Good day, ladies and gentlemen, and welcome to the Allstate fourth quarter 2016 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 then zero on your touchtone telephone. As a reminder, today's program is being recorded. I would now like to introduce your host for today's program, Mr. John Griek. Please go ahead, sir.
Well, thank you, Jonathan. Good morning and welcome, everyone, to Allstate's fourth quarter 2016 earnings conference call. After prepared remarks by our Chairman and CEO, Tom Wilson, Chief Financial Officer, Steve Shebik, and me, we will have a question and answer session. Also here are Matt Winter, our President, Don Civgin, the President of Emerging Businesses, Mary Jane Fortin, President of Allstate Financial, and Sam Pilch, our Corporate Controller. In December, we announced that John Dugenske will be joining Allstate as Chief Investment Officer in early 2017. John will join the team next month and will be part of our quarterly earnings calls beginning next quarter. Yesterday, following the close of the market, we issued our news release, an investor supplement, and posted the results presentation we will discuss this morning. These documents are available on our website at allstateinvestors.com.
We plan to file our 2016 Form 10-K later this month. As noted on the first slide, our discussion today will contain forward-looking statements about Allstate's operations. Allstate's results may differ materially from these statements, so please refer to our 10-K for 2015, the slides, and our most recent news release for information on potential risks. Also, this discussion will contain some non-GAAP measures for which there are reconciliations in our news release and our investor supplement. We are recording this call, and a replay will be available following its conclusion. I'll be available to answer any follow-up questions you may have after the call. Now I'll turn it over to Tom.
Good morning. Thank you for investing your time to keep up on our progress at Allstate. Let's start on slide two. We delivered excellent results for the year and finished 2016 with another strong quarter as we continue to effectively execute our short-term plans and build on long-term strategies. As a result, Allstate is well-positioned for continued success. Auto profit improvement plans over the last two years have enabled us to begin focusing on growth in 2017 while being able to react to further auto loss cost increases. The homeowners business continues to generate attractive returns despite higher catastrophe losses. Investment results for the year were good but bounced around from quarter to quarter, reflecting volatile external conditions. At the same time, we're investing in growth, both with existing businesses and new opportunities. Net income was $811 million for the quarter and $1.76 billion for the year.
Operating income was $2.17 per share for the fourth quarter, and $4.87 per share for the year. The recorded combined ratio for the year was a touch over 96%, and the underlying combined ratio is at the favorable end of the range we provided shareholders a year ago. Shareholders received $1.8 billion in cash through a combination of dividends and share repurchases. We also welcome SquareTrade into the fold, providing shareholders another opportunity for profitable growth. Going to the box at the bottom, total revenues of $9.3 billion for the fourth quarter reflected 2.8% increase in property liability insurance premiums, driven by the continued implementation of our auto profit improvement plan and higher performance-based investment income. Net income for the fourth quarter was $811 million, and operating income was $807 million.
Operating income benefited from favorable underlying loss performance in both auto and homeowners insurance, lower catastrophe losses, favorable prior year reserve releases, and strong investment income. The property liability insurance business performed well as a result of the successful execution of the auto profit improvement plan across the three underwritten brands and continued excellent performance in the Allstate brand homeowners and other personal lines insurance. Allstate Financial had a strong quarter with $130 million of operating income as the annuity businesses benefited from the very high investment income from the performance-based portfolio. Moving over to the full year columns at the right, the 12-month operating return on equity was 10.4%, and that's down slightly from the prior year, largely reflecting higher catastrophe losses in 2016.
Consolidated policies in force declined modestly over the year as strong growth at Allstate Benefits was more than offset by a decrease in the property and liability business. Looking forward to 2017 on slide three, we're in a position to achieve an improved underlying combined ratio and achieve our five operating priorities, both of which have both short-term goals and long-term objectives. We expect to build on last year's insurance margin improvement, resulting in an annual underlying combined ratio of between 87% and 89% in 2017. That range is comprised of a number of key assumptions. First, there will be continued improvement in auto insurance profitability as increases in average premiums and filed rates give us the flexibility to deal with increases in the frequency and severity of auto accidents.
Secondly, we assume the homeowners' underlying combined ratio will deteriorate slightly from 2016 but will still be well within our target range of profitability. Encompass and Esurance are assumed to stay on track to improve auto profitability. At the same time, we'll continue to invest in growth across the company. Our priorities for 2017 remain largely consistent with 2016. The first three priorities, better serve our customers, achieve target economic returns on capital, and grow the customer base, are intertwined and ensure the corporation has multiple paths to profitable long-term growth. In 2017, we expect to grow our customer base through continued policy growth in Allstate Benefits and Esurance, rapid growth in our newly acquired consumer product protection plan business, SquareTrade, and by reducing the number of policy losses under the Allstate and Encompass brands.
As you know, we proactively manage the $82 billion investment portfolio to achieve the best risk-adjusted return over time. Net investment income is relatively stable with the large fixed income portfolio. Additional income comes from the performance-based investment. The total return on the portfolio will be largely dependent on U.S. interest rates and economic growth. Our fifth priority continues to focus on building long-term growth platforms. The Allstate Agency Platform is being strengthened by rolling out the Trusted Advisor initiatives. Esurance will continue to expand in auto and homeowners insurance. Allstate Benefits will continue to leverage position in the high-growth voluntary benefits market. Arity will build out the telematics business, and SquareTrade will continue to gain new retail partners. Let's go to slide four to cover property and liability results.
Net written premium grew by 2.3% in the fourth quarter as average premium increases were partly offset by a 2.8% decline in policies in force. Fourth quarter catastrophe losses of $303 million were 15.4% lower than the prior year quarter. Catastrophe losses for the year, however, were nearly $2.6 billion, which was $853 million higher than 2015. The reported combined ratio for property and liability was 89.9 in the fourth quarter of 2016. When we exclude catastrophes and prior reserve re-estimates, the underlying combined ratio for the fourth quarter is 87.7, bringing the full year to 87.9. The four customer segments of the property and liability market are shown on the diagram at the bottom of the page. As you know, Allstate is the only company that provides a differentiated value proposition to each of these customer segments.
The Allstate brand, which is in the lower left, comprises 90% of premiums written, serves customers who prefer a branded product and value local relationships. Underlying margin improvement throughout the year in the Allstate brand was driven by the progress made in the auto insurance business. Homeowners insurance and other personal lines continued to have strong profitability. Esurance, in the lower right, serves customers who prefer a branded product but are comfortable handling their own insurance needs. The underlying combined ratio for auto insurance improved for the year but is above our long-term target, with the fourth quarter being somewhat elevated. The homeowners business is growing rapidly. The loss ratio is within expectation, but the underlying profitability was negatively impacted by startup advertising costs. Encompass, in the upper right, competes for customers who want local advice, are less concerned about brand experience, and are served by independent agencies.
We remain focused on improving returns and executing our profit improvement plan in this brand. Answer Financial in the upper right serves brand neutral self-serve customers and is an aggregator that does not underwrite insurance risk. We'll cover the results of these three underwritten brands in more detail on the subsequent slides. Let me turn it over to John.
Let's go to slide five to cover the results for Allstate brand auto. The recorded combined ratio for the fourth quarter was 95.3, which was 3.3 points below the prior year quarter and benefited from prior year reserve re-estimates. The quarterly underlying loss ratios and combined ratios are shown on the chart on the top left. The underlying combined ratio of 96.1 in the fourth quarter of 2016 improved by 1.5 points compared to the fourth quarter of 2015, driven by a 2.8-point improvement in the underlying loss ratio. This brought our full-year underlying combined ratio to 96.4, nearly a full point below year-end 2015 results. The chart on the top right shows the drivers of the improvement in our underlying combined ratio. Annualized average premium increased to $978, or 7.1%, while underlying loss and expenses increased by 5.5%.
This resulted in a favorable gap of $38 comparable to the third quarter. In the fourth quarter, loss trends benefited from favorable current accident year reserve development worth 0.7 points. To provide additional insight into our loss trends by coverage, the bottom half of the page shows the paid severity and frequency trends for both property damage and bodily injury. Property damage paid frequency, shown by the blue bar, continues to show an improving trend. After experiencing elevated levels in the second half of 2014 through the first quarter of 2016, results in the second and third quarter of 2016 were essentially flat. The fourth quarter experienced a decline of 1.2%. Property damage paid severity increased by a modest 1.9% in the quarter, and the rate of increase has come down throughout 2016. Bodily injury paid frequency decreased 19.2% in the fourth quarter, while paid severity increased by 18.8%.
These results are consistent with the trends experienced in the third quarter. Frequency and severity should be looked at in combination to get a sense of the true underlying loss trend. As we discussed in the third quarter call, bodily injury trends reflect payment mix and claim closure patterns that were impacted by changes in claims processes in the second half of 2016. These changes involve requiring enhanced documentation of injuries and related medical treatment and resulted in a reduction in the mix of smaller dollar claims paid. The increase in severity of larger dollar claims are more consistent with medical inflation, and we continue to be comfortable with our bodily injury incurred severity trends. Slide six provides detail on premium and policy growth for Allstate brand auto. The chart on the top left highlights our premium trends.
Average premium, as shown by the blue bar, increased 7% compared to the prior year quarter, a slight deceleration compared to the upward trend beginning in the second quarter of 2015. Average net earned premium, shown by the gray line, was up 6.8% and continued to increase, reflecting the lag between written and earned premium recognition. Since the end of 2014, auto average premium has increased by over 11% in total. We've continued to get approval for higher auto prices where appropriate as part of our comprehensive auto profit improvement plan. In the fourth quarter, approved rate increases totaled 1.3% for the Allstate brand, bringing the full-year approved rate increases to 7.2%. While our comprehensive auto profit improvement plan has improved margin, it's also reduced growth as we anticipated.
New business applications, shown on the top right, were down 21.9% for the full year, however were flat to the prior year quarter. Additionally, auto retention, shown on the bottom left, was down 0.8 points in the quarter and for the year. The combination of these two factors resulted in a decline of 2.9% in auto policies in 2016. However, both trends have stabilized in the quarter. Slide seven highlights the continued strength of Allstate brand homeowners. The top part of the page provides detail on our profitability results, showing that we have continued to generate attractive returns with a recorded combined ratio of 68.7 in the fourth quarter, benefiting from lower catastrophe losses despite Hurricane Matthew and the Tennessee fires. For the full year, the recorded combined ratio was 83.7, generating in excess of $1 billion in underwriting income.
The underlying combined ratio of 59.1 for the quarter and 59.5 for the year continues to reflect strong underlying profitability. While our underlying combined ratio results for the year have been excellent, our long-term target for this line remains in the low 60s. The bottom half of the page provides detail on our growth trends, which follow the auto line, since many of the customers in this segment prefer to bundle their purchases. New business and retention levels declined, leading to a 1.2% decline in policies in force. Slide eight provides a holistic view of our Esurance results. We continue to remain focused on improving auto profitability while investing in product and geographic expansion. The recorded combined ratio of 105 in the fourth quarter was two points below the prior year quarter.
For the year, the Esurance combined ratio was 107.5, lower than the prior year period by 2.8 points, driven by lower expenses. The lower expense ratio for the year reflects reduced advertising expense. The underlying loss ratio of 76.3 is higher than our long-term target, primarily due to higher auto claim frequency and severity. We will continue to take actions to enhance Esurance returns. Esurance growth trends are highlighted on the bottom of the page. Net written premiums continue to grow on increased rate actions, while policies in force were slightly higher than the prior year end. Slide nine highlights results for Encompass. Encompass remains focused on improving returns, which continues to adversely impact policy growth trends. The recorded and underlying combined ratios of 90 and 90.7 in the fourth quarter were better than the fourth quarter of 2015.
For the full year, the underlying combined ratio was 90.3, a 2.3-point improvement compared to prior year. Policies in force declined by 13.4% from the same quarter a year ago, reflecting both continued lower new business and retention. The majority of decline in policies in force was in six states. Now I'll turn it over to Steve.
Thanks, John. Turning to slide 10, Allstate Financial premiums and contract charges totaled $574 million in the fourth quarter of 2016, an increase of 4.9% when compared to the prior year quarter. For the year, premiums and contract charges increased 5.4% as Allstate Benefits surpassed $1 billion in premium with an increase of 442,000 policies. Allstate Financial's operating income of $130 million in the fourth quarter of 2016 was $32 million higher than the fourth quarter of 2015. Net and operating income trends by business are shown on the charts at the bottom of the page. Allstate Life net income was $58 million, and operating income was $66 million in the fourth quarter of 2016. Operating income was consistent with the prior year quarter, as higher premiums and improved mortality were offset by lower net investment income.
Allstate Benefits' net income was $22 million, and operating income was $23 million in 2016's fourth quarter. Operating income was flat to the prior year quarter, as higher premiums from policy growth were offset by higher benefits, debt amortization, and technology-related expenses. We expect to make additional investments in technology as this business grows. The $33 million increase in operating income in Allstate's annuity business compared to the fourth quarter of 2015 is a result of higher returns from performance-based investments. The chart on the top right highlights the growth in carrying value and annualized yield for Allstate Financial's performance-based investments, which primarily back the immediate annuity business. Net investment income of $101 million on these investments in the fourth quarter was very strong and was almost $30 million higher than the average of the preceding seven quarters.
We anticipate continued variability given the timing of investment origination and disposition and market conditions. Let's go on to slide 11 and our investment results. The chart at the upper left shows a shift we made in the risk profile of our $81.8 billion investment portfolio. The portfolio remains largely in corporate fixed income securities. To support long-dated liabilities, we are replacing market risk with idiosyncratic risk with an emphasis on ownership over lending. The conservatively positioned high-yield portfolio is being used in part as a bridge to fund performance-based strategies that will be reduced as this portfolio increases. Performance-based investments now comprise $6 billion or 7% of the portfolio. These assets require higher economic and regulatory capital, but we expect them to deliver attractive long-term economic returns for our shareholders.
In the upper right, total return for the quarter was a negative 0.7%, reflecting the decline in fixed income investment values due to the post-election increase in interest rates. The full-year return was a strong 4.4%. As you can see in the chart, solid investment income from both our market-based and performance-based portfolios has delivered a consistent contribution to return of approximately 1% each quarter. Net investment income and investment yield by business segment are shown in the bottom two charts. Property liability investment income reflects its interest-bearing yields closer to market yields and the portfolio's relatively short duration. Allstate Financial's investment income and yields reflects its portfolio's longer duration based upon its liability structure and the impact of last year's immediate annuity portfolio repositioning.
Performance-based investment results are the primary source of variability in income and yields between quarters, as you can see in the numbers above the chart in the bottom left. Performance-based investment returns in the fourth quarter were above our long-term target, as we had strong results in the quarter, which included income realization on direct real estate investments and substantial appreciation on a few funds. Slide 12 illustrates the continued strength of our capital position and highlights our financial flexibility. Shareholders' equity of $20.6 billion at year-end increased $550 million over year-end 2015. The debt-to-capital ratio of 23.6% reflects the issuance of $1.25 billion in senior unsecured debt in December to help fund the acquisition of SquareTrade, which closed January 3rd. We held $2.4 billion in deployable holding company assets at year-end, which excludes the funds utilized for the SquareTrade acquisition.
Book value per share of $50.77 increased by 7.2% over year-end 2015, primarily due to net income and increased unrealized gains in the investment portfolio, partially offset by capital return to shareholders. We returned $1.8 billion in cash to common shareholders in 2016 through a combination of $486 million in shareholder dividends and the repurchase of over 5% of our shares outstanding at the beginning of the year. As of December 31, 2016, there was $691 million remaining on the $1.5 billion repurchase authorization. I'll ask Jonathan to open up the line for your 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. Yes, good morning. My first question, what showed the favorable development within your auto book in the quarter, and did that come from some of the most recent accident years?
Steve, do you want to take that answer? That's okay. Okay. You can have it, man.
Whatever you want.
Elyse, this is Steve. The results that we look at, as you know from our reserving processes, is we do a quarterly review, bottoms up. We have the reserving department reports to me, separate from our business unit. We look, as you indicated, year by year, we look at virtually every state. We have a very specific program methodology, both mathematically and qualitatively as we walk through the results. We look at where by line of business. We look at a tremendous amount of detail. The actual results that you're seeing are based on across multiple lines and across multiple of our prior years. You'll see in the 10-K when we file, we have some additional information that will give you a little more detail of the specific years. Clearly, it's not just like 2015 or 2014.
It goes back three or four years, which is the nature of our business and the length of the tail of our bodily injury type coverages, in addition to the shorter tails you might see that we generally would report earlier in the year.
Okay. Some of it did relate, I guess, to the more recent accident years than we were seeing some of the higher frequency trends like 2014 and 2015?
Well, the frequency shows up immediately, right? We do it based on counts. Some people do target loss ratios. That's not the way we do most of our reserving. We do it for some lines of business. What happens is, of course, some people say, "Well, we think we're going to get a 70% loss ratio," so they take 1,000 of premium, and they book 700 of losses. That's not the way we do it in our business because of the short-tail nature of it. What you see is what you get. There's also a little bit of movement between quarters, which doesn't show up in the triangles in the 10-K, but we adjust as we go throughout the year to reflect what we're seeing in the specific year.
Okay. Thank you. As we see the frequency trends got a lot better, again, in the fourth quarter, you guys reported a pretty good improvement within your auto margin. Now, some of your commentary points to looking to pursue growth. Can you just provide some more color? Are you looking in certain states, I guess, that are showing better trends than other? Or is this broad-based? How do you think about both new business, which I did see flattened out in the quarter, new business growth from here, as well as just pursuing growth and higher retentions across your overall auto book of business?
Well, I'll make one overall perspective, then Matt will give you some specifics. I would not interpret this quarter as that was then and this is now, that we're going back to where we were. Miles driven were still up in the quarter. People continue to drive more than they have in the past. Accidents tend to be more frequent, whether that's distracted driving or anything else. I would not want you to take away from the fourth quarter that we're headed back to where we were. Matt can talk about where we are in pricing, and we're up almost $100 per policy in auto insurance premiums, which is quite substantial, which gives us more flexibility to pursue the growth you're talking about, and Matt can talk about that.
Morning, Elyse. It's Matt. Let me follow on to Tom's comments. First, I think about this in 3 different components. First, the rate environment. The way I would describe it is that we have completed some of the work necessary to kind of catch up to that spike in auto frequency that we saw over the last 18 months. You saw a fairly dramatic increase in rate taken in order to catch up quickly and try to get back to appropriate margins. We believe, with some exceptions in a few states, but in the vast majority of places, we have now caught up, and now we're in the mode of keeping up with emerging trends.
We don't know what this year will bring or next year, but we're now positioned very well in order to be able to react quickly and to keep up with trends as they emerge. Having caught up, it gives us a little freedom to do some things that we were unable to do over the last couple of years. One of which is to begin adding agencies again, points of presence is extremely important for us to enable growth. That's points of presence, not only of agencies, but the licensed sales professionals and financial specialists that go along with them. It's hard to do when you're focused intensely on taking rate and dealing with the disruption that causes. Now that we're in a slightly more stable environment, we're able to focus on that strategic deployment once again.
The second thing that we're hoping to do is to stem the retention losses and declines that we've had. Growth, overall growth in items in force, as you know, is a combination of new business growth and retention. It's a vicious treadmill if retention keeps declining because it's very hard to add enough new business to make up for that loss. We saw a slight stabilization in the decline in this last quarter that makes sense to us since we took a huge amount of rate in the second quarter of 2016. Most of that has burned through already and impacted customers, and we've dealt with the disruption already. To the extent we're able to keep more moderate rate increases at this point on, we believe customer disruption will be mitigated. We'll be able to hold on to more customers.
On the new business side, the third component is the competitive environment. While we took a fair amount of criticism or questioning for reacting quickly about 18 months ago, I think that has been to our benefit. I think most, if not all of our competitors have talked and recognized some increased frequency, and in many cases, are reacting only recently from a rate-taking perspective. That type of rate taking from our competitors creates shopping behavior in their customers. To the extent we're better positioned, we have increased agency capacity, and our agents are ready, willing, and able to serve those customers. We should be in a very good position to pick up more of those customers from a new business perspective.
It's really a combination of all those items, of improving retention, being in a position to capture more new business, and hopefully in a more stable rate environment, disrupting fewer customers.
Okay, great. Thank you. One last just quick question.
You know what, Elyse? Let's go to the next question because we got about 12 people in line here, okay? We'll come back to you if we have time.
Certainly. Thank you. Our next question comes from the line of Greg Peters from Raymond James. Your question, please.
Thank you for the call and taking my questions. I'll keep it to two questions, I guess.
Thank you.
Just as a follow-up, Matt, you were talking about agencies and points of presence. I was wondering if you could provide some demographics around your agencies, how many are close to retirement. Should we infer from your comments that because you took pricing actions earlier than many of your peers, that your product and price position should improve throughout 2017? Then I'll have one follow-up to that.
Okay, Greg. Well, let me try. I don't have the actual agency demographics in front of me. I don't want to go based upon memory. I can tell you that we have, over the last several years, engaged in a different type of recruiting and selection process for our agency owners, both in terms of geographics and in terms of demographics. I think, as you know, we have a historical presence on the coasts. That has served us well, we've been able to grow well there. We have not done as well in terms of strategic deployment in some of the heartland states, in the lower premium states. We have a concentrated effort to do that.
In terms of the people that we're bringing on, I would say that we've done what I consider an excellent job of using data and analytics and some of the predictive tools available to us to better select people who will thrive in our new environment. We've talked a little bit about our shift to a Trusted Advisor mode. The type of agency owner and licensed sales professional that would have thrived maybe 10 or 15 years ago will not be the same type of person that we're looking to now to thrive in the next 10 to 15 years. The Trusted Advisor model is much more relationship and advisory oriented. It's much more technology driven.
It requires people who really want to get out there and use data and analytics and technology themselves to serve customers, to be able to really advise them about the risks that they take on in their lives and how to mitigate those risks. It requires somebody who is more sophisticated about money management so that they help our customers manage their cash flows as well as manage their risks. The type of agency owner we're bringing on and licensed sales professional we're bringing on, I feel very confident is one designed for the future. We had a decline in most of 2016. We saw a pickup in the last quarter of about 200 licensed sales professionals, and that's really key for us.
When you look at our agencies, we've always had a focus on increasing the number of agencies because we believe physical points of presence are important. Agency capacity is just as important. The number of licensed sales professionals and their specialization, their ability to sell life and retirement, commercial insurance, all of the different consumer household lines, homeowners, their ability to specialize, segment their licensed sales professionals so that they can really focus on serving customers, is going to be key. I think from a demographic perspective, we are certainly creating the type of a field force that's better able to serve customers in the future.
Excellent color. Thank you. I guess the follow-up question to all that, I just go back to the product positioning for 2017 in light of your comments about your early actions on price versus some of your peers. Obviously, the focus will be going forward, for Wall Street and some of your investors will be PIF growth and improvement in retention. Just to clean up on that. Thank you very much for your answers.
Sure, Greg. Well, first, I'm going to ask you to resist the natural tendency to think that growth is only a factor of price. Certainly, rate and price competitiveness is a component, but it's certainly not all of it. Some of the things I just talked about with Trusted Advisor and better serving our customers and a better claims process and a more focused advisory system, the ability to do in-force reviews and annual reviews on their holdings, that's all a piece of retention, and that's all a part of growth. You are correct, shopping behavior is typically triggered by price changes. To the extent Our competitors are now taking additional rate in fairly significant chunks because they are catching up, as we were about 12-18 months ago. That it's likely to trigger shopping behavior on their customers.
If we have already gone through that and are stable, I believe we will be better positioned to potentially add some of those customers, and begin serving them as Allstate customers.
Thank you.
Thank you. Our next question comes from the line, Randy Binner from FBR. Your question, please.
Hey. Thank you. I just wanted to follow up on the response to Elyse's question on reserves. I guess the way I heard it was that the favorable PUD was primarily from auto BI over the last 3 to 4 years. What do you think caused that? Is it a change in your claims approach that you mentioned earlier? Is there some other aspect that caused that favorable development? Because it was notable, it's not really the pattern we've seen from others.
Well, I can't speak to what it means with respect to others. It's as you described, it's mostly auto. But it's really not a big percentage, by the way, of total reserves and loss costs. When you look at it, there's normal variation in making estimates when you're looking out to anywhere from 2 to 4 years to get claims resolved, particularly in BI, which tend to stretch out to 3 to 4 years. There's nothing unusual about it. We've had reserve releases much greater than that over time. I don't think you should read anything into it other than we conservatively estimate what we think our loss costs are. To the extent we wind those years up, not as much cash went out as we estimated will go out, it shows up in the P&L.
The other question is, I appreciate all the commentary about Allstate being kind of closer to loss trend with all the initiatives you've done over the last couple of years, that's clear in the numbers. What's your sense of what's actually happening out there from an accident frequency perspective? Meaning, are accidents still worse? Are we still seeing distracted driving impact the frequency of accidents? I'll leave it there and have one quick follow-up.
Randy, it's Matt. I'll try to answer that. There's loads of data out there about the correlation between miles driven and additional accident frequency. There's also data out there about increased deaths from increased fatalities, because the seriousness of the crashes is greater. We have talked about, I think we're seeing mounting evidence of the increased influence of distracted driving on that combination. The issue is this, people refer constantly to the fact that there have been cell phones for many years, actually, cell phone use hasn't spiked up that dramatically. In fact, smartphone ownership and use has spiked up dramatically. When we look at smartphone ownership statistics and smartphone use and compare that to accident frequency, the correlation is striking.
When you have increased miles driven due to increased economic driving, lower gas prices, so increased leisure driving, you have increased number of drivers on the road. That means lower margin for error because you used to have maybe 10 cars packed into a space, and now you have 15. Even if there was distraction before, you had greater reaction time before. Now with increased density, you don't have as great a space to react in, and so we're seeing increased accidents. We saw a spike in that, and that's what we reacted to. When we say, and I think the comment Tom made a few minutes ago is really important, when he said there's a difference between saying things have stabilized and saying things will revert to where they were before.
We don't think accident frequency is reverting back to where it was in 2012 or 2013 or 2014. What we're saying is the spike is over, and it has now stabilized at a new norm. We think, at least our belief is, and our assumptions for all of our future work is, that we are at that new norm in heightened accident frequency due to the greater economic activity, the greater density of cars on the road, and the greater use of smartphones and distracted driving in the car.
If that's a new plateau, is the kind of penetration of these cases by the trial bar also at a new high that we should expect to continue?
I'm sorry, could you say that again?
Yeah, sure. It was a trial bar question.
Oh, a trial bar. That's I think more of a severity question than a frequency question. Yeah, you've heard some of our competitors talk about the combination of medical inflation and the trial bar influence on severity costs. I think to some extent, we've seen many of those same things. I'd like to think we had been doing work on claims management and claims handling for the last several years as we saw those trends emerging. That led to some of the work that we talked about that influenced the BI frequency and the BI severity trends that you see that we reported on last quarter. I'll remind you that we did have a process change. We talked about that last quarter.
We encouraged you then. I will encourage you now to look at BI frequency and severity in combination, because the decline in frequency has led to the increase in severity. We start really high. I'll remind you that our number one commitment is always to pay customers and claimants what we properly owe them. That being said, there are claims management techniques to ferret out and eliminate fraud and wastage that don't compromise on that promise. That's what we've done. We began requiring some enhanced documentation of injuries and related medical treatment to ensure we were paying properly. That did change the mix of claims paid. In 2015, about a third of the BI claims paid were below $1,000. In 2016, that has dropped to about a quarter, leaving us with fewer but higher severity claims. That's what you see, a changed payment mix.
It's still in line when you move all that around. You rationalize the two. We believe it's still in line with medical inflation. We're comfortable with our incurred loss trends for BI. As Steve said in his response to the reserving question, don't confuse some of these operational claims statistics with profitability and reserving. When we are setting reserves, when that team is working on the reserves, they know about the claims process enhancements. They take the impact of those changes into account when setting reserves. They are doing very sophisticated segmented analysis. While the claims process changes will influence those statistics you see on paid severity and paid frequency, they're not influencing the reserves. Okay, next question, please.
Thank you. Our next question comes from the line of Amit Kumar from Macquarie. Your question, please.
Thanks. Good morning. I'll make this quick. Two questions. The first question goes back to, I guess, Elyse's question on the trends. When you talk about catching up to the spike and you also address miles driven, distracted driving, pedestrian fatalities, I'm curious, why shouldn't I be nervous that these trends will continue at elevated levels, and hence it's too early to start normalizing pricing? We've just caught up to the past trend, but it seems all the indicators point to a continued change in the slope of this loss cost trends.
Well, thank you for that question. It's Matt again. Look, this is what we do. I don't mean to be flippant about it, but we have really talented teams of people monitoring these trends on a state and sub-geography, sub-state level, whose job it is to watch net trends emerging and look at indications and ensure that we are keeping up with them as quickly as possible consistent with actuarial standards so that we can file for the rates and convince our state regulators to approve those rates. This is what we've always done, and we think we're well-positioned to do it.
In fact, coming out of a spike like this, one of the benefits, there's always a benefit of going through something like this, is it hyper-focused us on those troubling segments of the business and allowed us to take segmented rates on those parts of the business so that we can improve our loss ratios as quickly as possible. We believe that we have a high-quality book at this point. We believe that we know where the segments that we need to keep an eye on are. We believe we've gotten those segments rate adequate at this point. We feel very comfortable with being able to manage some continued inflation and some continued rise and escalation in some of these costs. I will remind you that if we go through that, we will not be going through that alone.
Our competitors will be going through the exact same thing. We think that we have an advantage with our skill in reacting quickly. We have an excellent relationship with the state regulators. Our teams have their respect. When we do a filing, it's a complete, thorough filing. We have an extremely high approval ratio and the ability to get rates into the book as quickly as needed.
Tom, at this time, let me add that. From our standpoint in pricing, it's only the now. It's like we look at what the frequency and severity is. We think we look at it at that moment. It might be March, April, May, might be August. Whatever it is, we look at the premiums we think we're going to get, and we adjust. We don't have a fixed assumption that's for the year on frequency and severity that is used in pricing. That's what Matt's explaining. It's very organic. When you're doing your models, of course, you're looking and saying, what's the frequency and severity assumption looking forward, and you have to put something in, and then you have to put in something to price. We do that continuously by state, by sub-piece of the state every day.
If frequency is two points higher, then we'll take two points more of rate. If it's two points lower, then we'll decide whether we think that's reasonable or not. Unlikely we will lower our rates, but we may choose to invest more in expenses to begin to grow the business. That's why our range is what it is. The 87-89 range, we're completely comfortable with. There's, as you know, obviously a one- to two-point variation if you add up frequency and severity over the course of a year. To extent, but we don't want you to think, so we lower the range from where it was in 2016. This does not mean that we think you should take 2016's numbers and lower that by the same amount. That's just not the way we do it. It's not the way we run the business.
Well, I guess I'm just a glass half empty kind of guy. The other question I had, and switching topics here, is the broader discussion on corporate taxation. If the taxes were to go down, does that get reflected in your pricing actions? I know it's all hypothetical, or should we anticipate that netting down to your bottom line? Thanks.
Well, taxes are, of course, a complicated thing as it relates to Allstate. The first and probably the biggest impact is when it relates to taxes. If it drives economic growth, that's good for the company because we are long U.S. growth, whether that's our investment portfolio or the growth in auto ownership and home ownership. Both of those could be very good for us. As it relates to, you're at then one component of it, which is if you look at our P&L, I'll go way up and now get back to pricing. We're a full taxpayer. Our industry does not have all kinds of special deductions. We think U.S. taxes should be based and corporations should pay based on what they use, not some centralized view of how to control the economy.
We know centralized planning doesn't work, using the tax code to try to centralize plan and drive certain industries doesn't make sense to us. As a result of that, though, there are a number of industries and companies that pay a lot less tax than us. We think it should be more of an even playing field, we think our tax rates should be lower. That would obviously immediately be good for our earnings because we would have less cash we have to pay out every year in taxes. What happens to pricing then really depends on the individual state you're in, where you are in pricing that piece. I would make a couple of comments. One, I would reiterate what Matt said. While price is important, we don't look at price as the only lever to growth.
We think price is what we have to charge relative to the value we provide, but Matt's working hard on providing more value rather than just, let's be the cheapest in town. It also depends, there's a variety of other facts which would rattle through the pricing besides just the tax rate used in determining our return on capital. What happens to interest rates? What happens to inflation in that scenario? There's a whole bunch of other, I would say, secondary and tertiary, but linked economic things that would rattle through pricing. Net-net, I believe a lower cash rate would be good for Allstate in the short term.
Got it. Thanks for the thorough answer, good luck for the future.
Thank you. Our next question comes from the line of Jay Gelb from Barclays. Your question, please.
Thanks. Good morning. The first question I had was just on expectations for share buybacks in 2017. Should we be thinking about the 2016 pace of $1.4 billion as a reasonable run rate?
Jay, this is Tom. I think you should think about, we still have $651 million yet to go. You can put that in the bank that we fully expect to get that done by the time its authorization is up, which is late in 2018. Based on the pace we're going, it looks like we'll be done sooner than the authorization. What we should have in there is then we'll look at where we are in terms of capital, and we'll make the same judgment we make every other year, which is between that and the dividend, how much cash can we return to shareholders? If we don't have a use for it, we return it. Share repurchases will continue to be part of what we do.
We don't have a set number that we're ready to tee up because we still got a ways to go on the existing share repurchase program.
Okay. On the Allstate brand auto rate increases, I believe the expectation now is for the pace of rate increases to moderate in 2017. How much do you think that could slow? I'm sorry.
How much do you think the pace of rate increases in Allstate brand auto insurance could slow from the extremely successful seven-plus % you had this year? Sorry, Jay, I was just helping out a question.
I like the way Tom-
Great. With the editorial comment. Thank you.
No, that wasn't editorial.
We're going to let Tom ask all the questions from now on. I wish I could tell you. I can only tell you that we're done catching up, as I said earlier. When you're in catch-up mode, you're sprinting, and you're taking as much as you can in order to get rate adequate and to clean up the book and get to where you're earning an appropriate return on the business. It feels like, in about probably three-quarters of the states, we will be reducing that level of rate. We will not be taking the kind of rate we took in 2016. That being said, if we see additional pressure, we're going to take that rate. I keep coming back to the same thing that both Tom and I have said during this call.
We do have some assumptions about where loss costs are going to go next year, but they're just that. They're just assumptions. They're not a plan. Our plan is to keep up with loss costs as they emerge and take rate necessary in order to do that. To the extent reality meets our forecast, we have a stabilization of frequency trends and the severity trends where they are now, I would expect us to see a moderated rate need as we're keeping up with what is more normal inflationary trends at that point.
That makes perfect sense. Thank you.
Thank you. Our next question comes from the line of Sarah DeWitt from J.P. Morgan. Your question please.
Hi, good morning. On the auto insurance margin, just to clarify, are you at your target auto margin on a GAAP basis now, or is there still room to improve? The reason I ask is because the Allstate brand auto margin would've been 97% ex the 2 points of favorable development.
Yes. Sarah, thanks for the question. I was hoping Tom would've asked this one, so he would've asked it differently. We still have room to improve. I think we feel really good about the progress that we've made, and we feel like we are earning an appropriate return on our investment at this point, but it's not our target return yet, and we will continue. That's multiple levers. That's the claims management, that's expense efficiency, that is additional rate. It's additional work on quality of the book. It's correct class work. It's underwriting work. It's all the things that we did really, really strongly to get back to profitability. All those levers are the same levers we need to continue to pull. Now we can dial them instead of pulling them really hard.
We can dial them carefully and thoughtfully to get back to our target returns, but we're a lot closer than we've been for a while to where we want to be, but our work is not over.
Okay, great. Thanks. Then just on the auto severity, I apologize if you addressed this, it was a bit confusing. My sense is your incurred trends are consistent. We've heard one of your competitors saying they're seeing much higher severity from more occupants per vehicle, more fatalities, higher legal costs. Sounds like you acknowledge some of those, it doesn't seem like you're seeing any of this in your incurred trends. Can you just clarify that?
I think you're overreading into what we said about there are normal inflationary costs embedded in our BI. Other people, they're saying they got lots of stuff, it's driving it. Ours are just normal, we incur it that way. We show you paid, that's not the way we use paid to help us think about how we reserve, it's not the way we track the P&L. We track the P&L on what we think is going to happen. We think it's normal. Maybe answer a question, give slightly longer perspective on the combined ratio. We had great returns in the auto insurance business for 12+ years, which showed us that the market was willing to give us the economic rents based on the value we provided.
In 2015, we had this huge spike in frequency and severity, as Matt said, we said, "Oh, we don't like that." We like those old returns better. We had to move quickly. Our team did. They executed well. We moved those returns up. Are we back to the amount of return we had for that period of time for that, like four years? No. Are we in a fundamentally different position today where we feel like now what we should do is thinking about how we get there and over what period of time we get there than whether we can get there? That's the process. That's the Allstate brand. Esurance is in a slightly different position. It's had good movement throughout the year Picked it up mostly in the first half of the year.
We're watching the fourth quarter because it bumped up a little bit if its frequency and severity trends increase, it does the same thing that Matt described. It's the now. It's whatever happens. Encompass, we feel like we've had some good success this year. As one of you pointed out, we finally managed to get some underwriting income out of that business we're happy that it's now positioned well. It's in the same place. We're feeling good about the process the way the machine works here to help us manage whatever comes our way. That's what you're buying. You're buying the machine, not somebody's projections on frequency and severity.
Great. Thank you.
Okay. Maybe we'll take one last question, and then we'll close. Bob, I think. Is Bob? Bob, I hate to leave Bob out. Cutting him short.
Certainly. Our final question comes from the line of Bob Glasspiegel from Janney Montgomery Scott. Your question please.
Thanks for squeezing me in, Tom. Two quickies. On SquareTrade, appreciate the increased disclosure. Is there any more color on what the goodwill's going to be, and is that going to be excluded a la Esurance or included in core earnings?
We just bought it, as you know, Bob, last month, so we don't have any updated numbers. We obviously are working hard at coming up with what the split between goodwill and intangibles are. I missed your second question.
Is it going to be included or excluded in core earnings, the goodwill piece?
The goodwill, only intangibles are amortized, and they are excluded from operating income.
Okay.
Our definition of operating income.
Got you. One other quickie. The life earnings were up by more than sort of the partnership delta. Mary Jane, was there anything unusual or recurring in that bump up in the earnings level in the quarter?
For Allstate Financial, Bob, really for the annuities, you had the benefit of the performance-based investments that Steve alluded to. That was at 13.8%. We also did have favorable mortality as well in both the annuity business and the life business. We did have a favorable quarter from a mortality perspective impacting the results.
What she's saying, don't multiply it by four.
No, I knew the bump up in partnership income was not recurring, but it was more than that. It was at $20 million-$30 million more than just the increase in partnerships. Mortality could explain some of that. Thanks.
Most of it was the partnership return, Bob. The mortality was worth about $6 after tax on the annuities.
Okay, I'll go through it with John. Appreciate it.
Yeah. Okay. Let's do that.
Thank you all for participating. If you go to slide 13, there's quick reasons as to why this is a good investment, why you should continue to hold Allstate. We're positioned for profitable growth. We know how to move in our marketplace well, we're building long-term growth platforms along the way as well. Thank you much. 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.