Good day, and thank you for standing by. Welcome to Allstate's third quarter investor call. At this time, all participants are in listen-only mode. After prepared remarks, there will be a question-and-answer session. To ask a question during this session, you'll need to press star one one on your telephone. Please limit your inquiry to one question and one follow-up. As a reminder, please be aware that this call is being recorded. Now I'd like to introduce your host for today's program, Mr. Mark Nogal, Head of Investor Relations. Please go ahead, sir.
Thank you, Jonathan. Good morning, and welcome to Allstate's third quarter 2022 earnings conference call. After prepared remarks, we'll have a question-and-answer session. Yesterday, following the close of the market, we issued our news release and investor supplement, filed our 10-Q, and posted today's presentation on our website at allstateinvestors.com. Our management team is here to provide perspective on these results. As noted on the first slide of the presentation, our discussion will contain non-GAAP measures, for which there are reconciliations in the news release and investor supplement and forward-looking statements about Allstate's operations. Allstate's results may differ materially from these statements, so please refer to our 10-K for 2021 and other public documents for information on potential risks. Additionally, we will be hosting our next special topic investor call on December 2nd, focusing on Allstate's auto and home insurance claims practices and reserving process.
Now I'll turn it over to Tom.
Well, good morning. Thank you for investing your time with Allstate today. As you know, we pre-released earnings several weeks ago and reported the net loss for the quarter. That reflected a small underlying underwriting margin that was offset by increases in reserves for prior years and a mark-to-market loss on public equity securities. Mario and Jess will go through the details of the quarter and the reserve changes after I set some context. Let's start on slide two. Allstate's strategy to increase shareholder value has two components, increase personal Property-L iability market share and expand Protection Services, which are shown in the two ovals on the left. We're building a low-cost digital insurer with broad distribution through Transformative Growth to increase market share.
We're also broadening protection offerings and leveraging the Allstate brand, customer base, and capabilities with expanded distribution. In the third quarter, we made progress executing this strategy while we continued to implement a comprehensive approach to improve auto profitability, which is shown in the right-hand panel. That includes broadly raising auto insurance rates, which you've seen in our disclosures. Operating expenses were lowered, including advertising and more permanent reductions in the operating cost structure. Underwriting guidelines have been adjusted to reduce new business volume where we're not earning adequate returns. Our claims operating processes are being modified to manage loss costs in a high inflation environment. We believe this plan will return auto insurance profitability to historical levels. While the current environment requires focus on improving margins, we continue to advance the Transformative Growth strategy to gain market share when profitability improves.
In addition, the Protection Services businesses are generating profitable growth. Investment returns were negative for the quarter and year to date, but better than the overall declines in the bond and equity markets. This reflects risk reductions implemented late last year. You'll remember we reduced the bond portfolio duration to lower exposure to high interest rates, which enabled us to avoid about $2 billion in losses in the bond portfolio. Our capital position is strong, and as a result, we were able to deliver attractive returns to shareholders. Jess is gonna discuss capital in his section, but let me provide a few summary points since this was covered in some of the reports issued last night. First, we have plenty of capital, and there's $4.5 billion of deployable capital at the holding company level.
Secondly, the significant reduction in risk with the sale of the life and annuity operations occurred last October, and that needs to be considered. This divestiture reduced assets by $34 billion and freed up capital. Thirdly, we use a really sophisticated approach to determining required capital that goes far beyond statutory capital and premium to surplus ratios. For example, if you just use statutory capital as a measure, the life company equity would be included in capital historically, which we did not believe was appropriate, so we never included it. Our methodology has led to those strong results. We did decide to complete the remaining $1.4 billion stock repurchase over more than the next six months, which was our prior target we had disclosed to, but we still expect to complete it in the second or third quarter of next year.
In summary, you know, we're really well capitalized, and this year's results have not changed our strategy or earnings power. Let's move to slide three to go through the third quarter performance in detail. Total revenues of $13.2 billion were 5.8% above the prior-year quarter as Property-Liability premiums earned increased by $5 billion or 9.8%, which reflected higher average premiums and policy growth. Lower net investment income and net losses on investments and derivatives negatively impacted the year-over-year comparison there.
A net loss of $694 million and an adjusted net loss of $420 million in the third quarter reflected a decline in underwriting income due to an increase in Property-Liability prior year reserve estimates, which was $875 million, and that excludes catastrophes and increased loss costs in the current year. Looking forward beyond improving profits in auto insurance, if you go to slide four, you'll see the flywheel of growth that'll increase personal Property-Liability market share. This is a multi-year initiative designed to build a low-cost digital insurer with broad distribution. That'll be accomplished by delivering on five key objectives, improving customer value, expanding customer access, increasing sophistication and investment in customer acquisition, deploying new technology ecosystems, and enhancing organizational capabilities.
We made significant progress on all these components, and we're well on the way to ending, really being in a position where we can dial up growth quite rapidly when profitability improves. Now let me turn it over to Mario, and he'll go through our Property-Liability results.
Thanks, Tom. Let's start by reviewing underwriting profitability for the Property-Liability business in total on slide five. Our underwriting results reflect the high level of inflation and the impact of reserve strengthening in the quarter, with a third quarter recorded combined ratio of 117.4% for auto, 91.2% for homeowners, 126.6% for all other lines, and 111.6% for total Property-Liability, which is shown on the left chart. Remember, our goal is to run the auto business with a combined ratio in the mid-90% and homeowners at around 90%. While homeowners was close to our target in the quarter, we continue to focus on improving auto margins through a comprehensive plan that is being implemented to get us back to our mid-90% objective.
The third quarter underlying combined ratio for auto insurance was 104%, as you can see on the right, so we're raising auto insurance prices, reducing growth investments, lowering operating expenses, and adapting claims practices to a high inflationary environment. While the homeowners business generated $245 million of underwriting profit, higher severity resulted in an underlying combined ratio of 74.6%, which is above where we manage it to, and we are increasing prices through both rates and the inflationary adjustment factor embedded in our homeowners product to improve underlying margins going forward. One of the reasons that we have an industry-leading homeowners business is because we proactively manage the risk and return profile of each market that we operate in.
Based on this approach, we have decided to stop writing new homeowners and condo insurance in California at this time, given our inability to fully reflect the cost of providing these products in the state, including both loss and reinsurance costs. We intend to continue protecting our existing California property customers by offering ongoing coverage to them. Other lines are mainly traditional small commercial auto and shared economy insurance, both of which have recorded an underlying combined ratios above target levels. As a result, we made the decision to exit five states in the traditional small commercial business and no longer provide insurance to transportation network companies unless pricing begins to utilize a telematics-based framework for pricing. These actions are expected to reduce commercial business premiums by over 50% next year. Let's move to slide six and discuss auto profitability in more detail.
As you can see from the chart on the left, which shows the auto insurance combined ratio and underlying combined ratio over time, we have a long history of meeting or outperforming our mid-90% combined ratio target, supported by our pricing sophistication, underwriting and claims expertise, and expense management. 2020 is an outlier with much better than target results due to reduced accident frequency in the early stages of the pandemic. In 2021, and again this year, we have experienced both higher frequency than 2020 and the impacts of inflation, which have dramatically increased the cost to repair or replace cars and raised the cost of settling injury claims with third parties who were injured in accidents with our customers.
In addition, this quarter we strengthened prior year reserves by $643 million, which Jess will discuss in more detail in a few minutes, and experienced higher catastrophe losses, mainly from flooding associated with Hurricane Ian. As a result, the auto insurance recorded combined ratio was 117.4%, with reserve strengthening and catastrophes contributing 8.5 and 4.4 points respectively to this result. The right chart quantifies the drivers in the year-over-year change in the underlying combined ratio, which increased from 97.6% to 104% and excludes catastrophes and the reserve changes. The red bar reflects the increase in underlying losses, primarily due to current report year incurred severity strengthening across major coverages and moderately higher frequency than last year.
The increase to underlying loss costs were partially offset by 4.3 points of average earned premiums from implemented rate increases and a 3.1 point reduction in underwriting expenses to get to the 104%. To add more clarity to current quarter results, we also highlight the 2.6 point impact of increasing full year claims severities, in the third quarter for claims that were reported in the first and second quarter of this year. This impact is noted by the green bar on the right-hand chart. Excluding this intra-year strengthening, the third quarter underlying combined ratio would have been 101.4%. Current report year incurred severity for collision and property damage claims were increased to 17% above the level recorded for the full year 2021, and bodily injury severity was increased to 12%.
Moving to slide seven, let's discuss key components of our multifaceted plan to deal with inflation, raising auto insurance prices. Growth in average premium per policy is accelerating due to implemented rate increases over the last 12 months, but the impacts to average earned premium per policy is on a lag due to the six-month policy term. Over the last 12 months, we've implemented Allstate brand auto rate increases across 53 locations for an annualized written premium impact of approximately 13.7% or nearly $3.3 billion, including 4.7% in the third quarter. The chart on the page is an estimation of when the rate increase is implemented in the last 12 months will be earned into premiums.
This illustrative example assumes only 85% of the annualized written premium will be earned to account for retention and the fact that some customers modify policy terms such as deductibles or limits when faced with price increases. As you can see, looking back at Q3 2022, the estimated impact of the $3.3 billion in annualized implemented rate had only an estimated impact of $660 million on earned premium, which is expected to grow by over $2.1 billion through the end of next year. Given ongoing loss cost inflation, we expect to implement additional rate increases in the fourth quarter of this year and into 2023, and those will be on top of increases implemented since Q4 of last year and additive to the increases shown here.
Moving to slide 8, let's discuss the timing of how these rate increases will impact the combined ratio for auto insurance. The chart on this page is an illustrative view to show our path to target profitability, along with the magnitude of actions already taken and required prospectively. Starting on the left, through the first nine months of the year, the auto insurance recorded combined ratio is 109.3%, as shown by the first blue bar. From this starting point, we remove the impact of prior year reserve increases and normalize the catastrophe loss ratio to our five-year historical average. This improves the combined ratio by approximately six points, represented by the first green bar.
The second green bar reflects the estimated impact of rate actions already implemented when fully earned in the premium, which is an additional $2.3 billion of premium across the Allstate and National General brands, or approximately 8 points. These amounts will be mostly earned by the end of 2023. Of course, loss costs will likely continue to increase, whether from inflationary impacts on severity or higher accident frequency, which would increase the combined ratio. Prospective rate increases must meet or exceed loss cost increases to achieve historical returns. Combined with other non-rate actions, such as reducing new business and expenses, we expect to achieve an auto insurance combined ratio target in the mid-90%. The timing of reaching this goal will be largely dependent on the relative increase in premiums and future loss cost trends.
Moving to slide nine, let's now take a look at our industry-leading homeowners business. As you know, a significant portion of our customers bundle home and auto insurance, which improves retention and the overall economics of both product lines. We have a differentiated homeowners product underwriting reinsurance and claims ecosystem that is unique in the industry. Our long-term underwriting results reflect this dynamic with a five-year average recorded combined ratio of 91.9%. The third quarter combined ratio for homeowners improved to 91.2%, primarily driven by lower catastrophe losses compared to the prior year quarter, as you can see by the chart on the left. Enterprise risk and return management actions reduced our Florida personal property market share to 2.6%, which, combined with a comprehensive reinsurance program, including our standalone Florida property coverage, significantly mitigated net losses from Hurricane Ian.
Estimated gross catastrophe losses due to the hurricane totaled $671 million and were reduced by $305 million in expected reinsurance recovery, primarily related to property reinsurance for our standalone Florida property insurance company, Castle Key. Of the $366 million net loss from Ian, only approximately 25% was from property lines. Homeowners insurance is certainly not immune to the rising inflationary environment as we continue to be impacted by increasing labor and material costs. In the third quarter, non-catastrophe prior year reserves were strengthened by $51 million, and current report year incurred severity was increased primarily as a result of increasing inflation in both labor and material costs.
The resulting impact to the underlying combined ratio from current year severity strengthening was 3.8 points in the third quarter, partially offset by slightly lower non-catastrophe frequency. Similar to auto insurance, there was an intra-year impact of 2.4 points related to claims reported in the first and second quarter of this year, which is reflected in the underlying combined ratio for the third quarter of 2022. To combat inflation challenges, our products have sophisticated pricing features that respond to changes in replacement values. The chart on the right shows key homeowners insurance operating statistics.
Net written premium has grown sharply throughout 2021 and into 2022, increasing 9.4% from the prior year quarter and 12.9% year to date, primarily driven by a more than 13% increase in Allstate brand average gross premium per policy and a 1.4% increase in policies in force. The Allstate brand increases are partially offset by lower National General premiums and policies in force as we improve underwriting margins to targeted levels in this brand. We are continuing to raise homeowners prices to address inflationary pressures, growth both through the impact of inflation on insured home valuations and filed rate increases.
Beyond these pricing actions, we have also decided to limit new business where margin targets cannot be achieved in the near term, including the action I previously noted of suspending the sale of new homeowners insurance policies to consumers in California. Let's delve deeper into improving customer value through expense reductions on slide 10. Let me start by saying we remain on pace and committed to our long-term objective to reduce our adjusted expense ratio, which is a metric we introduced about a year ago to track our underlying progress to improve customer value. This metric starts with our underwriting expense ratio, excluding things like restructuring, coronavirus-related expenses, amortization and impairment of purchased intangibles, and investments in advertising. It then adds in our claims expense ratio, excluding costs associated with settling catastrophe claims, because catastrophe-related costs tend to bounce around quarter to quarter.
Through innovation and strong execution, we've achieved almost 3 points of improvement since 2018. Over time, we expect to drive more than 3 points of additional improvement from current levels, achieving an adjusted expense ratio of approximately 23% by year-end 2024, which represents a 6-point reduction compared to 2018. The chart on the slide shows the Allstate Protection underwriting expense ratio since 2018 and quantifies the impacts from third quarter 2022 compared to the prior year quarter, reflecting actions we've taken to address the current operating environment. The first green bar on the left shows the decline in advertising spend as growth investments have been reduced given our focus on improving margins. The next green bar shows a decline in the amortization of deferred acquisition costs, primarily driven by the phase out of enhanced compensation models for new agents.
Our future cost reduction efforts are focused on digitization, sourcing, and operating efficiency, and continuing to reduce distribution costs. Let me now turn it over to Jess to discuss our reserving actions in the quarter and the remainder of our business results in more detail.
Thank you, Mario, and good morning, everyone. On slide 11, let's begin with our prior year reserve development. Property- liability prior year reserve strengthening, excluding catastrophes, totaled $875 million in the third quarter. The pie chart on the left breaks down the impact by line, with $643 million driven by personal auto, $120 million run off Property-L iability from our annual reserve review related to environmental and asbestos exposures, $63 million in commercial, largely related to auto bodily injury, and $51 million in homeowners. The chart on the right breaks down Allstate Protection auto prior year reserve strengthening of $643 million in the third quarter, which was primarily driven by non-customer claimant bodily injury claims. The total cost to settle these claims continues to be impacted by more severe accidents and higher medical and litigation costs.
Increases to commercial and homeowners insurance can also be attributed to these factors. Physical damage prior year reserve increases in the third quarter from property damage, collision, and comprehensive coverages, excluding catastrophes, were largely offset by higher subrogation collection estimates. Now let's move to slide 12 to discuss the drivers of bodily injury development and our claims operating actions to manage loss costs. Bodily injury severities have increased as the mix of claims shifted to more costly claim segments. The chart on the left shows the relative severity of bodily injury claims by type of treatment, major versus non-major, and whether the claim is unrepresented, attorney-represented or litigated. Major injuries have more expensive medical treatments, greater non-medical related damages, and often more attorney involvement.
As a result, paid severity for major injury claims and litigation, represented by the first bar on the left, costs approximately 3.9x the average paid bodily injury claim. Non-major claims, shown on the right-hand side of the chart, have less medical and other related costs and tend not to have attorney costs, so unrepresented non-major injury claims are roughly 10% of the average cost. Let me be clear, in all cases, we settle the cases for what is fair and equitable, regardless of attorney involvement. The table below the chart shows a significant shift from non-major claims that have below average cost to major injuries that are represented or in litigation in comparison to historical levels. This shift is partially attributable to more severe accidents. This shift to larger and more complex cases has also resulted in greater variability in paid and case reserve development patterns.
As part of our actuarial process, we review changes in claim development patterns to define an appropriate range of estimated outcomes based on weighing historical and more recent trends in the data. The chart on the right side depicts the value of two standard deviations to the average paid in case severity development over the last six report years. As you can see, this measure of variability has almost doubled over the last two years, resulting in a wider range of estimated outcomes. The third quarter reserving process showed a continuation of these development patterns. Therefore, we increased reserves for prior years to reflect the persistence of the trends in major injuries, increased settlement costs, and greater variability in case reserves. We're proactively responding to these trends by leveraging sophisticated models, increasing medical expertise, reviewing settlement processes, and assessing litigation risks.
Now let's move to slide 13 and briefly discuss physical damage loss costs, which continue to pressure profitability. Rising inflation and delays in third-party carriers' subrogation demands are driving higher expected severity in the property damage coverage, leading to an increase in the current report year variance from 12% to 17% when compared to 2021. The left side of the slide includes a chart we have shown before, which indexes inflation to year-end 2018 for a few of the main inputs to physical damage severity. While used car values are below their recent peak, which is a positive indicator, they continue to run more than 50% above pre-pandemic levels. Conversely, labor and parts prices continue to accelerate from the prior peak levels seen just last quarter. This continues to put upward pressure on severities in the near term.
The right-hand side of the page shows third-party subrogation demand dollars paid, again, indexed to the year-end 2018. Third-party demands are when our insured is in an accident and the claimant files a claim through their carrier rather than us. As the other carrier evaluates the claim, if the Allstate insured is wholly or partially at fault, they will reach out to us with a subrogation demand. We have recently experienced an uptick in the volume and severity of these demands compared to prior year trends and expectations. It's worth noting that a similar dynamic is also impacting our first-party collision coverages. We are demanding and receiving elevated subrogation collections from other carriers following the declines during the pandemic and backlog in claim settlements due to delayed repairs.
Shifting gears now on slide 14, the Protection Services businesses in the lower strategic oval are growing revenues and increasing shareholder value as we invest in future expansion. Revenues, excluding the impact of net gains and losses on investments and derivatives, increased 7.2% to $640 million in the quarter, primarily driven by a 12.2% increase in Allstate Protection Plans. Adjusted net income of $35 million for the third quarter of 2022 decreased $10 million compared to the prior year quarter due to increased severity on appliance repair for Allstate Protection Plans. In the absence of one-time restructuring expense at Allstate Identity Protection in the prior year quarter, as well as investments in growth.
Policies in force declined 5%, reflecting the expiration of protection plan warranties, primarily due to a high volume, low premium per policy retail account and overall decline in retail sales. Moving now to slide 15, Allstate Health and Benefits is also growing an attractive set of businesses that protect millions of policyholders. The acquisition of National General in 2021 added both group and individual health products to our portfolio, as you can see on the left. Revenues of $570 million in the third quarter of 2022 increased 1.2% to the prior year quarter as growth in group health and employer voluntary benefits was partially offset by a reduction in individual health. Adjusted net income of $54 million increased $21 million from the prior year quarter, reflecting a lower benefit ratio, lower restructuring charges, and increased revenue.
Shifting now to investments on slide 16, we'll review the performance and the portfolio risk and return position that we've taken given higher inflation and the possibility of a recession. As you may recall, we reduced our portfolio risk beginning in the fourth quarter of 2021. This included shortening the fixed income duration from 4.6 years to three years through the sale of bonds and use of derivatives, which resulted in a reduction to the portfolio's sensitivity to higher interest rates caused by increasing inflation. We also reduced our exposure to recession-sensitive assets through the sales of high-yield bonds, bank loans, and public equity. We maintain this defensive positioning in the third quarter, which continued to preserve portfolio value given ongoing market volatility, rising interest rates, and a further decline in public equity markets.
As shown in the table at the bottom left, our total return for the quarter was -0.8%, and year-to-date is -6.4%. While adverse market conditions negatively impacted the portfolio, we estimate our duration shortening mitigated portfolio losses of approximately $2 billion. These proactive actions and the broad diversification of our portfolio produced results that were better than the S&P 500 index, which is down 23.9% this year, and the Bloomberg Intermediate Corporate Bond Index, which has declined 11.8%. Our net investment income, shown in the chart on the left, totaled $690 million in the quarter, which was $74 million below the third quarter of last year. Performance-based income of $335 million, shown in dark blue, was $102 million below a strong quarter in 2021.
Three individual investments generated approximately 97% of the performance-based investment income in the quarter, including two sizable cash realizations. Excluding those assets, results for the broader performance-based portfolio were largely flat, with negative valuations in our private equity fund investments, which have a higher correlation to public equity markets, offset by increased valuations on other asset classes such as real estate and infrastructure. Our market-based income, which is shown in blue, was $50 million above the prior quarter, benefiting from reinvestment into market yields that are significantly higher than the overall portfolio's current yield. The table on the right demonstrates how our shorter duration fixed income portfolio is positioned to generate higher levels of investment income as we reinvest into higher interest rates.
Our fixed income yield has begun to rise and was 2.9% at quarter end, but is well below the current intermediate corporate bond yield of 5.6%. Now, let's take a few minutes to discuss Allstate's financial condition and capital position, starting with slide 17. Allstate's corporate organizational structure provides sources of capital to the holding company from multiple reporting entities and intermediate holding companies. We manage capital at all levels using economic capital, rating agency models, and regulatory requirements to guide decisions and maximize flexibility. We commonly report a view of capital that includes both statutory surplus and parent holding company assets. We prefer to dividend money up from subsidiaries to the holding company when possible, as it provides more financial flexibility for the organization while maintaining adequate capital levels in subsidiaries to support operations.
The chart on the left shows an overview of our capital position since 2016. As you can see, it grew substantially beginning in 2019, following strong results leading up to and during the pandemic. While the current level of $19.8 billion is approximately $6 billion lower than a year ago, this is largely made up of two specific items. First, $3 billion, or roughly half, is related to the sale of the life and annuity business, as represented by the first red bar on the chart. This transaction reduced our statutory capital as we sold the legal entities and significantly reduced our overall risk profile, freeing up an additional $1.7 billion of capital. We return this capital to shareholders as part of the current $5 billion share repurchase authorization.
The second bar reflects our cash returns to shareholders, excluding the impact of the life and annuity sale. Together, these factors reduced capital by $5.4 billion, with more than $4 billion going back to shareholders. The last red bar primarily reflects the impact of current auto insurance profitability challenges, which have resulted in a statutory loss and then changes in unrealized gains and losses on equity investments due to recent market volatility. We also added a line to this chart that represents our average capital from year-end 2016 through Q3 of 2021, excluding surplus related to the life and annuity businesses. Our current capital position of $19.8 billion is approximately $1 billion higher than this average, demonstrating that returning cash to shareholders after adjusting our risk profile in recent years of profitability has left us in a strong capital position.
The right-hand side of this page isolates holding company assets, a key component of our capital, relative to the remaining authorized repurchases and fixed charges. At the end of the third quarter, we had $4.5 billion in holding company assets. With $1.2 billion remaining on the current share repurchase authorization, we would still have $3.3 billion remaining in comparison to our annual fixed charges of $1.3 billion. We believe holding company assets and capital resources available from statutory operating companies provide significant financial flexibility as we continue to implement profit improvement actions and invest in Transformative Growth. Now, let's move to slide 18 to discuss Allstate's strong cash return to shareholders. Adjusted net income return on equity of 4.3% was below the prior year, primarily due to lower underwriting income.
Achieving our targeted combined ratios for auto and homeowners insurance will bring adjusted net income returns on equity back to our long-term targeted range of 14%-17%. Through the first three quarters of 2022, we've returned $2.8 billion to shareholders through $2.1 billion in share repurchases and $698 million in common shareholder dividends. Over the last year, shares outstanding have been reduced by 7.7%, providing more upside per share as profitability has improved. There's $1.2 billion remaining on the current $5 billion share repurchase authorization as of September 30th, which we expect to be completed in the second or third quarter of 2023, as we moderately slow the pace of our repurchases. With that context, we're gonna open up the line for your questions.
Certainly. Once again, ladies and gentlemen, if you have a question at this time, please press star one one on your telephone, and please limit yourself to one question and one follow-up. Our first question comes from the line of Greg Peters from Raymond James. Your question, please.
Good morning, everyone. I'm gonna focus the first question on reserves. I was looking at the information you provided on slide 12 and some of the earlier slides. I guess what we're trying to do is reconcile the charge that you took in the third quarter with the information we're getting from some of your peers, and then additionally trying to understand why the data that you're showing now here for the third quarter results, you didn't start to see it in the second quarter and make adjustments then. I mean, it's a long-winded question, but ultimately we're trying to get at, you know, is there risk of additional reserve charges going forward?
Greg, well, good morning. Thank you for your question. Let me provide just a quick overview, and then Jess can give you some more specifics. First, of course, obviously, we estimate what's gonna happen. That depends on trends for the numbers. We can't make a comment as to what other people's numbers look. People do reserving all sorts of different ways. We believe ours is highly precise, specific. We have external people look at it, and we put up the numbers when we think we need to put them up. Of course, this quarter we did increase it from prior years.
That's largely due to the injury trends that Jess saw, which have really been unfolding over the last couple of years. These are claims that take, you know, four years before you get 80% paid, so it takes a while before it develops. Jess, how do you wanna address this?
Yeah. Thanks, Tom. You know, Greg, what I would start with is, you know, I think it's important to be clear on one thing. At the end of every quarter, we record reserves at an appropriate level based on all the information that we have in front of us. You know, we did that at the end of Q3 and every quarter leading up to Q3. We followed the same process that we have in the past. It's a rigorous process. It leverages internal actuarial expertise, close collaboration with our claims team, and third-party reviews to analyze the most current data and assess the impact of that data on our reserves. You know, as I look at the quarter, you know, the variability that we've seen continued to come through in the data that we reviewed as part of our actual process.
Q3 data supported more recent trends and continued variability in reserve development. While these trends weren't new, an additional quarter of data did provide new insights into the persistent nature of the trends that have been emerging. Insights from actual claim development in the quarter led us to strengthen both prior year reserves and increase our report year 2022 ultimate severities. You know, as I take a step back, and think about how, you know, where we're at from a reserve perspective, you know, we record appropriate reserves based on what we know at the time. We used current data, and all known factors to establish the reserve. I'm confident in what we set up, and I think that, you know, the new insights that we gleaned in Q3 caused us to make the move.
Okay. Makes sense. I guess my follow-up question is on capital. You know, just looking at it from a macro perspective, I really don't remember in recent history a time where you guys have been growing your top line, you know, almost at a double-digit rate. You know, that by itself puts pressure on capital resources. You know, if we look at your capital position, it's outlined in the statistical supplement, and we see total capital resources having declined year-over-year due to a variety of issues. Just wondering if you can help frame how we should think about traditional metrics around premiums to surplus in the context of all the different moving parts.
I'll start and then Jess can add as well. Greg, first, we don't use the traditional. Well, I mean, we look at the traditional metrics like premium to surplus and as we go, but we're much more sophisticated than that. The way we allocate capital is from an enterprise standpoint and looks at specificity on risk levels down to the state-based level by line. For example, some people would blend their premium-to-surplus ratio for all Property-Liability products, auto and home, at the same. We don't do that. We think capital is much higher for homeowners insurance. That's why, when Mario went through our target ratios for homeowners insurance, they're lower than they are for auto insurance. Other people just assume that same.
We're very sophisticated in the way we do it. We manage it from an enterprise standpoint. When we, and the answer would be we have plenty of capital. Like we got tons of money, and it's not gonna do anything to our strategy. It has no impact on our future earnings power, which is, of course, what drives the company. We're in the middle of a massive share repurchase program, and we don't feel like we have to back off on it based on what we know about our business at the granular level. We feel very good about where we're at. We've generated good returns for shareholders by doing it that way.
I feel like using broad measures like that doesn't really reflect the economic reality that we're managing to. Jess, where would you go from there?
That's a pretty complete answer, Tom. I, you know, I think the important point is that the proactive capital management that really relies on our robust economic capital approach, which looks at risk on a granular basis and takes that information to understand capital needs at an enterprise level. You know, premium to surplus only looks at one dimension of risk and capital, and we use, you know, a more complete set of measures and metrics to establish capital levels as Tom laid out. I just think it's important. You know, we're cognizant of and we monitor regulatory capital requirements, rating agency capital benchmarks, all in our proactive capital management process.
you know, I feel the same way that Tom does, that we're, you know, we certainly have plenty of money to execute on our strategies and continue, you know, to implement our profit improvement plan. I, you know, nothing more to add, Tom.
Got it. Thanks for the answers.
Thank you. One moment for our next question. Our next question comes from the line of Elyse Greenspan from Wells Fargo. Your question, please.
Hi. Thanks. Good morning. You know, my first question is on the capital side of things. As you guys came to the decision, I guess, to more moderate your buyback, are you assuming that there's any dividends that you're gonna take out of Allstate Insurance Company over the next year? Within that question, I guess, you know, did you guys think about pausing the buyback program completely just to have more capital flexibility within that subsidiary?
Elyse, let me answer that and Jess, you can jump in. First, in terms of the dividends, we make sure that each of the insurance subsidiaries, the large ones, are appropriately capitalized based on what we think economic capital is just so with the rating agencies, whether it's AM Best or others think we should have in those, and then what regulators want. We're really well capitalized with those. To the extent there is extra capital in those, we then move that out of the insurance companies into the holding company, as Jess pointed out, because that gives us more flexibility. As we look forward next year, it'll depend how much money we make. Like, I think, you know, in some of the scenarios, like we have really strong earnings power.
When you look at our profitability of our auto insurance, we think it's headed up. We think there's plenty of earnings power. Whether that and where we think risk is and what we need to do with risk will depend on the overall enterprise risk portfolio. For example, not covered in some of these things, we dialed down the risk for our investment portfolio late last year, as a percentage of our total enterprise capital, because we didn't think it was good risk return trade-off. We're constantly managing, where do we wanna move capital? Where do we wanna make sure we get a good return on it? We don't feel like we are capital constrained at all. Did we consider shutting the program down in total? No.
We think we have plenty of money. I mean, this is a $5 billion massive share repurchase program, a large portion of which was funded because we sold the life insurance company, and we said we should give money back to the shareholders. You know, we have a historical track record of doing this extremely well. We're a top decile amongst the S&P of providing cash returns to shareholders. And we do that without putting our customers at risk or the company at risk. We feel like we're in really good shape.
Okay, thanks. My follow-up, I guess would be, right, you talked about that you don't manage to premium to surplus, but I know one of your peers has mentioned looking to write their, you know, all business to a 3-1, home to, you know, 1.5-1. I'm not sure if you have frames of reference that you look at for your businesses. If, you know, rating agencies and regulators, what are they looking at? Are they holding you to a 3-1? Or are there other metrics that they're holding you to relative to premium to surplus or something else?
We've had, essentially, say, from certainly my standpoint, no conversations with regulators about our capital levels because we're so well capitalized. Let me start there. That's been true forever, and it will be true far into the future. The regulators really, we're so far above their standards that it's not really been a conversation. I can't speak to how other people look at their premium to surplus ratio. I think, some of our competitors who don't make money in homeowners, should have, even more than we do. Because when you're looking at your capital, it's, okay, what do we think the risk is? Where is the risk of loss? You also want to factor in your earnings power.
If you're losing $0.10 on a dollar on a line of business, then you have to hold more capital than if you're making $0.10 cents on a line of business. You know, I think it's all very idiosyncratic to a specific company. We factor all of those things in by state, really, and even sometimes looking down at different components of the state, to decide what price we should get for a customer, from customers, how much business we want to write, and then how much capital we have to keep in the company. I feel very good about where we're at.
Okay. Thanks for the color.
Thank you. One moment for our next question. Our next question comes from the line of Brian Meredith from UBS. Your question, please.
Yeah, thanks. A couple of them here for you. First, Tom, just curious, this is, I think, the first quarter in a while that I've seen auto PIF actually decline somewhat sequentially. Is that due to some of the actions that you're taking in California? Or, you know, is it just in general, the price increases you're taking? Should we expect the, you know, PIFs to kind of decline here for a couple of quarters?
Well, the PIF decline was of course in the Allstate brand versus in total, because we went up in the independent agent channel. It's intentional. I would tell you that it's actually the decline was less than we thought it would be. When you look at historical price sensitivities on how much, what your customer retention is, our retention has held up better than you would think from historical trends. It's hard to do attribution, you know, down to the specific item. When you look at it, we'd say first, you know, the competitive position, our competitors are also raising rates. As we raise rates, we thought more people would leave, but less did. It could be because our competitors are also raising rates.
Consumers still have a fair amount of cash in their bank accounts, so that helps. They also know that their houses and cars are worth more, so it makes sense to them that they should have to pay more for insurance when explained to them. I think that's the value of our Allstate agents at this point. They're out working hard to make sure our customers understand why the prices are going up. As Mario mentioned, they'll help them to figure out how they get the right price. This is a case where like our Milewise product is really helpful for people because, if you're, let's say you're a senior citizen, you don't drive much, then you should go to Milewise and save a bunch of money and not pay more.
When you look through all those together, we expected our auto PIF to go down more in the Allstate brand than it did. We're happy that we're keeping the customers because with the price increases we've put through, that'll be good shareholder value creation when we start earning the rates.
Great. That's helpful. Thank you. My next question, I'm just curious, and I think this question has been asked in prior quarters, but when you're pricing your auto insurance and homeowners insurance now, what type of loss trend are you expecting in the future? Are you expecting inflationary trends to moderate here, or do you think they're gonna stay relatively high here for a while?
Mario, will you take that?
Yeah, sure. Good morning, Brian. Yeah, I think, Brian, in terms of what we're seeing, you know, we're factoring in obviously the inflation we're experiencing currently, but also projecting it going forward so that we can reflect the full cost of loss costs prospectively into our prices. So, yes, we're not making any kind of significant assumptions around, you know, a deceleration in inflation going forward. You know, given the current inflationary environment, that's why we made the statement that we expect to continue to take rate increases, certainly for the balance of this year, but into next year.
That's really a reflection of the environment we're operating right now and the continued elevated level of inflation, which we need to kind of catch up with and then surpass going forward. We're not assuming any, as I said, significant reduction in inflationary trends going forward.
Great. Thanks for the answers.
Thank you. One moment for our next question. Our next question comes from the line of Tracy Benguigui from Barclays. Your question please.
Thank you. All right. I'm used to asking this question for casualty line writers. Given the consecutive adverse reserve development charges, have you worked with any external actuaries to review reserves? If so, what is your management estimate relative to central estimate?
Tracy, we always work with external people in looking at our reserves. We obviously have Deloitte & Touche, our auditors, but we also have an external actuary called KPMG, which provides the statutory reports for our regulators. We look at all of their stuff. We just had a detailed review with Deloitte & Touche, a couple of weeks ago, and their view ties closely to ours.
Do you have a management estimate above the central estimate at the moment, or are you closer to the central?
You know, we don't put ranges in the financials. We put up what we, as Jess said, like we put up what we think the future liability is, and we pick a number, and that's where we do it. We're comfortable with the number. That number is very close to what our external participants or external help thinks, and thought historically, by the way. It isn't like. We both have very similar views at the end of the second quarter, end of the first quarter and end of the third quarter. They believe that the actions we've taken are appropriate.
Got it. Just going back to capital management, given you manage capital more efficiently at the opco level and the way you like to hold cash at the holdco level, so you could flex that up and down. I'm just curious, when was the last time you downstreamed capital to the operating company? Have you used that lever often?
Good question. I don't remember. Certainly in the last decade, I don't remember having done that at all from down to Allstate Insurance Company. Would we move money into the health and benefits companies because they were growing or do we have to? We do move money around. If you said, if you really talk about Allstate Insurance Company as the largest business we have, I don't remember anything in the last 10 years. Jess or Mario, do you have any other perspective on that?
I don't have anything more than that. You know, I think we did move some capital down into the life company at one point, Tom, but I think that, you know, certainly is no longer an issue. But that's the last thing that I remember. Mario?
Yeah, I would concur. The last time I remember any meaningful movement of capital down into an operating company would have been during the financial crisis, which obviously was a while ago. As Tom mentioned, we moved capital around, but nothing in terms of shortfalls within any of the insurance companies.
Right. I'm not thinking about this the right way, where you have capital at the parent company that you could flex it up and down. In your slide, when you talk about your fixed charges, is there a multiple that you wanna keep, like through time to this, SAP being a minimum level?
No, we don't manage it that way. I just showed the level of cash we would have after we used deployable capital to finish the share repurchase program. You can see it's still well above our fixed charges, which we intentionally manage to keep at a modest level, which is even though we increased the dividend by about 50% a year ago, we try to give a lot of money back to shareholders through share repurchases. We don't have a multiple. If you had a multiple and you also have to factor in how much money you're gonna make over the next 12 months, and obviously our fixed charge coverages historically have been terrific. We don't have a like don't go below this because we've never even been close.
Thank you.
Thank you. One moment for our next question. Our next question comes from the line of David Motemaden from Evercore ISI. Your question, please.
Hi. Thanks. Good morning. I think it was Jess who spoke about just, you know, the mix shift that you're experiencing on the BI claim side, to more expensive claims, and as part of the reserving process, you weight both historical and more recent trends in the claim development patterns in the data. I guess I'm wondering, after the changes that you've made this quarter, are you I guess, how much are you weighting more recent experience? Is it 100% weight on these trends that you're seeing? Or, you know, is there still some weight being placed on more historical experience?
David, let me answer that, and Jess, you can add anything else to it. First, the reserving process uses all types of statistical analysis, triangles, link ratios, all sorts of different things. There's no real, like, just 1 percentage thing. We look at it by state, by line of business, by coverage. It's sliced and diced a whole bunch of ways. There's no really simple way to answer that. I think what I would say about, you know, you're trying to get comfortable with the reserve, it really starts, you got to go back to say, well, what's happening in the world?
During the pandemic, we noticed when people weren't on the roads, people were driving a lot faster because they could zip around and there was no traffic and off they went. Once we get through the pandemic, at least in our data, we see people still driving pretty fast, but as a result of that, you have more severe accidents. That trend appears to be holding. We thought that trend might come down because when the little ones are sort of bumps and scratches and stuff which happen in congested traffic. Today, those, as you saw from Jess's numbers, have not gone back up and the major ones have not gone down. People are just driving faster and hurting people more.
You have to figure out, okay, well, what are you going to do to take care of it? Those are really complicated cases. I mean, people have surgery, they have all kinds of services, and those services are more expensive, and they take longer to develop. You know, if you're really severely injured, it could take six months, nine months, two years before you really figure out how you get back to where you should be. It costs a lot of money and takes a lot of time. Those cases develop over a longer period of time.
It isn't so much that it's just that we use the same processes, procedures, but as these things develop, it really comes back to, you know, our customers are just in a lot more severe accidents and people are getting hurt. We need to make sure we have the liability up to cover that. That's what we did this quarter. We said, okay, this is really continuing. Most of these are still severe. As they develop, then you have to put the money up.
Got it. Okay. It sounds like, you know, you had assumed that the mix would normalize somewhat away from some of these more severe accidents. I guess now the assumption is that there is gonna be no mix away from these more severe accidents, and that's sort of the new normal. Is that correct?
Yeah, it's not just one item, I would say. You can't pin it on just, geez, there's now more majors. Majors are harder to estimate. They're taking longer to settle. There's more legal costs associated with settling those, and so you have to factor that in. There's a whole bunch of factors that relate to it. We look at it, we're comfortable we've put up the right amount of money. You know, what other companies do and what they're reserving are. Some people use less specific processes than we do. You know, some processes react faster or slower to trends in the marketplace.
The important thing is we use the same processes, have external views, and we all think this is the right amount of money.
Got it. Okay. Maybe just a quick numbers follow-up here. You know, see that you guys are now assuming a bodily injury severity of 12%. I guess I'm just wondering, what was the report year incurred severity on bodily injury for 2021 after the changes that you've made?
We have not broken out the reserve changes by prior years. We won't break those out until we publish the 10-K. I mean, is that right, correct, Jess?
That's right, Tom. We don't disclose the splits, and we don't have that.
Suffice it to say, David, that it's higher than it was before.
Yeah, I mean, I was looking at a 5%. I'm just trying to get a sense for, you know, the compound, you know, if we say, okay, up 12%, but the base does matter. I think the base was set at 5% in the first quarter for all of 2021, which has since been changed. I was just trying to get a sense for where that's gone. I guess I'll look in the 10-K for that. Thank you.
Okay. Why don't we take one more question.
Certainly. One moment for our final question. Our final question comes from the line of Yaron Kinar from Jefferies. Your question, please.
Thanks, good morning. I'm gonna go to my specialty of beating dead horses here, if I can. On capital, do you expect to deploy some of the holdco liquidity into AIC over the coming year?
No.
Okay. Shifting more to the auto and home side. You guys shifted the exclusive agent comp structure to be more weighted to new business. Now that you're kind of maybe taking a little bit of a step back on growth and really focusing more on fixing the margins, how is that playing out with the agent comp structure with your conversations with them? I'm assuming there can be a little bit of murmurings and rumblings around that. How are you handling that?
Well, Yaron, I'm glad you asked. Embedded in all of this, there's a lot of good news on Transformative Growth that we really don't get a chance to talk about. One is that what you talked about, which was expanding customer access, is the second key lever. And that included selling direct under the Allstate brand at 7% less than was sold to Allstate agents. There was some concern among investors as to would the agents walk away, and would you have a decline in volume there? The answer is no. That is the underlying assumption that they would continue to be focused on getting more new customers given what we did to the compensation plan was true.
If you look at new business from the Allstate agents, you can see that's where it was a year ago, even though there are fewer Allstate agents out there. If you look at the retention numbers, as I mentioned, I think our agents are doing a great job for us, talking to customers whose price changes. We feel good that part of the expanding access, all of our underlying assumptions prove true. We also have really improved our web-based and the call center close processes, so we're getting much better at selling through those two vehicles. We obviously dialed advertising way down, in this year, because we don't wanna take on new business and then have to raise the price 15% or 20%, the first time.
While you don't see the benefit of those improved processes come through new business, but when we get auto profitability improved, we feel good about the underlying assumptions we made in Transformative Growth and our progress in making those a reality. We're feeling good about where that's headed. Thank you all for dialing in. As we move forward, you know, we have a couple of things in front of us. One, we have to improve auto insurance margins while making sure we continue to invest in Transformative Growth so that we can grow market share. Then continuing to expand our other Protection Services businesses, which also had a great quarter. Thank you, and we will talk to you in December.
Thank you, ladies and gentlemen, for your participation in today's conference. This does conclude the program. You may now disconnect.