All right, we're going to try and get back on track a little bit. We'll move into the next presentation. I'm Alison Jacobowitz. I work with Jay Cohen, for anyone who's not met me yet. I'm extremely pleased to bring up Allstate now. We have Tom Wilson, Chairman and CEO of Allstate, as our next presenter. Tom has been with Allstate for over 20 years and has been CEO since 2007. Under Tom's leadership, Allstate has emerged as a leading innovator in the personal lines industry, and the company continues to evolve in reaction to changing consumer appetites. Tom has been a regular presenter here, and I always look forward to hearing his comments. With that, I'm going to turn the podium over.
Thank you, Alison. Good morning, everybody. Thank you for investing your time to learn more about Allstate. Joining me today is John Grieco, who leads our investor relations team. I'm going to quickly breeze through some slides to set some context about Allstate, then we can have a conversation with Jay and Alison. Before we begin, is that the Surgeon General warning? Yes, it is. This just says I will be making some forward-looking statements that are in non-GAAP measures. If you want this presentation and lots of other specific information is on our website. Let's start with 2017. We delivered really strong results in 2017, and we're positioned for profitable growth. Net income for the full year was $3.1 billion, or $8.36 per share.
That did include a $506 million increase in net income in the fourth quarter, which was a result of the recently passed tax legislation. That one-time tax increase was the result of a reduction in our net deferred tax liability, which was primarily related to deferred acquisition costs, which we've already expensed for tax purposes at the higher rates. Some of the adjustments are bookkeeping. This one is cash. Also some impact on our unrealized gains. Adjusted net income was $2.5 billion in 2017, which was 34% above the prior year, reflecting strong performance across all of our market-facing businesses and investments. Auto and homeowners insurance margins remained strong, and our performance-based income did quite well also. Adjusted net income return on equity was 13.3%, as shown on the bottom, that was a significant improvement over the prior year.
This operational strength will support accelerated growth in 2018 while maintaining attractive returns. We also had good results on our five operating priorities. We post out our operating priorities for the year by better serving our customers, and that promoter score increased customer retention for the Allstate brand auto insurance improved in the second half of 2017 and for Esurance auto and home across the whole year, which if we maintain that, results in increased growth in 2018. We continued to deliver excellent results on capital, as we just saw on the last slide. Allstate Life and Allstate Benefits generated attractive returns, annuity income was up, the returns are still low. Total policies in force grew to 82.3 million. Property liability policies, in force, however, declined due to the impact of some profit improvement actions that we began in 2015.
We expect those lines to return to growth in 2018. SquareTrade policies grew $10 million, or nearly 36%, in 2017. Allstate Benefits continued its 17-year track record of growth, with policies in force increasing by 7.4% last year. Total return on the $83 billion investment portfolio was 5.9%, reflecting strong results from both a market-based and performance-based portfolios. We also built long-term growth platforms. SquareTrade achieved our early goals on the three factors of success we laid out when the acquisition was closed in January. It already signed its first third-party insurance customer to expand its platform outside of the Allstate entities. Our priorities for 2018 are consistent with those in 2017. The first three of those priorities, that's better serve our customers, achieve target returns on capital, and grow the customer base are all intertwined and ensure that we have multiple paths to profitable growth.
In 2018, the Allstate brand, Allstate Benefits, SquareTrade, and Esurance are all expected to contribute to growth. These priorities will generate shareholder value through profitable growth of our core businesses, leveraging our connected car and investment expertise, and entering high-growth markets through our services segment. I think everybody probably knows this slide. I think Jay can probably fill it out if I just gave him the axes. This is the way we segment the property liability customer base. Let me just go through it quickly to set a little more context. The lower left are our customers that want local advice and assistance in selecting their personal insurance and believe there's a difference in insurance companies, so the brand matters.
This segment is served by the 10,400 Allstate agencies that are present in nearly every community in America, that we compete aggressively with State Farm, Farmers, and Nationwide in this segment. The lower right are customers who prefer to handle their own insurance needs, but they still want a branded product. They're served by our direct company, Esurance. We compete aggressively with GEICO, Progressive Direct, and USAA for these customers. Encompass in the upper left competes for customers who want local advice and are less concerned about a branded experience and are served by independent agencies. We're the only company that has a specific offering for all four segments in this chart. The Allstate agencies, that's our largest business, of course, and it has a multifaceted profitable growth plan.
Their Trusted Advisor initiative will better serve this segment of customers and is expected to raise retention levels and consequently, policy growth. Growth investments in agency deployment and marketing are being increased to accelerate new business growth. We're building an integrated digital enterprise which uses technology, data, process redesign, and advanced analytics to improve effectiveness and efficiency. An example of this is our QuickFoto Claim, where people get in an accident, they take 22 pictures of their cars, takes them about five minutes. They send us the pictures, and enable us to shut down drive-in facilities and not pay adjusters to drive around looking at cars all the time. We now settle claims in hours, not in days, and at a lower cost. We're also using our telematics pricing expertise to improve our growth and pricing position.
Allstate's Drivewise telematics product is expected to give us a substantial competitive advantage through more sophisticated pricing and a better customer value proposition. Our approach is a little different than some. Ours is to establish a continuous connection with our customers through either their OBD port devices or mobile phones. That gives us the ability to provide customers a more accurate price. But it will also give them a better value proposition. This will give us more profitable growth, just as the introduction of financial responsibility into pricing did 15 years ago. Drivewise will have a broad set of customer offerings that improve their driving experience later this year. For example, customers can review their driving, they can find out where their car's parked, they can get information on their car's performance. You won't have to look at that annoying maintenance required sign anymore.
You can earn rewards for being a safer driver. It can also be used to settle a debate with your spouse or significant other on who's the better driver, but you can use facts in this case. We were early into auto telematics, having initiated Drivewise in 2012, and are connected with about a million customers. Our strong profitability and return to growth of the property and liability business will create additional shareholder value. As shown in the top left chart, Allstate Protection had great results in 2017, reflecting improved margins in auto insurance, which are shown in that blue, as we successfully reacted to an increase in auto claim frequency in 2015, which reduced margins. Homeowners profitability, shown in gray, declined last year due to a $649 million increase in catastrophes, but still generated over half a billion dollars in underwriting income.
Favorable prior year reserve re-estimates of $621 million, that's just shown in orange on that chart, were primarily the result of lower expectations for future loss costs than we had estimated for auto insurance. It's particularly focused on accident years 2015 and 2016, that was a large part of that orange bar, which means that the blue bars would've been higher in 2015 and 2016. The top right chart shows the Allstate brand auto insurance property damage gross frequency. You can see the large increase in frequency in 2015, which we reacted to quickly by increasing prices and decreasing new business levels. Hence, that's why our growth went down in 2015 and 2016, and last year. You can also see the large decrease in frequency in 2017. Since 2013, year-over-year changes in gross frequency have been as high as 6% and as low as -5%.
Last week on our earnings call, there was a number of questions on why the underlying combined ratio outlook we provided for 2018 was above the actual level in 2017, which you can see by that little dot for 2017 on the bottom chart. Let's look at that issue in a little more detail. The box on the bottom shows the underlying combined ratio guidance we've provided over the last five years, which has been from a low of 86 to a high of 90. This range, when you add in catastrophe losses, generates very attractive returns on capital anywhere in that range then.
If you look at the top right-hand graph and the bottom graph at the same time, you can see when frequency increases are low, such as 2013, '14, and '17, we tend to be at or below the range, at the bottom of the range or below it. When frequency increases are high, such as in 2015, we tend to be at the higher end of the range. There are many factors, of course, that go into establishing the annual guidance, some of which are in our control and reflect on how well we operate the business. Factors such as frequency, of course, are not controllable, and that volatility needs to be factored into the range as well. In 2018, the property and liability business is expected to have an annual underlying combined ratio between 86%-88% as we accelerate growth while maintaining strong returns.
When you factor in historical catastrophe losses, this represents attractive returns on capital, so growth is accretive to shareholder value. Shareholder value is also increased by disciplined, balanced, and proactive investing. The investment portfolio is managed on a segmented basis. The market-based core accounts for about 80% of the portfolio, and it's largely a high-quality, investment-grade fixed income portfolio, generated $2.4 billion of income last year. Market-based active is 11% of the portfolio. And our goal here is to really outperform the market through active management by leveraging our capabilities in public markets. For example, in December, the muni market got a little loose because everybody was rushing to issue bonds before the tax stuff because they didn't know what was going to happen. Spreads widened out a little bit.
We picked up some extra munis because, and we'll keep them for a couple of months, and then when the market tightens up, we'll sell them. The performance-based portfolio focuses on idiosyncratic return and is largely private equity and real estate holdings. That portfolio had great returns last year, generating $900 million in investment income. In 2017, this largely fixed income portfolio produced a 5.9% total return. Allstate Benefits had a compound annual growth rate about 8% since we acquired it 17 years ago and generates very attractive returns on capital. Allstate Life's products complement the property liability business, earn good returns, and broaden our customer relationships. Our services businesses, which is a new reportable segment that we're breaking out now, offers a broad range of products and services that support our customers and enhance our customer value propositions.
In the first year under Allstate ownership, SquareTrade had very strong performance with policy growth of 10 million and nearly 36%. In the fourth quarter, Arity established relationships with insurance and shared mobility companies, expanding its platform outside of the Allstate entities, which is Allstate and Esurance have Drivewise and DriveSense, respectively. Allstate represents an attractive investment opportunity because we have a competitively differentiated strategy, we generate attractive long-term returns from our broad-based business model, and we're focused on creating economic value for our shareholders. A return to profitable growth of the Allstate brand property and liability business and executing on the rest of our strategy will generate additional shareholder value. We're financially strong. We raised our quarterly dividend by 24% in the first quarter of 2018, and we repurchased over 26% of our shares outstanding in the last six years.
With that, Jay and Alison will do whatever you'd like to do.
Let me throw some questions. Alison and I will ask some questions, but if you have questions, raise your hand and we'll get you a mic as well. I think, Tom, one thing that's affecting the stock, based on my discussions with investors, is that in 2017, Allstate clearly was ahead of their competition in dealing with the higher frequency, really 2016 and 2017. The feeling was that would improve your competitive position because you had already dealt with the margin issues, and now you could pick up some share, and we saw that a little bit. The concern people have is now that others have dealt with their own margin issues, that everyone's going to become more competitive from a pricing standpoint. How would you respond to that thesis?
I don't think everybody's dealt with all their issues yet. When I look at the state-by-state profitability of some of our big competitors, they still have a ways to go. If you look at State Farm, we're in the lower left, we compete quite aggressively. If you look at our underwriting profit and their underwriting profit in auto insurance over the last five years, John, I think it's like a $20 billion difference. They got some room to go. If you look at Progressive, seems to have their act together. GEICO's got a mixed bag in terms of their profitability by state. Then there's a host of other people. I think it's like you're always competing with everybody. I think we're in a better position now going into 2018 than we were coming into 2017 to grow. I'm feeling good about it.
The other question I had was on the services business, which we appreciate the added disclosure. That's really helpful. Should we expect this segment or these businesses to be a continual drag on earnings? It loses money, which is fine, it's a new business, but should that continue for the foreseeable future?
We're running those businesses to grow them and make sure they're growing them economically, not for the P&L. I mean, the P&L, they lost, I think, $59 million last year, which was completely. We can fund that kind of stuff. It's really about what kind of value we're creating. If you look at some of those businesses, they're worth a lot more than if you value our company on either book value or P/E. If you did it on P/E, it'd be a negative. Yet we paid $1.4 billion for SquareTrade. Arity's got a huge position in telematics, which if it was a separate company, it'd probably be one of your unicorns. Allstate Roadside has an Uber-like fulfillment model where we're in 100 markets where we can get to somebody's car twice as fast as everybody else if it's not a tow.
We have some really good businesses, all of which are value, and we're like, "We got them, let's build them, let's grow them," and as long as we're creating value. Now you got to create economic value. With SquareTrade, for example, we're always making sure that the business we write today has good economics on it. Same thing with Esurance that we're growing. The services business, I think, will be a high return, high growth set of businesses for us.
Got it. The other question was on interest rates. Over the past five years, you guys have talked about the benefits of higher rates, and we didn't pay that much attention to it because rates were never going up.
Yeah.
Rates are going up now. Can you remind us again the leverage you have both in the property casualty business and in the life business to higher rates?
It's a really good question, and I think The Street has sort of forgotten about it. A couple things that people will say about Allstate is, "Jeez, why would I invest? You get peak auto insurance margins, like game's over. I should've got in at 50." The answer is, yeah, you should've got in at 50, but there's still a lot of room to go. If you look at last year's performance, our homeowners business, as I said, was down a huge number. We've been making over $1 billion out of the homeowners business for a long period of time. We just had high caps that actually offset some of the benefits we had. It was offset by the benefits of auto insurance, we would expect homeowners to offer opportunity.
From interest rates specifically, we have an $83 billion portfolio, property liability's about $42 billion, duration on the fixed income portion, that's 3.3 years. We've been short for four or five years because we just don't like to trade. We just didn't think it was right going long. Let's take $83 billion. You got some property liability, some life, fair amount of fixed income. If you took $40 billion, add 100 basis points on it, that's $320 million after tax, about $0.88 a share, and about over a point and a half ROE. We used to make a lot more money on investments. I've been here 23 years, we used to make a lot more money on the investment portfolio. It's there when we think rates are right. It's not going to happen in a day.
If rates went up by 100 basis points today, we wouldn't roll the whole portfolio. We are positioning ourselves for increased rates.
Any questions in the audience? I guess I want to touch on Arity. By the way, following this session, we've got a panel on telematics.
I'm going to stick my head in. I'm kind of curious.
Give us a sense of if we're here in five years at this conference, what do you want to be telling us about Arity at that point? What kind of products, leadership, revenue, anything?
Let me start first with our auto insurance business, and I'll do Arity if that's all right. In five years, I think most people that are sophisticated in auto insurance will be using telematics to price. It's much more accurate. It's every bit as powerful as credit was when we rolled that out 15 years ago. When we started rolling out back in 2002, 2003, we had win rates in some sales of 80%. We were knocking them dead, and we're all in on telematics at this point. Not everybody else is. There's a couple other people that are pushing hard on it, but we think it'll be the way you price insurance because it gives you a more accurate price. A more accurate price leads to less volatility, more growth, better retention, better customer value proposition. That's, I think, going to happen.
The Arity is really our way to participate and help that along, as well as participate in the growth from connected cars. Arity does all the work for our insurance companies, getting the device, doing the telecommunications, doing some of the math on the pricing. We built a rating service organization inside Arity that gives information to our insurance companies. What we said is, we've got a big enough base, we're going to go offer that to other insurance companies. We think we can do that, still maintain our competitive position in the Allstate brand, but we think that the network effect of getting other people's data together with ours will make them better and us better, and we think we can make money on it.
I would expect to see Arity start to turn a profit, not just on Allstate business, but on other people's business. It'll be a connected car company. We also think that it has the potential to get into what we believe will happen with the personal transportation industry. We think that's going to change dramatically. The math on that, the personal transportation industry, our cars and trucks, we got $4 trillion tied up in it right now. We spend $2 trillion a year using it. Direct cost, that's gas, maintenance, auto insurance, some interest on the cars and stuff like that. $1 trillion in indirect costs. Capacity utilization, 3%-4%, peak hours 34%, and only 30% of the time is there more than one person in the car.
If it was a manufacturing plant, you'd shut it down today, you'd redo it, you'd turn it into something new. It's got economically dispersed ownership, that makes it harder because everybody owns those $4 trillion worth of cars. That said, we think the economic benefits for households are so big that it's going to happen. A 20% improvement in that cost structure is a 5% improvement in household income in America, every household in America, $3,000 per household, $62,000 average household income. That's real money. People will go for it. We think it's going to change. Arity is designed not just to help insurance companies, but to help other people take advantage of that. Whether that's shared car, whether that's data, whether that's government data, whether that's managing transportation networks, we have a whole series of efforts. I don't know which will work, Jay.
Some will work, some will be just an idea, some will be a great idea and bad execution, others will be great ideas and great execution. Arity is our way to not just help our insurance businesses get better, but also ways to capture value from what's going on in connected cars.
Has the opportunity at Arity. Well, let me say it this way. Has Arity's mandate changed since you first decided to form the business?
Our mandate really started 2012. I think we might have created Arity, John, in 2014 or 2015. It was last year we decided we would open the stuff to third parties. It's really been about a year that we've had to hire marketing people and figure out what we're going to do and get data protection stuff. We like it. We're getting good appetite. In fact, in some ways, we've changed the nature of that industry. Some other telematics providers who do devices, they manage the OBD port or have some software you can put on your cell phone.
When we went into the rating service organization piece to actually provide customers not just that backbone and infrastructure, but actually data, and let them use their data to price better, you saw Octo bought a data company last year, I think because we're starting to change the basis on which we compete. We're excited about it. We think it's good potential. We're going to make a lot of money in auto insurance with telematics. This is extra return based on what we know how to do.
I want to just hit on autonomous vehicles. I've been one who's been poo-poohing this whole concept. I just thought it would take so long for autonomous vehicles to have any impact on driving, and accidents, and frequency, that it really wasn't worth talking too much about. I just sat down with our auto analyst last week, and he said it's happening faster than even he thought. My question to you is, are you seeing this happen quicker than you had expected a couple of years ago? If so, are you reacting to it?
Some cars are already autonomous, partially. It's sort of like what's autonomous? Does that mean it does nothing, or it's got the little beeper on the side of your mirror when you see somebody in that lane? It's assisted driving. Cars are continually getting better, and some ways they get better, there's fewer accidents. Some ways they get better, and it's more expensive. When that little mirror gets ripped off, now it's $1,000, and it used to be hundreds of dollars. As it relates to auto insurance, we think it'll take some time. When we first did our forecast on autonomous vehicles, and it's really the state of personal transportation, back in, I think it was 2013. We looked forward then and said, looks like we got a good 10-year run before that has any real big impact on auto insurance.
We just redid that last year, and it still looks like it's a way out. That said, all the things we thought would come true four years ago are coming true. We think it's opportunity there, but I don't think it's like we can do it to 240 million cars. Even if the cars were all ready today, nobody's got $4 trillion to replace the whole fleet. Even if you could, you had $4 trillion, what would you do with the other 240 million cars that are sitting around here? It'll take a while. Our point is, lean into it. Right? One of our offerings with Drivewise is upgrade the technology in your car. You don't need to buy an iPhone X to upgrade your car.
You can just put Drivewise in your car, you can put our OBD port into it, and your car is connected to the world. We're leaning into it as opposed to thinking like, "Oh my gosh, our business is going away." On the commercial piece, we are looking hard at what we do with some of the shared mobility companies on that, and I think you'll see us grow in that space.
Got it. Any other questions? Just wait for the mic. Right there.
Do you want to talk about the annuity business, the first time you disclose some numbers around it? Yeah. The annuity business is 2 types of annuities in it, right? We sold our VA business in 2006, didn't think the market was going to crash, glad we did it. We have two other annuities, all fixed. One are deferred fixed annuities, which are relatively short-term consumer liabilities, and the other are what we call structured settlements, which are payments you make to people over the course of their life if they are severely injured, by and large. That first piece we have pretty much matched off, and we get about a 11% return on equity in that chunk. The other piece is much longer duration, and we've invested in that long duration, much like you would a pension fund.
Most of our investments in that segment are equity, except we say we want to be cash matched for the next seven years. I don't want to ever have to worry about not having enough cash to pay people on a rolling basis for seven years in that portfolio. Every quarter we update it and make sure we're good for seven years. That said, the past seven years we've invested in equity because when you look at equity returns versus fixed income returns, of course, let's say you take a 10-year period, the return is twice and the risk is actually less. The risk is less because in fixed income, you have defaults, and you lose it. In equities, it can come back in total, and over time goes up. That's a good thing. That's how we've invested.
The bad thing is that makes our return in that segment about 4% return on equity, because there's huge capital charges with investing in equity, which is not right. It ought to be changed. That said, we're going to invest for shareholder return, and accept that lower return. We've looked at should we just dump it, and take the write-off, and I'm like, well anybody can write their way off to a higher ROE. Real management teams work hard and create value by getting to a higher ROE. We've decided to invest in the right way in this and accept that.
We're also trying to work to try to get the NAIC and other people to change their rating standards because what they're doing is actually encouraging many people who are investing in fixed income for 30-year liabilities, which as you all know, is not a good idea in today's interest rate environment as Dave was talking about.
Any last-minute questions? All right. Join me in thanking the folks from Allstate. Thanks, guys.