We're very pleased to have with us today Tom Wilson, Chairman and CEO of The Allstate Corporation. Allstate is one of the largest insurers of vehicles, homes, and personal devices in the U.S., and also has a life insurance and retirement savings business. Tom will begin with opening remarks, and then I'll moderate a Q&A session. With that, it's my pleasure to turn it over to Tom Wilson.
Thank you, Jay. Good morning, everybody. Thank you for taking time to invest to learn more about why Allstate is a good opportunity for you. Joining me today is John Griek, who leads our investor relations team. Before we begin, I'll give you the obligatory, I'm going to make some forward-looking statements today. In reference to some non-GAAP measures, this presentation, and then more specific information about all of our business, you can go to our website at www.allstateinvestors.com. Let's begin with the top of the news. Hurricane Florence is currently expected to hit portions of North and South Carolina, probably maybe some of Virginia as well, with sustained winds of over 135 miles per hour, which would make it a Category 4 storm. The first thing is to know is that we are prepared both operationally and financially.
Our first priority is make sure we take care of our customers. We have an exceptional track record of being early to the scene, reaching out to our customers, helping them rebuild their lives as quickly as possible. We have 700 people pre-positioned at three different cities right now. We have the technology available at three different levels of aerial technology. We'll be there delivering on the Good Hands promise. We're also prepared for this event financially. We shift an extensive amount of this risk to reinsurance markets, which reduces our capital requirements and protects annual earnings. The current reinsurance program covers catastrophe losses from homeowners and auto insurance across all three of our underwritten brands, providing $4 billion of limits above a $500 million retention per event. Now let's move on to a broader discussion of Allstate's strategy to create shareholder value.
Our strategy has two parts. One is to grow market share in the property liability businesses, and the other is to expand our other protection businesses. The property liability market has four consumer segments, and we serve each of those with a unique and competitively differentiated product offering. Allstate agencies provide customers with personal guidance through 36,000 professionals located in 10,400 agencies in virtually every community in America. Esurance, on the other hand, provides over 1.5 million policies to customers who prefer to purchase their products through the web or through a call center. We use sophisticated analytics across all those businesses to ensure we grow profitably and are in the forefront of using telematics-based offerings. We're building an integrated digital enterprise that uses data, analytics, technology, and process redesign, and that process redesign piece is really important to improve our effectiveness and lower costs.
For example, we're the leader in using customer-generated photos to settle auto insurance claims. Our strategy also includes expanding other protection businesses. We do that by leveraging our brand, customer base, investment expertise, and capital. This began in 1957 with life insurance. In 1999, we acquired American Heritage Life, which was rebranded Allstate Benefits and provides protection products such as life and disability insurance at the work site. By leveraging our resources, this business has had a compound annual growth of 7% for 18 years running, with 4.2 million policies in force, now four times the size it was when we bought it. We purchased SquareTrade at the beginning of 2017, and it is growing rapidly. This year, we began offering insurance to transportation network companies such as Uber through Allstate Business Insurance. We recently announced a potential expansion in personal identity protection through the acquisition of InfoArmor.
InfoArmor fits well with our focus on protection, and its products are distributed through voluntary workplace programs. That fits very well. We'll leverage the Allstate Benefits distribution and the Allstate brand. This two-part strategy creates shareholder value through customer satisfaction, unit growth, and attractive returns on capital. It also ensures we have sustainable profitability and a diversified business platform. Making sure our largest businesses generate attractive returns is key to performance. Allstate has a long track record of generating attractive returns from auto and homeowners insurance. The graph on the top left shows the auto insurance combined ratio for the past 10 years in comparison to our three largest competitors in the overall industry. The three lines on the bottom show that Allstate's auto insurance margins are amongst the best in the industry and substantially better than our largest competitor, State Farm.
Now, results obviously vary from year to year. As you can see in 2015 and 2016, when combined ratios increased due to higher frequency of auto accidents. Allstate's results since then reflect an early and aggressive focus on improving profit margins. This did have a negative impact on policies in force, but today policies in force are increasing. The graph on the top right shows homeowners insurance results and highlights the significant higher profitability we've generated on this risk. In 2008, we began repositioning the homeowners business to reflect an increase in severe weather. This included raising prices to ensure rates were adequate, re-underwriting the existing book, creating a new homeowners product, and increasing our reinsurance coverage. At that time, those actions led to large decreases in number of policies in force.
That business now has a very attractive combined ratio, which is substantially better than the industry, as you can see from the blue line. Over the last 12 months, premiums were $6.9 billion and underwriting income was $932 million. Today, Allstate brand homeowners insurance is a very attractive business. We're also enhancing customer value propositions through increased connectivity through telematics-based insurance products. We introduced telematics in 2011, and we're the first in the industry to roll out a mobile phone version. Allstate's Drivewise and Esurance's DriveSense products are expected to give us a substantial competitive advantage by providing a more accurate individual pricing and enhancing the driving experience. Drivewise is currently available in 48 states, and DriveSense is in 32 states. We have over 1.3 million customer connections, and our data set is growing rapidly.
We have over 38 billion miles worth of Allstate customer data, which is growing by 1 billion miles per month. Arity, which is our captive telematics company that provides the services and the technology to get into telematics, is also expanding through third parties such as Life360, which has exponentially increased our access to driving data. We continue to innovate around connectivity and introduce Milewise, which is a usage-based insurance offering where customers pay a daily rate plus a per mile charge. If you live in New Jersey, and you take the ferry or train to the city, you've undoubtedly been frustrated by paying so much for insurance while you're working, and have likely seen our ads. Milewise is in three states, with several more planned for implementation over the next 12 months.
We're also using new advertising for the Allstate brand to highlight the benefits of Drivewise. We've launched a new advertising with Dennis Haysbert sitting in the middle of an intersection in a chair. If you see it and you like it, call us, we can help you. When auto accident frequency increased, Allstate brand policies, in force, decreased due to the profit improvement actions I just talked about. We did that so that we could improve the combined ratio. Policies, in force, are now growing again due to increased customer retention and higher new business levels. Allstate has additional growth opportunities beyond the Allstate branded property liability businesses. The Esurance brand has doubled in size since it was acquired with compounding annual premium growth at 12%.
This trend continued in the second quarter of 2018 with net written premium growth at 12.5% compared to that prior year, which reflected increases in average premium and a 4% increase in total policies of force. Policy growth at Esurance has also benefited from not just auto insurance, we've been expanding into homeowners insurance as well. Esurance recently launched an ad campaign targeted towards those self-serve customers who prefer a branded insurance product under the theme Simply Painless. The ads highlight ease, simplicity and affordability and use Dennis Quaid as our spokesperson. We also have significant growth opportunities in other businesses like Allstate Benefits and SquareTrade, which sometimes get overshadowed by just the sheer size and scale of the property liability businesses. Nevertheless, they create substantial shareholder value. Allstate Benefits premium and contract charges have grown for 18 consecutive years and exceed $1.1 billion.
SquareTrade revenues of $122 million were $52 million above the prior year in the second quarter of 2018, with half the increase driven by organic revenue growth and the remainder due to the adoption of a new accounting standard. SquareTrade continues to make progress on the three acquisition objectives we set, growing the U.S. retail business, improving returns, and creating growth beyond U.S. retail. The potential acquisition of InfoArmor will further expand customer protection offerings in the identity protection market. InfoArmor provides a full suite of protection services that includes monitoring, alerts, remediation, and then a proprietary customer identity health score. Their products are primarily distributed through employee benefit brokers and benefit administration platforms to 1 million employees and their families at 1,400 companies. They have a strong competitive position in what is a low-cost distribution model.
As part of Allstate, InfoArmor will be able to accelerate its growth by leveraging Allstate Benefits’ relationships with brokers, companies, and customers. The acquisition expected to be accretive to net income, and neutral to return on common equity by year two. The acquisition price of $525 million will be funded with wholly company cash, and the transaction is expected to close in the fourth quarter. There is no impact on our share repurchase program. A disciplined approach to capital management balances investments in growing while simultaneously returning cash to shareholders. There is an excellent track record of returning cash to shareholders. The dividend yield has remained steady as dividends per share increased. We have also repurchased between 4% and 10% of outstanding shares each of the past five years. Return on common shareholders’ equity was 17% for the 12 months ended June 30th, and book value per share increased by nearly 10%.
Over the last five years, book value per share has increased at over 6% annually. The result is total shareholder returns of 78% over the last three years, which exceeds our peers and the overall market. While there are many ways to value a company, Allstate is positioned to create additional shareholder value. Using a regression analysis of returns and book value multiples, Allstate currently trades at a 20% $20 discount relative to our peers. Allstate’s competitively differentiated strategy, broad business model, operational and investment capabilities, and capital management will enable us to create value and win far into the future. With that, Jay, we will open for questions.
Tom, thanks very much for that overview. It was really helpful, especially around the potential Florence exposure. If I did my quick takeaways on Florence, based on Allstate’s substantial reinsurance protection, it looks like a $500 million pre-tax retention, and then 95% protection around the losses. Essentially, each $1 billion of gross losses for Allstate is around $50 million.
After the $500 million.
Correct.
Yeah.
Yeah. With a $4 billion per event coverage base and a reinstatement.
Yes.
When we think about even something with the magnitude and breadth of potential damage coming out of Florence, it seems like something would be less than a quarter's worth of earnings.
One, of course, you never really know, and so I don't want to predict what Florence will do to us. Yes. You're correct, and what we've tried to do is set up the reinsurance program to basically divest ourselves from the severe catastrophe risk. The reinsurance markets, as you and I were talking about earlier, are robust. A lot of capital moved into it. People view it as alternative asset classes. We take advantage of that to shed that large catastrophe risk while still taking care of our customers.
Excellent. I think that puts you in a great position. Talking about the broader business, Allstate is among the largest auto insurers in the U.S. The Allstate brand auto business has shown a sharp recovery in profitability and policies in force are growing again. Can you discuss the prospects for further margin improvement and top-line growth in that business?
Yes. The auto insurance business on a long-term basis has modest growth to it. First, it would be how many cars are on the street. In North America, it's not a really high growth in terms of it being low single digits, 1%-2%. Then of course, it's how much does it cost to fix them, and how much does it cost to fix the stuff that those cars bash into. We call that severity. That has gone up. That's up a little more than inflation for a couple of reasons. One is the cost to rebuild a car has gotten a lot more expensive because of all the sensors in a car. We've just done some math on the cost to buy a car hasn't gone up that much in the last 10 years, but the cost to rebuild a car has about doubled.
When your car gets wrecked, then you got to pay for that, and that goes into insurance premiums, which raises the average premium. The same is true for medical costs. The other factor is what happens to frequency. Frequency was coming down for a long period of time, then it kind of flattened out. Came down for about 10 years, flattened out about five, six years. In 2015 and 2016, went up. Last year and this year so far, it's gone down again. A little hard to predict. Not a little. It's very hard, impossible to predict frequency. I would say on a long-term basis, I would expect frequency to come down. What you have is a business that's got relatively modest industry growth for those reasons. The question is, how do you win in that?
How do you continue to take shares since we don't have 100% of the market? We have a lot of competitors who aren't as good as we are at making money. It's about telematics. Telematics pricing is every bit as powerful as credit. That will be the way insurance is priced 10 years from now. Secondly, it's about pricing sophistication. We're good in both. We have more work to do in both. The way you grow and you do better is don't keep doing what you're doing. Just the use of data analytics in resolving the claims is worth a tremendous amount of money. Whether it's QuickFoto claims or anything else we're doing, it helps you reduce the variability of what you pay for a claim. You always want to pay the right amount, not too much, not too little.
It's really about reducing the spread on those claims. Each of our businesses has different strategies. The Allstate brand is working hard to improve the effectiveness and efficiency of its distribution. We call that a trusted advisor initiative. Esurance is expanding both in certain states, putting more growth in states, and expanding into homeowners. Encompass really needs to expand geographically, whereas the Allstate brand is everywhere, and it needs greater density in some of the middle parts of the country. Encompass is just not in some parts of the country, they have to geographically expand. Then we're hard at work on continuing to reinvent the product. Insurance is not the simplest, most consumer-intuitive product out there. We think there's great white space there.
Just like we did with Your Choice Auto 10 or 12 years ago, we're now working on redoing our products in a more consumer-friendly way.
Very helpful. If we think about market share dynamics within personal auto, it appears that GEICO and Progressive have been generating faster auto policy in force growth than the industry average. Do you expect this market share shift to persist?
Well, Progressive clearly has had good results right now. Their growth and their profit margins look good. They clearly have good pricing technology going, so they seem to be having a good run. I think they'll continue to do that. Now, if I traded at 4 times book or something. If you look at what GEICO has done, GEICO's combined ratio has gone up. If you go back to that chart. They've decided to keep growing and not make as much money in auto insurance and make it on investments. I think you'll see both of those are strong competitors. They're like us. They're down in that bottom. They know how to run their business well.
The question is, as we start to grow, will we all keep taking share away from those people who are up in that industry average where they're just not that efficient and effective? I believe that'll be the case.
That makes sense. Price increases in auto insurance have been substantial, although State Farm-
I'm going to just move this past, I don't want to just-
Okay
keep you guys to death on the discount.
It's like a pie chart. Price increases in auto insurance have been substantial, although State Farm recently announced potentially nationwide rate cuts. Can the industry keep raising prices?
Well, State Farm went down 2%-3% in, I think, 23 different located states. What you've seen is a rapid increase in auto insurance prices over the last couple of years because of the increased frequency that I mentioned in 2015 and 2016. On top of that, the increased cost of settling claims. Auto insurance CPI, I think, is down about 7% right now over the last 12 months. It was 9% last year. I don't think it will keep going up at that level, in part because the frequency has come down. Ours won't be up anywhere close to that because we took our rate increase. I think, though, the market's pretty rational. To the extent costs go up, we ought to be able to raise our prices.
We went down a little bit in growth when we raised our prices 6% on average over two years when nobody else was. That was intentional on our part. We didn't add as many agencies. We cut back advertising. We did it on purpose. I think there's plenty of capacity in the market to raise prices to cover costs. Yeah, I don't see that.
That makes sense. Okay. Turning to margins, underwriting margins, Allstate's target for 2018, now an underlying property liability combined ratio of 85%-87%, which improved relative to the initial guidance. What's your comfort level for achieving or perhaps exceeding that target?
Well, we wouldn't say it if I didn't think we could do it. We've never missed it.
Right. 10 years in a row?
I think so, yeah. To be clear about what we do there is the underlying combined ratio doesn't include things like catastrophes and prior year reserve releases, which do impact profit. We give it to our shareholders and potential investors as a way to say, look, this is what we think we can deliver this year. There's a range on it because frequency bounces up and down. Frequency is the known unknown. Nobody can determine what frequency of auto accidents is on an annual basis or quarterly basis. The reason we just lowered our combined ratio guidance, aka increased our margin projection for this year, was because frequency was down. We had not predicted auto frequency was down this year. It is down this year. It's hard to determine exactly why. Therefore you can't predict it. I'm really confident we'll be there, yeah.
That's great. Okay. You mentioned catastrophe impacts. One of the things that's blooming now, at least currently, is another pretty tough California wildfire season. Perhaps not as bad as a year ago, but any initial thoughts there in terms of what industry or Allstate's impact could be?
The Carr Fire was the big one out in California. The Carr Fire was included in our July monthly cat release, where it's $220 million, I think, was the July number for just what happened in July. Of course, that develops over time. You find you have to spend more or less than what you thought you were going to do when you first made an estimate of it. We're comfortable with that number. In terms of just wildfire exposure in total, we've gotten a lot smaller in homeowners in California over the last seven or eight years, in part because we went to customers and said, "Look, if you don't have bushes pretty far from your house, if you have bushes close to your house, trees close to your house, your house is going to burn down.
We don't want your house to burn down, we suggest you cut your bushes back." Some people like that and listen to us. Other people don't. We say, "Well, if you don't cut them back, we don't want to insure you." We view that actually in an odd way as a way of helping our customers. We're willing to give up the business to have them know how serious we are about it because, for the most part, you could write the business, and your odds of losing money on it aren't that great in any individual state. When you look at catastrophe management, you have to look at the risk you cover, the concentration of risk. We manage that in California by individual house as well as risk concentration in various areas. The fires don't burn the whole state.
They just burn parts of it. We manage that wildfire risk pretty aggressively and feel good about where we're at.
That's great. Okay. Big picture topic. Can you discuss how Allstate is effectively employing technology, putting data and analytics to distinguish itself from competitors?
Sure. First, there's just a lot going on in data and analytics in our space. To set the stage, we have about 1,700 people who are just, part of their time is dedicated data and analytics, everywhere from embedded in the business everywhere, to we have a centralized group of 250, 280 people now. We have them located in the United States, in Northern Ireland, in India. We've got them everywhere. It's a big effort for us. We have a comprehensive strategy on it, everywhere from how do you collect, make sure you have the right data, and have it governed properly, what kind of tools do you use it, how do you have people do it, how do you redo your processes, how do you use continuous improvement to embed it? It's a very high priority and comprehensive enterprise project for us.
It would be, let's take Florence. Florence, right now, as the weather clears, we'll be using three levels of aerial data collection to start to analyze claims. We'll use satellites, fixed-wing aircraft, and drones. We'll use that information, collect that information, use that data to then have people sit in locations, centralized locations, to adjust claims. Sometimes we have to have the drone driven out to the house, and the person flies a drone over the house. That information, then we start accumulating that information, and we can use that to better adjust claims. We do it obviously in pricing.
In our telematics pricing with data and analytics, we can tell if you're driving, and we can tell which hand your cell phone's in, there's a whole bunch of stuff we can tell about your driving behavior. We use data and analytics in telematics to tell whether your hard braking is hard enough to likely cause an accident, or it's hard braking that's not likely to cause an accident. While you're getting all this data driving information in and we're tracking you, Jay, and you say, "Well, that hard braking was because some jerk cut me off." You're like, "Yeah, but had you not been going 80 miles an hour, maybe you wouldn't have had that problem." We can put that data together, and build a risk score for you as an individual.
Data and analytics is ripping through certainly an insurance business and service business, doing everything from making us smarter, more effective, to taking our costs out.
Sounds very important near-term and long-term. Going back to some of the other Allstate businesses, the smaller auto insurance businesses of Allstate, which are Encompass, focuses on sales through independent agents, and Esurance, which is a direct writer. Can you talk about plans to improve profitability in both directions?
Sure. They're both different, right? Let's start with Esurance. That's the bigger business. We bought that business because we wanted a customer value proposition to go head to head with GEICO and Progressive Direct. For example, the customer value proposition is tools that get me smart. When you get on the Allstate website, it takes you right to tools. You can tell what other people buy. If you get on an Allstate website, it takes you to an agency with a picture of a person, because that's what you want to buy. We went at that market aggressively with Esurance. We've doubled it in size. We backed off a little bit in 2015 and 2016 because we had the same trend on frequency in Esurance that we did in the Allstate brand. That said, we're back to growth again.
What I would say is during that entire period we've owned it, all the business we've written as a cohort by year has been above the cost of capital. Not as high a return as we get in the Allstate channel, but above our cost of capital. We won't write it if we think we're going to lose money on it. We don't believe in growing to lose. Profitability, though, in that channel is different because as you know, you spend a bunch of money advertising up front, and then you don't have to spend expense money each and every year after that. Your profitability, you lose money in the first year. We manage it economically. We've now got the profitability down where the combined ratio is around 100 ±.
Combined ratio first six months of the year is down, I think, four and change end points. I kind of like it around that 100 level. I want to grow as fast as we can at that level. As long as the advertising and the close rates are good, we'll drive that business. Profitability and economics will look different there. Encompass is a different cup of tea. Encompass, we did not have the pricing sophistication we should have had. We sort of missed a turn back about four years ago, maybe five years ago in the independent agency business. We've been working to catch up there. The profitability in the auto insurance business has gotten better. It's about where it should be. The homeowners business is still not where I'd like it to be.
It's profitable, it's not as economic as I'd like it to be because-
That's for the
For Encompass.
Encompass.
Yeah. Encompass homeowners. When you look at the Encompass homeowners business, it is profitable. The reason it's profitable is because the catastrophes just haven't gotten to wherever they've written their business yet because they're not everywhere in the country. We still need to improve the profitability of the homeowners business. The bigger challenge for us in Encompass is how to build a sustainable and profitable growth. That means we have to still get the pricing sophistication done, and we're working on redoing the distribution capability. We still have some work to do. I would say that's the, of the three property liability businesses heading towards growth Allstate is the biggest and will drive the most policy growth and has very sustainable ability to sustain growth.
Esurance, smaller, higher growth, will grow both because it can advertise more and has a different cost advantage, and because it can get into homeowners. Encompass is the third one in the line in terms of growth. It'll take longer for Encompass to grow.
Okay. Allstate's most recent bolt-on acquisition, or not most recent, but one of its most important was of SquareTrade, which expanded the company into insuring personal electronic devices. How would you characterize the performance of that business since acquiring it? Maybe that could lead into a discussion about Allstate's appetite for further acquisitions, both large scale and bolt-on.
Okay. Good question, because people are asking about allowing you to do an InfoArmor, how does that relate? SquareTrade, we wanted to be in the cellphone and personal device insurance business. We had missed that turn in 2007, 2008. We were a little preoccupied with the financial crisis and $120 billion of assets at the time. We wanted to get in the business. SquareTrade enabled us to get in the business using different distributions because they sell at retail. You go to Costco, Target, now Walmart, and you buy an Allstate Protection Plan. Walmart is an account we just launched with SquareTrade in August. We're really excited about that business. I really like what it does. It does claims differently and better than other people in the industry. It's very customer focused. We like that business a lot.
When we look at acquisitions, we start with, okay, do we know anything about this business? If you go into auto insurance or homeowners insurance, we know a lot about it. When we looked at the electronic device market, we thought we don't know that much about it. Is there a way we could buy somebody that did? Yes. Does it leverage our skills and capabilities? The answer to that case was yes. Can it be economical and get good growth? There are times when we look at that and can't do it. Telematics, we've been a leader in telematics. We've looked at just about every telematics company that's out there and been on the market, we decide to make it ourselves. We decide, you know what, we're as good, smart, we got good data analytics.
Some of these startups, it seems like the pricing has gone to about, it's like $10 million per person. If they have five people, it's like a $50 million fintech company. We don't need to do that. We can make it ourselves. When it comes to InfoArmor, we're already in the identity protection business. We have about 250,000 policies out there. We looked at some of the large players in the market who sell on a direct basis. They advertise. Like a magazine model, advertise, and then do the subscription. We didn't think that was economic. We thought the acquisition costs were too high. We decided to make our own. For the last two years, we've been working on the digital safety product, which is different, it's consumer focused, and it's different than what's available in the market.
InfoArmor came along, we said, you know what? We can buy this company. We can make it. We're a better owner because we can help them grow through our Allstate Benefits distribution. Our brand will help them. We can help on a variety of other things. They can take the product we've been working on and really accelerate it, bring it out to market. It's a combination. We're always looking. We don't feel like we have to buy stuff, but if we think it adds to it and it's economic, and we've then set priorities for. When we close on InfoArmor, that's assuming we close on InfoArmor, and I have no reason to think we won't, then we'll say, here are the three things we're trying to do with InfoArmor, you can judge whether we're doing a good job or not.
Helpful. All right, let's go to the audience response questions. I think the audience knows the process for this. We ask a question, and you have 10 seconds to key in. If you, the audience, currently don't own the shares of Allstate or happen to be underweight, what would cause you to change your mind? Start the clock, please. It's faster auto policies in force, greater improvement in the underlying combined ratio, perhaps partial or full exit from Allstate Financial, lower valuation, or none of these. The audience view on these. Bit of a bell curve here. Around 30% each either saying maybe scaling back or exiting from Allstate Financial or a lower valuation with less people focused on improvement in the underlying combined ratio or growth . What do you make of that?
Well, everybody always likes to buy stuff when it's cheap. I'm not too shocked by that. What I would say is our stock's lower now than it was at the beginning of the year when we've bought back more than 4% of the stock. Just on a heads-up basis, it ought to be up by 4%. We're making more money because we had a tax cut from $500 million. I think people always want to look at it on a how much is it run up lately. I think we struggle a little bit with the, "Geez guys, had a great run over the last three years. Did I get in too late?" I'm like, it's always point forward. You all know that. On the Allstate Financial business, let me describe that business. We have four businesses there.
One is a really great business, two are good businesses, and one's not so good. The one that's really great, Allstate Benefits, I talked about that already. The two that are okay are life insurance business returns about 12%, modest growth, good diversification, leverages our investment capabilities. It's good business. Our customers like buying life insurance from us. We sell only through the Allstate agency channel, downsize that business substantially. When we downsized Allstate Financial, we got out of the annuity businesses. We still provide annuities to our customers, we just don't sell our stuff. We decided we could still serve our customers, we didn't have to deploy our capital. Two separate conversations. We still serve our customers, all that stuff, but we decided not to deploy our capital in annuities. Variable annuity business we got out of in 2006.
That was near down right before the crisis. Not that we predicted, it was just pure luck. We got out of the fixed annuity business kind of beginning in 2008 to 2011 to 2012. We have two annuity businesses which are closed blocks. One is a decent business, one is not such a good business. The decent business is a single premium deferred annuity retail business. It's about $7.5 billion of assets. We earn about an 11% return on that business. It's pretty easy to manage it, not very complicated. The business that got a lower return is an immediate annuity business, which are structured settlements, people who have spinal cord injuries, and they get an annuity for life, or they have payments for it. These are really long-dated liabilities. The returns on that business are about 3%.
They would be higher if we didn't run our business economically, though. Let me be very clear about that. With that kind of business, it's about $11.5 billion, John, I think, of liabilities. Some of that stuff's 20, 30 years out, 40 years out in terms of payments. Once you get past about seven years, you should invest like a pension fund. I just think about it that way. Like anything past seven years, the right economic answer is invest in a pension fund because the risk on equities past 10 years is lower than the risk on bonds, and the return is twice. That's what you should do. Unfortunately, the regulatory scheme puts a huge penalty on you for investing economically like that. We chose to ignore that. We put the money up. We do what's right for our shareholders. I'll accept a 3% return.
If I could find a way to reduce the drag on that of the overall company without giving up the return, I would do that. We've been working on it for about five or six years. We've repositioned the business now. We've got different operating processes. We've lowered our costs, and we've got the investment portfolio mostly in performance-based investments on the long-dated stuff. That one, I would agree with, some partial exit would be good for shareholders. I'm not going to do it if we lose money, if it's going to be uneconomic. You can't sell a 3% business and not lose money. I'm not going to do it if it's not economic, just won't.
Excellent. Unfortunately, we're out of time. I'm afraid we'll need to stop there. Please join me in thanking Tom Wilson from Allstate. Great job, Tom.