Good day, ladies and gentlemen, and welcome to the Allstate fourth quarter 2018 earnings conference call. At this time, all participants are in listen only mode. Later, we will conduct a question and answer session and instructions will be given at that time. If anyone should require assistance during the program, please press star and then zero on your touchtone telephone. As a reminder, today's program is being recorded. Now I'd like to introduce your host for today's program, Mr. John Griek, Head of Investor Relations. Please go ahead, sir.
Well, thank you, Jonathan. Good morning and welcome everyone to Allstate's fourth quarter 2018 earnings conference call. After prepared remarks, we will have a question and answer session. Yesterday, following the close of the market, we issued our news release investor supplement and posted today's presentation on our website at allstateinvestors.com. Our management team is here to provide perspective on these results. As noted on the first slide of the presentation, our discussion will contain non-GAAP measures for which there are reconciliations in the news release and investor supplement and 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. We are expanding the quarterly earnings call to not only review quarterly operating results, but to highlight other value creation initiatives.
Today, we will do a deeper dive on our success in using telematics in auto insurance. Now I'll turn it over to Tom.
Well, good morning. Thank you for joining us to stay current on Allstate. Let's begin on slide two. Allstate continued to deliver strong operating results while building the future. We achieved all of our five 2018 operating priorities, we generated excellent returns. The property & liability underlying combined ratio of 85.8 for 2018 was better than the range we established with you at the beginning of the year. For 2019, the property & liability business is expected to have an annual underlying combined ratio between 86 and 88. If you move to the table at the bottom, revenues excluding realized capital gains and losses were $10.4 billion for the quarter and $40.7 billion for the year, driven by increased insurance premiums in the Allstate and Esurance brands. Adjusted net income was $430 million, which was $1.24 per diluted share in the fourth quarter.
For the full year, adjusted net income of $2.9 billion was 15.6% higher than 2017, which reflected strong insurance margins. Net income return on equity was 10.5%, and adjusted net income return on equity was 14.8%. If you could turn to slide three, we delivered on all five 2018 operating priorities. The first three, better serve our customers, achieve target economic returns on capital, and grow the customer base, are all intertwined. What it does is ensure we have profitable long-term growth. Customers were better served as the Net Promoter Score improved across all of our major businesses. Higher customer retention across the three underwriting brands was a key driver of growth last year. Returns remained excellent. Property & liability policies increased by 784,000 in 2018, which was a 2.3% increase for the Allstate brand and a 10% growth at Esurance.
When you combine that with the significant growth at SquareTrade, policies in force surpassed 113 million in 2018. The $81 billion investment portfolio generated $3.2 billion in net investment income in 2018, which was a result of slightly higher yields on the market-based portfolio and good performance-based results, which was compared to a very strong 2017. The total portfolio return was 0.8% in 2018, reflecting the stable contribution from net investment income that was offset by lower fixed income and equity values, particularly at the end of the year. We also made progress in building long-term growth platforms in 2018. I'll discuss telematics next, and Mario will discuss SquareTrade's performance. We also accelerated our expansion into personal identity protection with the acquisition of InfoArmor in October. These operating priorities will remain unchanged for 2019. Before we discuss telematics, let me set the context within our overall strategy.
Allstate's strategy is to grow by protecting people from life's uncertainties. We start with the upper oval, the personal Property & liability market has four consumer segments and provides protection by ensuring automobiles, homes, boats, and personal liability. We use differentiated products, sophisticated analytics, telematics, and are building an integrated digital enterprise to grow market share in this protection space. Our strategy also protects people from a range of other uncertainties, which are shown in the bottom oval, and we leverage our brands, our customer base, investment expertise, distribution, and capital. It began in 1957 with life insurance. In 1999, we acquired Allstate Benefits, which provides protection products such as life and disability insurance to employees at the work site.
That business is now four times its size from when we bought it, and is with 4.3 million policies in force, an adjusted net income of $119 million in 2018. We purchased SquareTrade in 2017, began offering insurance to Transportation Network Company last year as well, and recently closed the acquisition of InfoArmor. This strategy creates shareholder value through customer satisfaction, unit growth, attractive returns on capital. If you turn to slide five, this quarter, as John said, we want to highlight the value created from the use of telematics in auto insurance. We've been investing in telematics for almost a decade, to increase auto insurance pricing sophistication, to improve the customer value proposition, and leverage our capabilities and data to create a new source of growth and profit.
Let's start with what we do now. We began to use telematics in auto insurance in 2010 and now have a suite of products in the market. Drivewise and DriveSense are telematics-based offerings from Allstate and Esurance, and represent the bulk of our proprietary connection. These products either use a customer's mobile phone or an OBD port device, which goes up underneath the dashboard to establish a connection with the car. We launched Milewise in 2016 in two states, expanded to four more states last year, which allows customers to pay for insurance by the mile. Streetwise is offered through our online insurance aggregator, Answer Financial, in conjunction with Arity, to enable other insurance companies to benefit from telematics-based insights. Arity is a telematics service provider to Allstate, and is separate from the auto insurance company.
Moving to slide six, auto insurance pricing will eventually be significantly influenced by telematics information because it's just better than existing approaches. From a pricing standpoint, if you look at the top of that chart, auto insurance policies today are priced by who you are, such as age or gender, and where you live, which is a proxy for where you drive. For example, if your car is registered in Montana, there's a low likelihood you'll be commuting from New Jersey to New York. Telematics, though, enables pricing to be based on how you actually drive, and telematics is also based on exactly where, when, and how much you drive. That leads to increased pricing accuracy, lower subsidization between risks, and creates a highly personalized risk-based price. Telematics will be required to effectively price auto insurance.
Allstate is also using telematics to improve the customer experience by staying connected with customers. We provide customers with rewards for safe driving, safe driving tips that can lower their premiums, decode the maintenance light needed in your car. When you have that light comes on that says maintenance needed, if you have one of our OBD port devices in your car, we can tell you specifically what's wrong, how serious it is, what is your cost to repair, and enable you to link to a repair facility. Given these benefits, we believe telematics will be integrated into auto insurers' business models in the future. As a result, we created Arity outside of the insurance companies to create more value for shareholders. If we turn to slide seven. In 2015, we defined a strategic platform to help us design Allstate's business model.
You can see that in our 2015 annual report, we lay that out. A strategic platform is a system of capabilities, assets, information, and shared intelligence. These platforms are tended to be broad and flexible and create multiple uses for a wide range of customers and partners. In our definition, we would consider Apple, Facebook, and Amazon's marketplace to be examples of platform businesses. Companies that control strategic platforms generate high economic returns. These returns reflect the benefits of reduced friction and cost between participants and the ability to improve returns through increased knowledge and analytics. Platforms are also rapidly scalable. The transportation system can benefit from such a platform. A telematics platform enables companies to increase their speed to market in a connected car world. If you want to price auto insurance with telematics, you need data, which is enabled by a platform.
As more companies in the industry use Arity, the breadth and depth of data and shared intelligence will grow. More data on the platform allows companies to refine and customize their specific business models to their specific needs. For example, ride-sharing companies can use the Arity platform to enable them to select safer drivers or better manage their operations. It also lowers the cost of collecting information. Just like with credit scoring data, it's inefficient to have many companies collecting the same information. We decided to build Arity as a telematics platform to capture a portion of these additional economic benefits. It is little downside to us, since we need to build these services for ourselves because we're so far ahead of most of the industry. To date, Arity has 12.5 million active connections, of which more than 1.5 million are through the Allstate entities.
They analyze over 300 trips per second and create a proprietary driving score that can be used by insurers or shared mobility companies. Arity's scale continues to grow, and it's now adding 10 billion miles of driving data per month. Arity generates substantial advantages for Allstate's insurance operations today, and we're actively working with other insurers to help them utilize telematics and auto insurance. We'll keep their information confidential, but all parties benefit from the network effect of a consistent and large data set. As Arity grows, it will also provide us with new sources of revenue from the transformation of the personal transportation system. Let me now turn it over to Mario, who will discuss our fourth quarter and annual results in more detail.
Thanks, Tom. On slide eight, you can see that property liability results remain strong. Net written premium increased 6.8% in the fourth quarter and 6% for the full year, driven by accelerated growth in the Allstate and Esurance brands. Allstate brand auto insurance net written premium grew 5.7% in 2018, reflecting a 2.7% increase in policies in force and higher average premium. Esurance brand net written premium grew 12.7% in 2018, and policies in force increased 10.4%. The recorded combined ratio of 97 was six points higher than the prior year quarter due to higher homeowners insurance combined ratio, primarily from catastrophe losses related to Hurricane Michael and the California wildfires. The auto insurance combined ratio remained solid as higher premiums earned and lower accident frequency more than offset higher physical damage severity.
The underlying combined ratio, which excludes catastrophes and prior year reserve re-estimates, was 86.8 for the fourth quarter of 2018. Excuse me. This was 1.1 points higher than the prior year quarter, primarily due to non-catastrophe weather-related losses in Allstate and Encompass brand homeowners insurance. The underlying combined ratio of 85.8 for the year was within the revised annual outlook range of 85 to 87. Allstate brand auto and homeowners insurance continued to deliver attractive returns and generated underwriting income of $1.7 billion and $472 million respectively in 2018. Turning to slide nine, let's review Allstate Life benefits and annuities. Allstate Life, shown on the left, generated adjusted net income of $68 million in the fourth quarter, up 19.3% from the prior year quarter as a lower effective tax rate and higher premiums more than offset higher contract benefits.
Allstate Benefits adjusted net income, shown in the middle chart, was $25 million in the fourth quarter. The $5 million increase from the prior year quarter was primarily driven by increased premiums and a lower effective tax rate, partially offset by higher expenses. Allstate Annuities, on the right, had adjusted net income of $31 million in the quarter, which was $24 million lower than the prior year quarter. This was primarily due to lower performance-based investment income. Moving to slide 10, our service businesses continue to provide strategic and economic value. SquareTrade written premium increased 107.1% to $323 million in the fourth quarter of 2018, and revenues increased to $137 million. Adjusted net income was $9 million in the fourth quarter of 2018. Arity continues to invest in advancing our telematics platform. In the fourth quarter, Arity had revenues of $24 million, primarily related to affiliate contracts.
Allstate Dealer Services revenue was $105 million for the quarter and $403 million for the year. Adjusted net income was $6 million in the fourth quarter of 2018, benefiting from improved loss experience. Allstate Roadside Services revenue was $74 million for the quarter, with an adjusted net loss of $7 million, in line with the prior year quarter. InfoArmor is the newest addition to the service businesses segment, contributing over one million policies in force. Slide 11 takes a closer look at the acquisition measures of success for SquareTrade. SquareTrade is delivering on the three objectives supporting its acquisition as domestic policies in force increase, loss experience improve, and the business is expanding beyond U.S. retail.
SquareTrade written premium increased 81.5% to $773 million in 2018 due to the continued growth in U.S. retail as SquareTrade became the exclusive protection plan provider for a leading U.S. retailer in the second half of the year. Adjusted net income was $23 million in 2018, reflecting full year improvement of $45 million. Slide 12 highlights our investment results. We proactively manage the investment portfolio considering market conditions, the nature of our liabilities, and risk appetite. As a result, we have a high-quality, market-based investment-grade bond portfolio that generates substantial annual income and cash flows. We also have a performance-based equity portfolio that generates higher returns for longer-dated liabilities. The chart at the lower left shows net investment income for the fourth quarter was $786 million, lower than the fourth quarter of 2017.
Market-based investment income increased, reflects higher purchase yields and a modest duration extension for the property liability and fixed income portfolio, partially offset by a reduced allocation to high-yield bonds. Performance-based investment income generated $145 million in income in the fourth quarter. Performance-based income for 2018 was solid with a yield over 9%, but fourth quarter was below the prior year quarter, reflecting a lower number of sales of underlying investments and more moderate asset appreciation. The components of total return are shown in the table on the right. The -0.2% return in the quarter includes a stable contribution from investment income, reduced by lower fixed income and equity valuations. The total return for 2018 was 0.8%. Slide 13 provides an overview of returns on capital. We continue to generate attractive returns on capital.
Adjusted net income return on equity, shown on the left, was 14.8% in 2018, an increase of 1.4 points compared to the prior year. Our capital position remains strong, and we returned $1.2 billion to common shareholders in the fourth quarter, bringing the total cash return to shareholders to $2.8 billion for the full year, as you can see in the chart on the right. As part of the $3 billion share repurchase authorization, we executed a $1 billion accelerated share repurchase program in the fourth quarter.
Additionally, in 2018, we completed the acquisitions of InfoArmor for $525 million and PlumChoice for $30 million and redeemed $385 million of our preferred shares. We continue to take a proactive approach to managing our capital. I'll open up the line for questions.
Certainly. Ladies and gentlemen, if you have a question at this time, please press star then one on your touchtone telephone. If your question has been answered and you'd like to remove yourself from the queue, please press the pound key. Our first question comes from the line of Elyse Greenspan from Wells Fargo. Your question, please.
Hi. Good morning. My first question is just on the severity trends that you're seeing within your auto book. PD severity remained elevated for the second consecutive quarter. If you can just tell us, is it the same factors that you saw driving that last quarter? Then within your underlying margin guide for 2019, are you assuming that severity trends will remain elevated?
Elyse, this is Tom. Glenn will answer for the Allstate brand. If you want to go into the other brands, let us know. As you know, we don't give breakouts on what's in the underlying combined ratio guidance.
Thank you, Elyse. I want to start with just a little bit of context on it and then talk about the issue itself and actions from it. From a context standpoint, as you know, the losses represent about 70% of the combined ratio, and within those losses, because we price Allstate auto in total, you have physical damage is about half of it, and then you've got the injury coverage, which is about half of it. Within those, you have frequency and severity impacting each of those. Of those four quadrants right now, we actually see three of them performing as expected or well. We've got favorable severity in both injury and physical damage, and then the injury lines are performing as expected, as you can see in our reserve releases.
No question there's pressure on the physical damage part of it, and they are similar trends to what we saw in the prior. The issue's an industry-wide one. We've seen more sophisticated cars costing more to repair. As a result of that, particularly at the high end of repairs, more cars are then reaching that capitation level of a total loss. You'll see across industry trends, not only in the Fast Track numbers of overall severity up, you'll see that more cars are reaching that total loss threshold, particularly in recent model years. I'll give you a specific example just to illustrate, because our team in claims, and our actuaries, and our product folks, they do a ton of work getting deep into make and model and what's driving different trends.
The specific example, take a Camry, which is a pretty common vehicle, and we looked from 2013-2018, what the changes were in repairing a front-end impact. The answer is, in that five-year window, it doubled. It's the impact of more sophisticated cars with sensors that are actually helping us on the frequency side, but they're costing more to repair. That added $1,700 alone, just the sensors that help avoid accidents. A really specific component taken on that car specifically is the headlight. The headlight went from $360 in 2013 to over $1,900 in 2018. I give you the specifics in the example for two reasons. One is to give you an idea of what's going on and what we see in make and model. Two is to illustrate we do go deep on this issue, and we look at it very thoughtfully.
That takes you to what do you do about it? One lever is rate, which obviously we look at on a market-by-market basis, where we need to take pricing. The next would be to be more sophisticated and laser-focused on that rate. How do we get better and better in the make and model pricing we do, which we do today, specific make and model pricing. The more we learn and the deeper we go, the more sophisticated we get on specific make and model pricing. The third one would be working with OEMs and the repair community on improving the overall cost to repair, because as more and more cars are reaching that total loss threshold, I don't think it behooves anybody to have disposable vehicles in the newer model years, and we want to be able to repair cars and return them to customers.
In answer to the question about whether it's included, we absolutely include all the factors that we look at in giving guidance, and it's included retroactively in all of our financials. Our alternates are projected in there.
Okay, that's very helpful. My second question, very recently, we've seen a whole host of transactions in the annuity space as companies have entered into deals that have helped to free up capital. Could you let us know your thoughts? Would Allstate be interested in entering into transaction with their annuity exposure? If you did, if there was a transaction that was able to free up capital, would you be more likely to use that for share repurchase or potentially to hold on for M&A opportunities?
Elyse, we obviously don't comment on any transactions that people think we could, should, or maybe are thinking about. I have no comment on that. What I can tell you is we've always been very aggressive about managing our capital. I'll do the second part first. Whether that's to do share repurchases, to buy companies where we're a better owner than somebody else, to continue to grow our business. That part won't change at all as it relates to that. We have, as you know, been working hard on improving the long-term economic returns on the annuity business, which have hurt the current book returns.
Specifically for the payout annuities, which are long-dated liabilities, some 20, 30, 40 years, the right way to invest behind those liabilities is much like you would a pension fund, put it mostly in equities, because once you get past 10 years, your return on equities is twice what it is in bonds, and your risk is lower. The downside to that is that the regulators require a significant capital requirement for equities because they think about the equities as being volatile on a day-to-day basis, not on a 10-year basis. We've been working with the National Association of Insurance Commissioners to come up with horizon-based capital standards, particularly as it relates to equities and the payout annuities, which would free up a tremendous amount of capital for us that we could then use the way I started to talk about.
We're looking at every possible way we can improve every element of our capital deployment each and every day, whether that's the annuity business, using reinsurance, using preferred stock to buy back stock. It's just a part of life for us. You should know we're focused on how do we improve returns, annuities, and to the extent there's opportunities where we can take advantage of them, we run for it.
Okay. Thank you. I appreciate the color.
Thank you. Our next question comes from the line of Gary Ransom from Dowling & Partners. Your question, please.
Good morning. I'd like to dig in on Allstate brand PIF growth. You've talked in the past about how it takes time to energize the agents, but we're now seeing more improvement. I'd like to just ask, how is that growth spread among the agents? Is that subset of growing agents building throughout the past year? Maybe that's one part of the question. Another part of the question is there any change in the quality of your new customers? Are they more bundled or, as you talked about, are more of them using telematics in that process? Sort of two parts, the spread among the agents and then the character of the new customer.
Gary, good morning. Let me maybe go up a minute and then come back down to Glenn, and he can answer your specific question on-
Okay
the agency owners, the breadth of distribution, what we're seeing. Obviously we have a variety of both short and long-term strategies to grow, and each brand is different. There's differences by geography and by component of the leverage you can have in each of those. Those growth drivers are customer satisfaction is up in all of our brands last year, and retention is up in all of our brands last year. That's a big driver of growth. Obviously, distribution is one you mentioned. There's also the level and quality of our advertising. There's the breadth of our product offering. We pursue all of those. Glenn will talk about the auto, home life in the Allstate brand. If others want to get into Esurance, Encompass, or the Transportation Network Company, Steve can handle that one. Don obviously heads the services businesses.
Glenn, why don't you deal with Gary's questions directly?
Thanks, Gary. First talking about the agents. It's a fun part of what's going on right now in terms of our growth, and I call it a system-wide growth because you're touching on an important part because you can see the policy growth, you can see the premium growth, but what's really happening underneath that is we've got 2,700 more points of sale. New agents join us to start a business or new salespeople join an agency because they want to win. They don't start it to stay flat. They want to grow. We're attracting them because right now we're competitive in the market. There's good health in the system, as Tom said. We're retaining more customers, which is helpful to the growth.
With those additional points of contact, we have good momentum sort of built into the system, and we're growing in a healthy way, and we're competitive in the market. You also asked about the new business coming in and how it's bundling, and you asked about telematics. Quickly on those, bundling has been very favorable. We're bundling more of our new business than we have historically. We also have more people buying full coverage than we've seen historically. Both of those are good bellwethers for the quality of business. From a telematics standpoint, we grew that by 30% this year. We added about over 300,000 active connections to our Drivewise. On the Milewise side, which is the pay per mile, as Tom mentioned before, I think about it like a startup. It's small numbers, but it's growing like crazy.
In the fourth quarter alone, we over doubled the size of Milewise as an offering for us. It's an exciting area where we're growing our business.
Just as a follow-up, are there any areas right now where the PIF growth is lagging but has the potential to do better in the future?
Well, this is Tom. I would say that if you look at the Allstate brand, Glenn's got a number of strategies, different states where he thinks he can grow faster. Obviously, retention is always a benefit that we get across. Esurance, you saw, had a 10% growth last year, so we're feeling good about that, and it's back on the picking up decent market share gains. Encompass, less so. That would be one where I would say we still need to come up with a sustainable growth plan that drives items in force growth.
Okay. Well, thank you very much.
Thank you. Our next question comes from the line of Greg Peters from Raymond James. Your question please.
Good morning. The first question is around the homeowners business. I was looking in your statistical supplement, I think it's page 25, where you talk about gross claim frequency and paid claim frequency, and it being up almost double-digit in the fourth quarter. Actually for the last couple of quarters, it's been up pretty strong. I looked at the Allstate homeowners' underlying loss ratio, and it improved in the fourth quarter. I'm trying to sort of bridge the gap there and understand what happened.
Greg, good morning. First, I'd like to set a little bit of context. The Allstate brand homeowner business is very attractive for us. $7 billion of premium, made $470 million of underwriting income in a year where there are a fair amount of catastrophes. It's a business we like and does well. Obviously, things bounce around from quarter-to-quarter. Glenn can talk about this specific question.
Yeah, Greg, I think there's a mix there of the quarter-over-quarter versus the year-over-year change. It did come down from the 3rd quarter to 4th quarter, but year-over-year, it was up a couple of points in terms of the underlying combined. To Tom's point, in a year with a lot of catastrophes, particularly in the 4th quarter, we produced a recorded combined. I like to go to the recorded because managing the overall risk over the long term of catastrophes is a big part of what we do, even though we focus on the underlying. 93.3 in a year that had the type of catastrophes we had was pretty strong, before the 4th quarter, you'd have to go back almost 7 years to have a quarter that didn't produce an underwriting profit.
Strong work across the system, which is the underwriting approach that we take. It's the pricing, the reinsurance, which obviously played a part in the 4th quarter for us and our claims approach that we think helps mitigate costs. Part of your question relative to the underlying, I would say 2018 was an extremely wet year. I mentioned that in the 3rd quarter, continued into the 4th a bit. It was a heavy rain year. We got some pricing for that. We've reacted to it. You saw the 1.1 of rate in the 4th quarter, which is rate, not premium, because we do have automatic inflationary factors in the policy as well.
Thank you for that answer. Tom, I know you highlighted the success of Esurance, I was looking at the underlying loss ratio, I guess it's seasonally high in the 4th quarter, but it certainly seemed to be trending up from what I used to think you used to like to target in the low 70s. The underlying loss ratio seems to be pushing up close to the 80. I'm wondering if there's anything that I should be thinking about with that loss ratio and the underlying loss ratio as it relates to future growth in that business.
Greg, it's Steve. You're absolutely right. Our targets have generally been, we said the lower of 70, below 75 range. For the quarter, it was elevated. A couple reasons for that, I think is, you see 1, given the growth we had in the business during 2018, we have a greater proportion of new customers, and there's some higher loss ratio with newer customers. That's only a piece of it, though. Clearly, we have had the same type of issues in terms of severity that we've talked about in the Allstate business. You look at the offset to that, we brought our expense ratio down a large amount over the last 3 years. In total, we actually have a combined ratio on an underlying basis below 100 for the whole year. For all but the last quarter of the year, we've had good results.
We are starting to get profitable in the business as we grow. We feel good about the overall profitability. You're correct. We need to work harder in that underlying combined ratio and drive it down a little bit from where it is today, and that will accelerate our growth. We feel really good about the growth and the profitability for 2018.
Thanks, Tom, Steve, and Glenn, for the answers.
Thank you. Our next question comes from the line of Mike Zaremski from Credit Suisse. Your question, please.
Hey, good morning. My first question is about auto accident frequency. Can you remind us what Allstate's long-term trend has been on frequency, and does that differ from your estimate of 2019 frequency? If it does, maybe you can talk to why.
Okay. By long-term, we can go real long because I've been hanging around the hoop for a long time. Mike, for, I don't know, 15-plus years, it just came down pretty steadily as we had third brake lights on the car, anti-lock brakes, airbags brought some stuff down. It worked on both frequency and severity for a long period of time. It kind of flattened out for a while, and I don't have the exact years in my head, but it kind of flattened out in about 10 years ago and wasn't really going up or down much. Of course, in 2015 and 2016, it just spiked way up. Still unclear exactly as to why that happened, somewhat miles driven , some by economic activity, which is related to miles driven.
There have also been people that speculated it's due to the big cell phones and stuff like that, distracted driving. I think the answer is it's really hard to get attribution as to why. Glenn talked about the decline in frequency since then, which has been in part due to less miles driven, but also safer cars and some of the trends that were present earlier in that long-term cycle. A little hard to tell exactly what it is. As it relates to the forecast for next year, I'm sure everybody has a thought about this. As you know, we try not to go through the components of the guidance. We do make an estimate of what we think is likely to be, and that includes some trends that are just external, we don't control. For example, frequency would be one of those.
We can't control whether people get in accidents or not. There are some which are partially controllable, and Glenn talked about that severity and the physical damage or bodily injury. You can control some of that. Some of it's inflation, some of it's the cost of parts. You can try to do a good job making sure your customers get the right amount, not too little, not too much. There are trends we create, like price increases and expenses that both Glenn and Steven talked about. When we're looking at the underlying combined ratio guidance, we look at the trends we can't control. We look at the ones we think we can have some influence on, and then the ones we do have influence on. We, as a result of that, set our guidance at 86%-88%.
It's obvious, of course, that that's just a small part of the value created. As mentioned, it doesn't include catastrophe losses. It doesn't include prior reserve changes. It doesn't include things like pension settlement charges, which I think led to some confusion in the third quarter. It doesn't include investment income, it doesn't include our life business, which makes almost $300 million a year. It doesn't include Allstate Benefits. As it relates to doing projections on the business, we think it's useful as a way to help you understand how we feel a large component of our business is operating. I always try to caution us against thinking that it's the only value creator and the only thing under which shareholders should decide as to whether this is a good story or not. Was there anything specifically on frequency in the fourth quarter that we can answer for you?
No. I guess it's just the answer was there's not really a conclusive long-term trend line to frequency, if you think I understood your answer correctly.
No, I think it just bounces around a lot, Mike.
Okay.
What you do want to do and what we try to do is just be flexible, quick, and adapt to it. If it goes up, you've got to raise prices. If it goes down, you have to make sure it goes down, and then maybe you don't raise prices as much and you get a little better retention. You've seen both of those stories in the last four years.
Okay, great. My follow-up is regarding telematics. Thanks for the information, and you gave out one stat earlier in the call saying telematics, Drivewise policies, I think grew 30% year-over-year. Can you give us maybe some more stats or color on what the adoption rate is for telematics and maybe how we can try to start thinking about the adoption curve longer term? How are you guys going to get these policies into the hands of your policyholders?
Glenn gave it a number of 30%, the Allstate brand. The Esurance brand was up substantially more than 30%. They really got focused on it this year. We're finding the adoption rate to be relatively high and going up, but you have to manage it. You have to manage the initial conversation, you have to manage the onboarding, and you have to manage, not just the conversion rate, but the retention rate. You also have to manage the ongoing relationship. We focus on all of those components. We've made progress on all those components in 2018. We don't give the specific numbers out because it's just too competitive, and we want everybody else to do it. As it relates to where it's going to go, I think most auto insurance is going to include telematics pricing. It's just way too powerful.
It's every bit as powerful as credit was when that was adopted. The industry is going to adopt it. We made that call when we decided to build Arity as a platform that we thought it's too powerful a rating factor. If you go back to just the location, if you price your auto insurance by where somebody's car is garaged, people figure that stuff out. They garage it one place and drive it someplace else because it's cheaper. Not everybody does that, but it leads to small inefficiencies in the market. To the extent you can be more efficient on when, where, and how they drive, it can give you much better pricing. It's going to happen.
It's hard to tell what the adoption rate will be, but I think you could look at credit as an example and see how that rolled out in the industry over a period of time. Basically, almost everybody's using some form of credit or some proxy for it today.
Thank you.
Thank you. Our next question comes from the line of Amit Kumar from Buckingham Research. Your question, please.
Thanks. Good morning. I wanted to go back to your example on auto parts in response to Elyse's question. I think not a day passes that we see new headlines on auto tariffs and tariffs on auto parts. Seems like we're getting closer to some sort of conclusions. Have your conversations internally or with the repair shops changed? How are you thinking about in case we do see some action on that?
I'm sorry. I missed the action on what? What was that? I missed it.
On the auto tariffs discussion.
Auto tariffs. Okay, yes.
Yeah.
I'll let Glenn talk about what he's doing specifically inside the business. Obviously, tariffs on steel, and to the extent parts are made of steel, becomes expensive. That rattles through the pricing. It's obviously an indirect cost.
In terms of the cost to repair and buy these vehicles, I think it's time to actually have a conversation about the cost to repair versus the cost to buy a car. If the auto manufacturers are giving away the razor to sell the blades as a way to make profit, and that shows up for individual consumers as higher insurance prices, which they don't really understand because it's embedded in the fact that Glenn's example that the headlight now costs $1,900 instead of $300. My first car cost less than the $300, certainly I could've gotten a better one for $1,900. I think that's an economic trend between industries that we need to sort out. Glenn, anything you want to say about the specific parts and body shops?
Yeah. There's a lot of different approaches we take, and our claims team does a really great job working on agreed pricing with large auto body shop aggregators, and there's a number of components in how they're trying to manage overall cost. Specific to the tariffs, it's something we're keeping an eye on, similar to other things we look at from an inflationary standpoint. Like, we look at the cost of petroleum because a lot of auto parts and certainly roofing materials and homeowners is petroleum based, and we look at price of steel, to Tom's point. The tariffs would be another inflationary factor we look at because about 60% of auto glass, for example, comes from China. You have a significant percentage of metallic auto parts coming from China as well. As those work through the system, it's essentially an element of inflation that we watch.
Got it. That's helpful. The only other question I have is just going back, I guess to the discussion on pricing. If you look at page 22 of the stat supplement, it says 25 number of locations, the Allstate brand change was only 0.3% for Q4. How should we think about broadly the trajectory of rates from here into 2019 in terms of planned rate actions? Thanks.
Yeah, this is Glenn. I'll take that, Amit. First of all, I think setting context on it, we had the same underlying combined ratio two years in a row. We look at our margins, and we manage for attractive return. I think we've shown that where we went after pricing when we had to with the frequency spike a few years ago.
We showed it in the fourth quarter on homeowners, where we saw some trends there and had to go after pricing. On a state-by-state and market-by-market basis, we're going to look at the pricing and keep in mind our competitiveness, our need for rate, and act accordingly with it. Where you saw the last couple of quarters with very low rate is we had taken a lot of rate a little bit ahead of the rest of the industry following that frequency spike, and we were more rate adequate in more places. It was reflected in good, stable rate environment for our customers, which helps with the retention, helps with growth. As that turns, we will on a market-by-market basis, make those decisions.
I can't project forward obviously for you what rate we'll take, but I think our history is good evidence of the fact that we'll take the rate necessary to make sure we're managing our margins appropriately.
Got it. That's helpful. Thanks, and good luck for the future.
Thank you. Our next question comes from the line, Kai Pan from Morgan Stanley. Your question please?
Thank you and good morning. My first question follow up on your guidance for 2019. The range 86%-80% exactly the same as a year ago, but the operating environment is a little bit different. Pricing today is less than a year ago, and loss trend side frequency better today, but severity a little bit worse. Just wondering, do you build in the same level of conservative in the initial guidance that will still end up to be at lower end or better end of the guidance as we did in 2018? This operating environment is totally different?
Kai, obviously, you're correct in saying that 2019 is not 2018, Wasn't the prior 10 years or 11 years, how long we've been doing this. We look at each year differently. We feel comfortable at 86 to 88. There was one word you said that I don't think is accurate, which is conservative. We set the range that we think we're going to be in the range. We don't set the range where we think we're going to be below the range. We pick a number we think is reasonable. This year, we revised the range down, as you pointed out, after the first quarter because frequency was down so much relative to the prior year that we didn't think that we would be in the range anymore. Obviously, we're not.
We were in the middle of the new range, but not where we sat at the beginning of the year. That's one of those trends, though, that I mentioned, you just can't forecast. We have no ability to forecast frequency in total. We have our estimates. We look at our trends. Glenn goes by number of times renewed by state. He looks at frequency, make a model year. He's got all kinds of work on it. If you said, do we have a solid number that goes into 86 to 88? We make an estimate on that frequency and what we can manage to. I think that's the other important part is we think we can manage to 86 to 88, depending on what happens in the marketplace.
We think that's the right place to be because we're earning incredibly high, or we're earning very attractive returns on capital at that level. We should be growing the business then.
That's very fair. My second question on SquareTrade, the top line growing for the full year, about 80% year-over-year in term of return premiums. If you're excluding Walmart, and what's underlying growth rate has been for that business, and also do you have a sort of long-term view on the profitability of the business? I just wonder, one, can this rapid growth, meaningful top line growth, will translate to a material impact on your Allstate earnings?
As Mario mentioned, we set three objectives when we did the acquisition, things we needed to do, and that included the growth you talked about, in the domestic retail business. We've achieved all three of those objectives. Don can give you some color on the breadth of the growth at SquareTrade.
When you look at the full year, the company obviously has been growing dramatically. Part of that has been new customers, as you pointed out, but part of it's been just doing a better job with the customers we have. Having them become more able to serve more of their customers. If you look at the fourth quarter, which was obviously a terrific year, there was a large component of the growth in policies that was the result of a large retailer. Even without that, we still would have had a healthy growth number just because we're growing the existing books that we have as well. As far as the profitability on a long-term basis, you're right, we did set as one of our measures of success, increased profitability, and returns on capital.
We've been very happy with the improvements we've seen, we're trying to do it in a balanced way that builds the business for the future. Tom said in his opening comments that Allstate's strategy is to grow by protecting people from life's uncertainties. SquareTrade is a key component of that because, it's not just auto and home people need to be protected by, from, it's also other things in their lives that are important. We could probably have made the numbers better. In fact, I know we could have made the numbers better in 2018 by skimping and not investing for the future. I think we're seeing the improvement in profitability in a nice, balanced, long-term, sustainable way. Loss rates continue to perform attractively for us.
We said earlier that we're going to improve our scale of expenses as we grow the business, the growth has helped us do that. We brought our paper in-house, we're underwriting the risk at Allstate, which we're obviously capable of doing, which helps improve profitability as well. I'm delighted with the improvement of profitability, I think it's the combination of profit and growth working together that is most attractive at this point.
Kai, as Don said, look, we create shareholder value by component, so you're right to ask about the components. We think about it that way, we also look in total. To the extent we think there's another component we can grow, even if it hurts the total a little bit, we'll do it. For example, SquareTrade did really good growth in European telco business.
That's great. Well, thank you so much for all the answers.
Thank you. Our next question comes from the line of Yaron Kinar from Goldman Sachs. Your question, please.
Hi, good morning. First question is on industry pricing and auto. Are you surprised to see that rate filings aren't really showing much improvement given the severity trends in the second half of the year?
Yaron, I don't know. We don't manage other people's books of business. We watch what they do. There are some people that run combined ratios higher than we think you need to. We think that the economic rents you can extract in the market relative to value creates are ones that we're very comfortable with. If you look at our combined ratio, whether that be auto or homeowners, we think it's very attractive. There are other people running combined ratios in homeowners insurance, which we wouldn't do either, and they're growing quite quickly. We try to manage our own book and look to them. I don't think these trends that you're seeing in severity will impact other people, is our assessment.
I can't say when they'll have to react to that.
Okay. Let me try a little differently then. I think heading into 2018, you had said that you thought that your pricing put you in a competitive advantage. I think that was a little bit of a change from the prior few years, where you had taken early pricing action. As you look at the environment into 2019, how do you think of your pricing relative to the industry in auto?
Glenn can talk about how he's thinking about pricing relative to auto, I want to just take it up a little bit, so it's both growth and profitability, and doing a better job for our customers. I think what you saw in 2018 in terms of our growth was not just fewer price increases, but the fact that for three or four years, we've been working hard on a Net Promoter Score. The benefits of those improvements got overshadowed by the fact that we did what you described, which is we raised prices faster than other people. Our retention went down. If you're looking at profitable growth, it's both what we price to, which Glenn can talk about, but also how much of our existing business we retain.
Okay.
I'll just add on, as we look at pricing, we see competitors as you do in filings. Some are still taking some rate, others aren't. The CPI most recent number, which is a trailing indicator, but was between five and six, and it's come down from much higher than that, and expect it to continue to come down because the filings are different now than they were late 2017 and early 2018, which create that five and change CPI in auto. The other thing I would look at is, we look a lot at the absolute price change, but we don't start with what the competitive position of a particular carrier is. Now we look internally at our competitive position by market and by competitor, and so somebody may take a 3% price decrease.
If they're 10% higher than the market and they take that, well, they're still higher than the market. In an absolute level, we're looking at our competitiveness on a more granular level than the price changes.
Okay. That's helpful. I appreciate that. Then one on capital deployment, if I can. I think this year you deployed well more capital than you generated in earnings, and I think that's a multi-year trend now that we're seeing. How sustainable is that, barring a change in asset requirement or capital requirements by the regulators?
The R&D deployable capital, are you talking about share repurchases in comparison to earnings?
I'm talking about share repurchases, dividends, M&A, all the deployment avenues.
Well, of course, we use preferred stock.
Right
We use preferred stock to restructure some of the balance sheet. I don't think you should assume that what I heard, it might not be what you were implying was, are you just jettisoning a lot of capital from the business more than you make? That's not the way we think about it. We think about deploying capital to the extent we don't need to deploy capital, or we can get alternative capital that's cheaper than having it from our shareholders, we do that. That could be, we use a bunch of reinsurance, it could be other alternative capital, it could be preferred stock. We're always seeking to try to optimize the return on our shareholders' capital by using other capital. We don't have sort of a metric that measures earnings relative to just dividends and share repurchase. We think about it broader than that. If that makes any sense.
Yes. The only thing I'd add is, we've been saying for a long time that our capital position's really strong, and we continue to generate capital organically. We start from a position of strength. We make individual decisions around acquisitions and buybacks. There's no formulaic way of approaching it, we still feel really good about where our capital position is, even after the $2.8 billion that we returned to shareholders this year.
Okay. Thank you.
Hey, Jonathan, we'll take one more question.
Certainly. Our final question for the day comes from the line of Brian Meredith from Autonomous Research. Your question please.
Hey, thanks. Just a couple. It's interesting because in the past, the homeowners hasn't been so volatile on an underlying basis, but this year it actually was a pretty big headwind. Just curious, when you think about your guidance for next year, directionally, what are you contemplating for margins at home?
Brian, we don't like to give underlying combined ratios for the different components. Let me just maybe make a comment about the pretty big headwind. The underlying combined ratio for homeowners, which is in the low 60s because there's such a large amount of catastrophe that come from that business, and as Glenn said, we manage it in total. We give you the underlying combined ratio because we can't estimate catastrophe. Yes, the underlying combined ratio was up some in 2018 versus the prior year. It's still though, we still earn a 93 combined ratio, made $479. I don't know that we thought of it as a headwind. Are we paying attention to it? Glenn can talk about the underlying combined ratio of components on homeowners.
We took it seriously. We talked about it last quarter that we were seeing more non-CAT weather, and we had to look at that. As we always do by market, we looked at pricing. You saw us react with some price, 1.1% in the quarter, when you annualize that's a meaningful move on pricing. On top of that's rate, we get an inflationary factor. The premium is a larger number than what you see in rate. We can't predict, much like Tom talks about frequency, we can't predict whether we're going to have another significant non-CAT weather year. We know that we look at trends, and our actuaries work on this, whether you look at a five year, 10 year, 15 year. If the one-year trend looks worse than the five, which looks worse than the 10, generally.
If weather patterns are getting worse and here to stay, then we're prepared to react to that with the right pricing.
Got you. On the autos side, was there a current accident year true-up at all there that swung the loss ratio one way or another in Allstate brand?
A credit or a true-up? Are you talking about prior reserve releases or?
Was there a current year, like a reserve release or a change in the loss pick in the fourth quarter?
We of course changed. Why don't we have John take you through that. Here's what I would say, is one, we think the numbers accurately reflect what we earned in 2018. We make changes in reserve picks for the specific year in which we're operating as we go through the year. Sometimes if you make them in October and you should have made them in June, then you got a little bit of a catch-up, and I think that's what you're referring to. It's not a significant element as it relates to the Allstate brand. Some of the smaller brands can have a little bigger quarterly swing, because of those true-ups. The Allstate brand is usually big enough where it doesn't really make much of a difference. John can help you sort out any additional pieces.
Thank you for being on the call. Allstate's strategy is obviously to protect people from life's uncertainties. We'll continue to reward shareholders who decide to be part of our story, which is by innovating, growing market share and earning attractive returns. Thank you all for being on the call. 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. Good day.