile over time?
Michael Zaremski, this is Tom Wilson. Maybe I'll take it, and Glenn Shapiro, if there's anything you want to add at the end, jump in. First, I understand the benefit of being able to determine what the trend is on frequency, and I know people have poked around this with our competitors as well. You can actually make a sector prediction, or you can assess the individual company performance. That said, frequency falls into the known unknown category as it relates to the future, whether that's its absolute level or the volatility in it. We think the decision on frequency really ought to be based on the business model. That is, how good are the processes and people.
What, of course, we do know is there's been a long-term decline for a lot of good reasons, whether that's drunk driving laws and the cultural unacceptability of drunk driving, anti-lock brakes, more cars per household. There's a variety of things. On the short term, of course, there's a number of drivers which make it more unknown there. It's the number of miles driven, which of course, impacted by economic activity, gas prices, weather at the particular time of day. If you have ice at 4:00 P.M., it's different than if you have ice at 2:00 A.M. We do know from our Drivewise customers that the number of miles has declined, the number of miles driven over the last couple of years, really starting in the fourth quarter of 2016. How that will be in the future, it's hard to tell.
You have seasonality, of course. Seasonality, frequency is typically a lot higher in the second half of the year than the first half, which if people want to talk about our guidance, we factor that in, that we do think frequency is typically up the second half versus the first half of a given year. One way I tend to think about it's baseball season, so I'll use a baseball analogy. You have a pitcher and you have a hitter. As a coach, you kind of know in general what the pitcher's good at, whether they got a fastball or a curveball or something like that. You don't know how good they're going to be that day, nor do you know what every individual pitch is going to be. To win, you just need the best hitter.
Someone who can pick up the pitch early, as soon as it releases from the person's hand, or react before it gets to the plate. In auto insurance, Allstate's a good hitter, right? In 2015, the frequency went up. We picked it up early. We made the choice. We saw the same pitch in 2016, and we continued to see it. Others didn't see it. They then caught up. Frequency's down in 2017 and 2018. As a hitter, our reaction is not to slack up and lower our prices to just get a bit more margin. What we do is we have a system that picks it up by state, by risk class.
That's really, I think, the best way to try to get a sense for how do you invest based on frequency, is think about the process, the people, the technology, and can they hit.
Okay, that's helpful. Thanks, Tom. I'm going to switch gears quickly to homeowners. I'm probably splitting hairs here, but if I look at the trailing 12-month underlying loss ratio in Allstate brand home, it's up a couple points over the last couple of years. I'm just curious, if the current levels of profitability, are you happy with those levels, or is there something that you're monitoring and might take more action to improve the underlying?
Well, you're living our life, which is we live our life splitting hairs, I'll let Glenn take that one.
Right. I would like to start with sort of a longer term view, because the second quarter marked the 24th consecutive quarter under 100 combined ratio in home. In a volatile line of business like homeowners, to have six years running of quarterly profit is remarkable. We do take a long view on it. Just like in auto, we're looking state by state at our rates. We look at our loss results. If you look over the last five years, even at the underlying, we always say we want to be in the low 60s. We're at the upper end the last couple quarters of low 60s, at 63 and change. If you look at two, three, four years ago, we're similar to those time frames, and a lot of that is driven by weather.
We talk a lot about catastrophes, but the non-cat weather, it drives our peril mix, which drives severity. It's just so much more volatility in the homeowners line than you see in auto between the frequencies and severities driven by weather patterns. The long-term view we feel very good about. We're monitoring from a rate standpoint, I would not say there are alarm bells based on the hair splitting or anything that you brought up there.
Why don't you talk a little bit about what you did see in peril mix and stuff in the second quarter?
Yeah. Definitely, whereas in the first quarter, we talked about a lot of water and fire. Continued with the water, definitely wind driving some of the peril mix, moving with even intra-quarter of decline on the fire. Like some fire in the quarter, but moving throughout the quarter downward.
Fire loss is a lot higher. Like once the fire gets into you, your severity's a lot more higher than if somebody runs in your garage door. We had elevated fire losses in the second quarter, the same as wind and hail, so that bumped it up. That said, we watch it, right? We have some processes built in by state to make sure if we think we need to increase price because of either frequency or severity, we do that.
Thank you. Our next question comes from Christopher Campbell with KBW. Your line is now open.
Yes. Hi, good morning.
Morning, Chris. We can't hear the question. Maybe if you'll repeat it.
Yes. Can you hear me now?
Yes.
Okay, great. I guess the first question is just on capital management. Allstate's been a strong repurchaser of shares and bought a bunch back this quarter. Is there a certain hurdle above which that you're thinking of where repurchases wouldn't be economical? How would your capital deployment strategy change if the shares got too expensive?
I'll let Mario talk about the overall capital plan, I would say at a business that's growing with a 17% ROE with its current multiples, we think it's cheap.
Yes, I would agree. Hi, Chris, this is Mario. I guess where I'd start is when we look at our capital position, it continues to be excellent, we view it as a source of both strength and flexibility for us. When I say that, I mean the absolute level of capital we have, our capital structure, and our capacity to source capital efficiently if we need to. When you look at those things combined with, to Tom's point, the fact that the business is performing really well, we expect to continue to generate meaningful amounts of capital going forward. We're long capital, and we feel really good about our capital position. In terms of the buyback program, we still have $376 million left on the existing authorization, we'll continue to buy back stock.
When we run out, we'll follow the same process we always have. We'll talk to our board, we'll look at different alternatives, and we'll make a call from that point.
Okay. Thanks. That's very helpful. Then just kind of one other question on SquareTrade. What's driving the year-over-year loss ratio improvement?
It's actually a variety of things. First, as they build up their scale, they're getting more effective. A combination of things like their claims handling, their ability to reduce escalations and so forth, are all leading to better experience for the customer and lower loss cost for them.
Thank you. Our next question comes from Joshua Shanker with Deutsche Bank. Your line is now open.
Yeah. I want to thank you hosts for taking my question.
Yeah.
The homeowners growth and auto growth seems to be going in line with each other. I assume that that's mostly bundled sales. I'm interested in knowing how the homeowners-only product is selling and what the marketplace looks like for growing in homeowners only.
Well, you're correct, Josh, in that you remember when auto was going down, the homeowners business got slightly smaller as well, but it lagged it, because it's an annual policy, not everybody has it. To the extent we were losing auto customers because we were raising prices, that impacted it. You see the other side that Glenn can talk about, how we bundle what we see in the homeowners market.
Yeah. We think it's a great opportunity to bundle. We've aligned our incentives for agents around bundling, and we think it's the right thing for customers. What we're seeing in this cycle right now is that, as Tom said, it tends to lag the auto in terms of whether it's going up or coming down. It seems to be pulling in a little more and, as you said, moving in tandem. We're growing at pretty close to equivalent rates in real time. That is a result of more bundling of the product. We think we are very competitive in home right now. We have a lot of markets where we have a strong competitive position from a pricing standpoint. Our product is part of the reason for that.
We have, with the House & Home, a favorable product, our roof rating schedule allows us to do that. I think good catastrophe management and spread of risk over a long period of time has positioned us well for that. We're in a really good spot to grow home.
I realize that having a home product increases the persistency of selling an auto, too. Is a home-only product terribly less persistent than an auto-home bundle? Ultimately, is the long-term value of a home-only customer high and worth pursuing?
Josh, let me take that. Before we do that, let Steve talk about homeowners and Esurance, and then I'll come back to your question on the lifetime value of a customer, either by policy or in total.
As you know, in Esurance, we also have a homeowners product. We rolled it out over the last three or four years. We have it in 31 states, I believe. Our PIF overall increase in Esurance is 4.1%, while it's about 27% quarter-over-quarter from prior year in homeowners. That's a really important sale for them because we spend money on marketing and advertising. As we can get more premium, so both auto and home across the we have, it significantly makes the acquisition cost better for us in terms of the customer acquisition. We've been growing that for a handful of years now. We feel good at where we are. Obviously, you see we still have a really good geographic spread in the country, so we do have catastrophe quarters where we have some pretty good catastrophes. This is one of them.
We think we're doing well and growing well there, and it'll be a good profitable addition to the Esurance business.
On the what does it do for lifetime value both on the product and so a couple of key themes there. We want to make sure we make money on every product we sell to everybody, so we don't lose money on one product and make it up on another. We have standards around how we underwrite and price our products so we make money on every and each product. That said, you do know that, and we know, actually, and you're obviously aware of as well, that the more things people buy from you, the longer they stay as a group. Now, some of that's just they buy more from you because they like you more. They like you more, so they stay more. There's a little bit you got to be careful on to what you read into it.
That said, the more opportunities you have to build connections with customers where they have ongoing relationships with you, all things being equal, they should stay longer and it's good for you. You lower your acquisition cost per customer, you improve their customer value proposition. They don't have to shop around to a whole bunch of people, which is why, particularly in the lower left, we have such a broad-based set of products and services. Whether that's auto insurance, home insurance, boat insurance, personal umbrella policies, life insurance, we sell them everything we can, and we have metrics and measures around trying to do that because we know it best meets their needs. If we do that, they're more likely to stay.
As Steve talked about, we're starting to expand that in the Esurance business as well, which particularly because of the high upfront acquisition cost, it helps you lower your acquisition cost dramatically and gives you a competitive advantage versus those people who only sell one thing. I sometimes say, sometimes we end up only talking about auto insurance. It's a big product for us. We need to compete aggressively in it, we also want to win on a broad basis with our customers.
Thank you. Our next question comes from Michael Phillips with Morgan Stanley. Your line is now open.
Thank you. Good morning, everybody. Tom, let's just take a step back and ask, I think it might be a bit of a goofy question. I'm going to throw it out there, hear your thoughts. As you look at the agency plan field and the folks that run all your exclusive agencies, how has the average age changed over time? Has it shifted any? Then has it been any more difficult to find replacements or recruits to replace what may be an aging population out there over the past, say, three to five years?
I'll answer your question specifically, then I think where you're headed a little bit, Mike, is what's the long-term view of the agency platform? Particularly when you look at some of the independent agency industry metrics, it's as you know they're all getting older, they're staying in their business, but they may not be there forever and who's going to take them over? Will that go away? We don't have that problem in the Allstate agency channel. We're out recruiting people all the time, whether that be our agency owners or licensed sales professionals. We don't have a, what I'll call, a demographic barrier to our future growth or even maintaining our current position.
That said, I think the business is going to change dramatically, that technology is going to enable us to have those people in the field who know our customers do a much better effective job for them by doing less data collection. If you go back to Glenn's view on Integrated Digital Enterprise and the work effort that went into claims and how much was spent driving around, if instead of driving around, you slipped in their data collection, how much time is spent by agency systems in data collection? We're hard at work at trying to make that a much smaller piece of what they do.
Maybe if it's data verification, even then you don't have to do a tremendous amount of it, change our products and processes and our technology to do that, use our technology to make the experience even better, then enhanced by a person. I do think there's going to be a lot of change driven in that channel. We're pushing hard on doing it, so we're not just sort of saying, "Oh, geez, can we still hire the same amount of people?" Our challenge will be to transition that system effectively and efficiently on behalf of our customers. We're hard at work on that. As we've talked about, Matt talked about, Glenn's picked up the trusted advisor work. We have a whole other series of work related to technology and product that we've embarked on, that we think will further accelerate that.
Great. Thank you very much for that detailed answer. Quick numbers question, if I could. The 135 development, I assume it's bodily injury and property, and I'm assuming that's recent accident year 2017, maybe 2016. I guess just to confirm that. Then I noticed you took out the severity numbers of your press release, unless I missed that. Is that true for BI liability?
Right. There's two parts there. I'll let Mario handle the reserve changes, the net he'll talk. You're correct, it is bodily injury, but it wouldn't be as the most recent years you're talking about. Mario can give you an update there. John can talk about what we've done on severity, because as we talked about last time, we want to be fully transparent with everybody. Transparency also includes giving you information that means something. We changed the bodily severity stuff. Mario, do you want to start with reserves and then John?
Yeah, sure. I guess the place I'd start on the kind of from a broad perspective on reserves is, as we've mentioned in the past, we have really solid processes in place to both establish reserves and continually reassess the level of reserves that we have really across the entire business. When we look at our roughly $26.6 billion gross reserve balance, we feel really good that it's set at the appropriate level. Having said that, as circumstances change both in terms of our claim processes and the external environment, we look at that reserve balance every quarter. Like we've done over the last number of quarters, to the extent there are releases, then we recognize those in the income statement. I think that's as much a function as our conservative approach to reserving as anything else.
I certainly wouldn't use that as a predictor of reserve releases going forward. It's just something we continue to look at and if reserves are redundant, we release them.
When you look back at, it wouldn't have been the 2016 and 2017 report years on bodily injury. It was before that. You would look back if you were to adjust the calendar year combined ratio numbers we gave you, which you'll be able to do when you get the information. You can see it was really prior years to then.
Hey, Mike, it's John. Just specific to your question on the BI severity disclosures. We're very transparent overall in our financial reporting. We're clearly trying to aim to provide you with the best metrics to evaluate the financial performance of our business. Last quarter, we did signal that we were going to be removing the BI paid severity statistic. Really on a long-tail coverage like BI, using a calendar year paid severity measure, it's much more an operational statistic than something that gives you a clear line of sight to the P&L. We do factor in overall severity trends in our guidance each year, so it's included in our underlying combined ratio guidance. This quarter, you can find some color in the 10-Q in terms of what our overall severity trends look like.
Thank you. Our next question comes from Bob Glasspiegel with Janney. Your line is now open.
Good morning. I'll give you a chance to throw some cold water on my enthusiasm, Tom. Your service business broke into the black, I think, for the first quarter, your life earnings were up a decent bit sequentially, I think year-over-year excluding partnerships. Is it onward and upward for both units, or were there some one-time items helping out?
Well, Bob, as you know, first, cold water would be a bad idea, right? Guidance would be equally a bad idea when it comes down to individual components of the P&L. I won't give you specifics. I would just say both of them. We feel really good about our life insurance business. It's earning good returns. It's well run. We're bringing our costs down. We just launched a new term product, which we think will really work well with our customer base, who tend to want something simpler. This is a monthly term, just pays you by the month, what you want for your family for three years, five years, as opposed to some big number, which scares people, some of our customer base. We feel good about where that business is positioned, both operationally and strategically. We feel the same about the service businesses.
They're of different characters, though. It's a broad enough portfolio. Don's got some that are doing really well and others that we'd like to do better, because we have high aspirations. You'll see this quarter, the Allstate Dealer Services, which has got almost 4 million policies out there, its profit turned up. It used to make a lot more than it made this quarter, so we'd like to see it make more. It made more than it did last quarter, or its losses, I think, last year. Overall, though, we feel good about all those businesses. The one business you didn't mention, which is our annuity business, where the results are down this quarter because the performance-based investments, the returns were down. That said, even though performance-based investments were down, the yield on an annualized basis was still 9%, so we're feeling good about that.
John can talk about it if you have an interest, but we feel good about where it is in total. Overall, that annuity business continues to earn low returns on capital because you have to put so much cap up. We've talked about that in the past. We're trying to come up with ways to put up less capital so we can raise the return as well as by investing differently.
That's helpful. Just on Sarah's question, opening question, closing the loop. One of your large competitors yesterday on the issue of declining frequency observed that the industry is, because of the volatility of frequency, competitors tend not to factor in down frequency in future pricing actions. I was just wondering, you lowered your guidance for the range because of lower frequency. Are you factoring in negative frequency, I suspect not, into your thought process for prospective pricing actions?
Well, everybody prices differently, Bob, so it's hard to tell what people do. What I would tell you is that we look at total loss cost, which includes everything from underwriting expenses to loss cost, which, of course, are a combination of frequency and severity. If you were to look at a graph of percentage change year-over-year of frequency and severity, you would see that they have what we'll call the alligator chops because it's gotten wide. Severities continue to go up at sort of an inflationary level in the 4%+, depending which numbers you're looking at, whereas frequency has gone down. Pricing on that kind of basis and assuming the lower frequency is there usually means you're going to get caught in the jaws because at some point, the frequency could come back and the severity is not going away.
We tend to look at it on a long-term basis. Everybody does it their own way. If frequency is down and stays down, will we take less and fewer price increases in the future? Sure. You see that this year. From a shareholder value creation standpoint, it still earns really high returns, even at our old guidance, which we tried to let people know in the first quarter. It earns really, really high returns at this level. We're comfortable with the way we've positioned stuff, and I can't really speak to what the other competitor, listen to their call as well, and other people think as well. Let's take one more question, if we do that.
Thank you. Our last question comes from Brian Meredith with UBS. Your line is now open.
Yeah, thanks. Bob asked the question I was looking for. Tom, just to kind of clarify things here, your pricing decisions right now that you're kind of looking at for your auto insurance book, you're kind of pricing in line with what you think loss trend is going to look like here going forward. You're not purposely providing below loss trend because you're looking for some market share gains?
No. Our growth, I want to make sure that we underline this, more than two-thirds of our growth in policies in force came from retention.
Okay.
We didn't really get into new business discount and all that kind of stuff, and loss ratio on that. We're very comfortable with our growth and our profitability at this point because of how and why we're getting our growth, whether that's the retention in which customers we're retaining, which states we're retaining them in, or where our new business has come from, whether that be risk class, state-based stuff. We're comfortable that we're in a good position. We're not attempting to lower profitability to grow. That's not been our strategy. Thank you, all. Let me close. I know it's a little last there. We had a strong first half. We look forward to catching up with you next quarter.
We're going to stay focused on our operating priorities to make sure we balance both the long and short-term initiatives so we can create more value for our shareholders, for our customers, and all our other stakeholders. Thank you very much.
Ladies and gentlemen, thank you for participating in today's conference. This does conclude today's program, and you may all disconnect. Everyone, have a wonderful day.