Good morning. My name is Ian, and I will be your conference operator today. At this time, I would like to welcome everyone to The Hartford third quarter 2015 financial results conference call. All lines have been placed on mute to prevent any background noise. After the speaker's remarks, there will be a question and answer session. If you would like to ask a question during this time, simply press star then the number one on your telephone keypad. If you would like to withdraw your question, press the pound key. Thank you. Sabra Purtill, Head of Investor Relations, you may begin your conference.
Thank you. Good morning and welcome to The Hartford's webcast for third quarter 2015 financial results. Our third quarter financial results, news release, investor financial supplement presentation, and 10-Q were all released yesterday afternoon and are posted on our website. Our speakers today include Chris Swift, Chairman and CEO of The Hartford, Doug Elliot, President, and Beth Bombara, CFO. Following their prepared remarks, we will have about 30 minutes for Q&A. Just a few notes before Chris begins. Today's call includes forward-looking statements as defined under the Private Securities Litigation Reform Act of 1995. These statements are not guarantees of future performance, and actual results could be materially different. We do not assume any obligation to update forward-looking statements, and investors should consider the risks and uncertainties that could cause actual results to differ from these statements.
A detailed description of those risks and uncertainties can be found in our SEC filings, which are also available on our website. Our presentation today also includes several non-GAAP financial measures. Explanations and reconciliations of these measures to the comparable GAAP measure are included in our SEC filings as well as in the news release and financial supplement. I will now turn the call over to Chris.
Thank you, Sabra. Good morning, everyone. Last night, we announced our financial results for the third quarter. While we delivered strong underlying performance in our commercial lines and group benefits businesses, we did experience some headwinds in several areas, resulting in a decrease in core earnings. Lower net investment income, prior year development in commercial lines, and higher cats and loss costs in personal lines were the primary contributors to a 19% decline in core earnings per diluted share. Net investment income was down 10% compared to the third quarter of last year. This decline is mainly due to hedge fund performance in the quarter. However, year-to-date results remain ahead of our outlook for alternative investments in aggregate. We also had some notable achievements, including increasing 12 months core earnings return on equity to 9.1% and growing book value per diluted share by 8%.
In mutual funds, net flows year to date were at the highest level we've seen since 2010. In Talcott, our execution remained steady, where we continue to successfully manage the risk of the book and return capital to the holding company for more accretive uses. Doug and Beth will cover our operating results in more detail, but I wanted to share a few thoughts on our results. In commercial lines, the underlying combined ratio, excluding cats and prior year development, improved one point over prior year. This result was driven by small commercial and specialty commercial, which delivered underlying combined ratios of 86.8 and 99.1 respectively, both better than prior year. The underlying results in middle market were steady versus last year. Prior year development in general liability and commercial auto contributed to the deterioration in the total combined ratio.
Personal lines cat losses were elevated versus prior year, although below our expectations. We saw and reacted to recent signs of increased auto frequency, which impacted our combined ratio. We also experienced elevated not-cat property losses compared to the prior year. This business has always required that we actively utilize data and analytics to monitor trends and make adjustments, and that is exactly what we're doing. Group Benefits delivered another strong quarter with core earnings margin of 5.5%, improving a full point from prior year. The year-to-date margin of 5.9% is the best we've seen since 2008. The Group Benefits team recently signed a renewal rights agreement for AIG's under 100 lives employer segment, which aligns with our objective to grow in the small and middle market areas.
Looking ahead, we are focused on executing on our strategy, even as we face a more competitive and dynamic industry environment. Let me share some examples of these efforts. Relative to distribution, in commercial P&C, we are on track to meet our goal of adding sales and underwriting talent in targeted geographies like the West and Midwest regions, which will help us be more responsive to customer needs. In Group Benefits, we increased the number of sales representatives in the under 500 lives employer segment to complement our national account segment. The goal is to have a more balanced portfolio of small to mid-sized accounts while maintaining our strength in national accounts. Early results of these efforts are encouraging.
In early October, Doug and I, along with a number of our senior leaders, attended the annual CIAB conference, where we met with more than 80 of our industry's top agents and brokers. These discussions confirm that the work we have been doing to expand our product offerings and risk appetite is beginning to pay off. The feedback we received also affirms our view that The Hartford has strong and growing momentum with many of these organizations, which will serve us well on our strategic journey. Talented people are the engine behind The Hartford's success, and I am especially proud of the care and commitment our employees show towards customers in their time of need. During the recent wildfires in California, our claims teams were able to quickly inspect 100% of claims, meet in person with all impacted customers, and get them into temporary housing.
No small feat, considering the impact of these fires. This is exactly the kind of claim service that differentiates us and earns The Hartford a 4.8 out of five-star ratings with our customers. Before turning the call over to Doug Elliot, I want to emphasize that while we had some challenges in the quarter, I am pleased with the progress that The Hartford has made during the year. Our year-to-date results reflect a 17% increase in core earnings per diluted share over the prior year, and we are focused on finishing the year strong. I am confident that we have the right people, capabilities, and underwriting discipline to succeed even in an increasingly competitive marketplace as we remain focused on delivering shareholder value by increasing ROE and book value per share. Now let me turn the call over to Doug Elliot. Doug Elliot?
Thank you, Chris Swift, and good morning. I'm going to provide additional details on the operating results of our property and casualty and group benefits business units. First, let me begin with a few observations about the market. The competitive landscape in commercial lines and group benefits is slightly more pressured than we experienced over the last four quarters. Markets remain largely rational, but there are more clear signs of aggressive new business pricing with some loosening of terms and conditions, particularly in commercial property. We continue to find opportunities to acquire adequately priced new business while remaining disciplined in our risk selection approach, mainly through more intense sales execution in our local markets with agents and brokers. In personal lines, competition is generally consistent with prior quarters. We continue to see opportunity for growth in the direct channel and with our differentiated AARP agency offering.
Price competition in the traditional agency channel remains the norm, driven by comparative raters. Third quarter core earnings in commercial lines was $216 million, with a combined ratio of 94.5. This was an earnings decrease of $52 million versus third quarter 2014, primarily driven by adverse prior period development in commercial auto and lower net investment income. The underlying combined ratio, excluding catastrophes and prior period development, was 91, improving one point from third quarter 2014, largely driven by continued margin expansion in workers' compensation. This improvement reflects the solid foundation we've built in recent years across our commercial businesses through rigorous underwriting and pricing discipline. Renewal written pricing in standard commercial lines was 2% for the quarter, down 1% from second quarter 2015 and down two points from third quarter last year.
Commercial auto continues to achieve high single-digit price increases as we and the industry address weak returns in the line. Pricing in other lines is more competitive, particularly middle market. In small commercial, written premium grew 4% in the quarter. Strong policy retention has continued, providing a nice tailwind as new business was up more modestly at 2%. The underlying combined ratio was 86.8, improving seven tenths of a point from a year ago due to better workers' compensation margins and favorable non-cat property losses. We continue to work on distribution initiatives with our agency partners to drive new business growth. Although the market is competitive, our business model is performing well, and we see the opportunity to deploy our capabilities to gain market share. In middle market, we posted a somewhat mixed quarter with an underlying combined ratio of 93.8, three tenths higher than third quarter 2014.
However, the overall combined ratio was 102.5, 8.8 points higher than last year due to adverse prior period development, primarily in general liability and commercial auto. The development in general liability was driven by a large loss in older accident years. In commercial auto, we continued to see increased severity on a relatively small number of losses, mainly from accident years 2010 to 2013. In several of these claims, there's been a pattern of significant buildup in medical costs without ongoing notification to us. It's important to note that our reserving estimates assume that these trends will persist into more recent accident years as well, but that certainly does not reflect the intensity of our actions to improve performance in the line. We've been working throughout the year to improve claim, product, and underwriting execution, and thereby better manage outcomes on the current accident year.
This includes implementing new underwriting tools and guidelines that we expect to reduce our exposure to these high-severity risk profiles for both new and renewal business. We continue to increase rates and improve our pricing segmentation to better address loss cost trends. We believe that we have begun to mitigate these trends in the current accident year, but we'll only make that call as the data develops. Moving to the top line in middle market, our metrics continue to show that we're making effective decisions to retain well-priced business and acquiring new business when meeting our underwriting and rate adequacy thresholds. Written premium growth was 2%, driven by strong renewals in marine, new business growth in large property and construction, and pricing increases in commercial auto.
We remain committed to improving and expanding our non-workers' compensation lines, recognizing that we must be thoughtful in our approach given recent market conditions. In specialty commercial, the underlying combined ratio is 99.1 versus 105.1 in the prior year. The six-point improvement was driven by better loss performance in bond and financial products. Last year, bond's accident year losses included a large loss, while this year has returned to our historical performance. Top line growth for specialty was 7%, driven mainly by strong account retention and renewal premium in national accounts, and to a lesser extent, audit premium adjustments. In personal lines, core earnings was $17 million for the quarter versus $71 million last year. Of the $54 million decrease, $23 million was due to higher catastrophe losses versus third quarter 2014. Our total catastrophe losses this quarter were below our expectation.
However, third quarter of last year was even more favorable, resulting in a challenging year-over-year comparison. Our most significant events this quarter were the California wildfires, which resulted in $56 million of pre-tax losses. The underlying combined ratio of 95.6 deteriorated 4.7 points from last year, driven by increases in auto frequency, non-cat homeowner losses, and marketing expenses. Let me provide more detail on each of those items. First, auto frequency increased 3% in the quarter after several quarters of flat to negative indications. We had anticipated some increase in our frequency this quarter, knowing that we had a very favorable frequency change in third quarter 2014. On a trailing 12-month basis, our frequency change is below 1%. There have been quite a few broad-based data observations, such as lower gas prices, improving employment conditions, and increased miles driven that point to roads being more congested.
It is very difficult to correlate this information with our specific book of business, which we generally find to be less susceptible to these factors due to our weighting towards mature drivers. However, we all travel the same roads, and our customers are not completely insulated from these conditions. As a result, we have reflected a slight increase in frequency with our loss estimates and pricing assumptions. The months ahead will provide more data, and we will continue to adapt our pricing and marketing strategy accordingly. Second, this quarter we saw an increase in non-cat homeowner losses, primarily from fires and water damage, although partially offset by favorable weather losses. These types of losses tend to be uneven from quarter to quarter. Recall that fire losses in the second quarter of 2015 were at their lowest level in seven years.
We've examined the loss profiles and at this point, have not seen any particular patterns in our data. Finally, direct marketing spend is up this quarter versus third quarter 2014. AARP Direct Auto has continued to perform well, and we've been planning for increased acquisition efforts in the back half of this year to take advantage of our recent product improvements. We're especially focused on driving online activity to our contact centers, where our sales teams provide outstanding counseling services and have demonstrated the ability to convert prospects to customers. Total written premium for the quarter grew 1%, including 4% growth in AARP Direct and 8% growth in AARP Agency. In other agency, written premium was down 10% versus the third quarter of 2014. Our efforts to engage with highly partnered agents who value the differentiated products and services we offer are continuing.
Shifting over to Group Benefits, core earnings in the third quarter was $47 million, up 24% over the same period in 2014, achieving a core earnings margin of 5.5%. The increase is primarily attributable to top-line growth and a lower disability loss ratio compared to prior year. Earned premiums, excluding association financial institutions, was up 3% in the quarter, driven by growth in our employer Group Life and Disability lines. For the quarter, fully insured ongoing sales was up 7% to $61 million. In addition, our employer group business continues to maintain strong book persistency, around 90% on a year-to-date basis. We're having success in competitive markets. Our flow of new business opportunities is strong, and we're working in a disciplined yet aggressive manner to win new accounts. Long-term disability continues to be the most competitive line, despite having underperformed across much of the industry in recent years.
Our disability book of business is performing well following several years of underwriting and pricing actions, and we're maintaining our steady course. Our Group Benefits value proposition has been significantly enhanced over the last several years, with improvements to our service and claims experience and the addition of a robust voluntary platform. We are well positioned to expand this business and are confident that we have built momentum across our target markets. With that, let me conclude my comments. This is a quarter where we experienced some volatility across our Auto and Property lines, and we're very focused on taking appropriate actions to strengthen performance in these areas. We continue to invest in product expansion, deepen our distribution capabilities, and deliver outstanding service to our customers. Markets are competitive, and we're responding with discipline and focus to stay on track with both near-term actions and our long-term strategic objectives.
Let me now turn the call over to Beth.
Thank you, Doug. I'm going to briefly cover the other segments, the investment portfolio, and our capital management actions before we turn the call over for questions. Mutual funds core earnings were flat with the prior year quarter as lower fee income due to the decrease in market levels from June 30th was largely offset by lower distribution and other operating expenses. Total AUM was down about 5% from September 30th last year, mostly due to Talcott-related AUM runoff. Excluding Talcott, AUM declined 2% due to the market decline this quarter. Fund performance remained solid with 58% of all mutual funds and 74% of the equity funds outperforming peers over the last five years, helping to improve net flows to a positive $307 million for the quarter and almost $1.1 billion year-to-date.
Talcott posted stronger than expected core earnings of $107 million, down from $122 million last year due to decreased fees and investment income, partially offset by lower expenses. The decrease in fees reflects the continued runoff in Talcott with an 11% decrease in variable annuity contract counts over the last year. Investment income was impacted by several factors, the largest item being lower limited partnership income, which was partially offset by higher bond calls and make-whole premiums than last year. In addition, Talcott's operating expenses and costs for contract holder initiatives were lower than the prior year. As I mentioned last quarter, Talcott paid a dividend of $500 million in July, in addition to the $500 million dividend paid in February. We expect to pay another $500 million dividend to the holding company in early 2016.
Corporate segment third quarter 2015 core losses of $63 million were up slightly from $58 million in 2014, which included a $7 million insurance recovery. Excluding this benefit, core losses continued to decline due to lower interest expense as a result of debt repayments over the last year. Turning to investments, the credit performance of our portfolio remains strong. Although impairments rose slightly to $40 million before tax from $15 million in the third quarter of 2014. About half of the impairments resulted from the decision to sell some lower credit quality securities in the high yield and emerging market portfolios. Our annualized portfolio yield, excluding limited partnerships, was 4.2% in the quarter, up slightly from 4.1% in both second quarter 2015 and third quarter 2014.
Our consolidated portfolio yield held up well despite the headwind from low interest rates as the impact of lower reinvestment rates was partially offset by the benefit of make-whole call premiums on bonds, prepayment penalties on mortgages, and other non-routine items. These non-routine items, which are largely correlated to the continued low interest rate environment, increased the year-to-date investment yield by about 10 basis points. Excluding these items and limited partnerships, our year-to-date annualized portfolio yield was 4%, down about 10 basis points from a year ago. The P&C portfolio also is experiencing a similar pattern, declining this quarter to an annualized yield of 3.7%, excluding limited partnerships, and it did not benefit from non-routine items to the same degree as the Talcott portfolio. Although P&C new money yields averaged 3.8% due to wider credit spreads, we continue to expect low reinvestment rates to pressure P&C and consolidated portfolio yields.
As Chris mentioned, lower limited partnership returns negatively impacted our results, particularly in commercial lines and Talcott. This portfolio had an annualized return of 3% this quarter. The year-to-date return is 10%, still well above our 6% outlook. This portfolio, which totals about $3 billion, is roughly 60% private equity and real estate partnerships, which have earned about 18% year to date, and 40% hedge funds, which were negative in the quarter and year to date. Hedge fund performance was adversely affected by the global decline in equity markets and the volatility in currencies. Looking at the fourth quarter, we can't accurately predict hedge fund performance because of their idiosyncratic nature. However, we do have some early visibility into real estate and private equity partnership returns, which are reported on a one-quarter lag.
So far, we expect fourth quarter real estate and private equity income to decline from the strong recent performance as their valuations will be impacted by the third quarter 2015 decline in equity markets. Considering this, we currently estimate that fourth quarter limited partnership and other alternative investment income will be lower than the third quarter 2015 actual result. We have always expected limited partnerships to have more volatile returns than the rest of the general account, and we have sized them accordingly. We also expect returns on the portfolio to be higher over time to compensate for this volatility, which we certainly have seen over the past several years. Turning to our capital management program. As you know, we increased our equity repurchase authorization in July. During the quarter, we repurchased approximately 6.5 million shares for $300 million.
In addition, since the end of the quarter through October 23rd, we repurchased an additional 2 million shares for $94 million, leaving roughly $1.7 billion remaining under the equity repurchase authorization that expires at year-end 2016. In addition, as a reminder, we intend to repay an outstanding $167 million debt maturity in November. I would note that our rating agency debt to total capital ratio was approximately 26.9% at the end of September, down from 28% at year-end. September 30th, 2015, book value per diluted share, excluding AOCI, rose to $42.99, up 6% from year-end 2014 and 8% from September 30th, 2014, reflecting positive net income after dividends to shareholders and the accretive impact of the equity repurchase program. To summarize, core earnings totaled $360 million for the quarter, or $0.86 per diluted share.
Even with some of the challenges we experienced in the quarter, we did improve core earnings ROE and grow book value per share. Over the last 12 months, we have achieved a core earnings ROE of 9.1%, compared with 8.2% at third quarter 2014, and in the range of 8.7%-9.2% provided to you at the beginning of the year. In addition, our year-to-date core earnings are $2.81 per diluted share, up 17% from 2014. I will now turn the call over to Sabra so we can begin the Q&A session.
Thank you, Beth. First of all, to those of you listening on the webcast, I want to apologize for some of the sound interference that you are getting. I just want to remind you all that the replay of the call will be available, and there's also a transcript. However, given the irritation of the skipping, what I want to do is repeat the dial-in number so that you can hear the Q&A session clearly. For the dial-in, it's area code 877-685-7362. For those international, it's area code 937-999-2389, and the conference ID is 505-761-63. I repeat, it's 505-761-63. If you dial those numbers, you should be admitted automatically to the call. Again, the dial-in is 877-685-7362.
Just a reminder to those of you who are on the call right now that we request that you limit yourself to two questions, after which you're welcome to re-queue for additional questions. We appreciate your adhering to this so that we can have as many people as possible ask their questions on the call. Ian, could you please repeat the instructions for Q&A?
At this time, I would like to remind everyone, in order to ask a question, please press star, then the number one on your telephone keypad. We'll pause for a moment to compile the Q&A roster. Your first question comes from Brian Meredith. Your line is open.
Yes, thank you. A couple questions. First, can I just dive into the whole frequency situation just a little bit more here? Perhaps you can give us some more color on maybe states that it's coming from. Is more of it coming from AARP versus the other agency? Is it balanced? Anything that you're seeing?
Brian, this is Doug. Let me start out by saying, clearly July and August were a little heavier on the frequency side than September. We think some of that is attributed to summer travel. We're very interested in how October will play out. A very important trend to stay on top of. There is obviously a state dynamic. If you look at miles driven that DOT has shared, there clearly is a West Coast and across the South dynamic of more miles driven over 2015. We're feeling some of that in our state-by-state analysis. Next point I would share is that as we look at the top 10 states from a personal lines liability perspective, essentially we're in line with market share. We're not over lined in any of the bigger states that are driving some of this frequency change, namely California.
We're aware of those states where we are, our work continues. As I look at this dynamic, we've had very favorable rolling 12 frequency changes over the past couple of years. We have seen a little bit of activity, certainly into the third quarter. We reacted to it in our financials. We'll stay on top of our trends into October, November, we're going to make sure that our pricing analytics are correlated with the trends we're seeing in our loss analysis.
Hey, Brian, it's Chris. Let me just add a larger context. You guys jumped into it right away. The AARP relationship has been a wonderful 30-year relationship for us, it's very strategically important, we pride ourselves on working together with them through their members and have coordinated strategies and outlooks that 80% of our personal lines book is AARP, whether it be direct or agent. Over a longer period of time, it has performed very well. When Doug refers to increases in frequencies, you got to think in terms of a lower different base than sort of a larger mass market personal lines company. As much as we reacted to it, there is a basis difference that our book is still relatively outperforming with lower frequencies from what we could tell at this point in time.
Got you. Thanks. My next question, I'm wondering if you could just dive a little bit more into the commercial auto adverse development. What are you all doing right now to make sure that's not going to continue here going forward? It's kind of been almost a quarterly event that we've seen recently.
Brian, again, this is Doug. What we've seen most recently is some pressure between months 24 and 42 in our triangles. As I mentioned in my commentary, primarily in years 2010 to 2013.
We're seeing late emergence of medical information as these claims are unfolding over the couple of years. A couple of things are happening. One is that inside our claim operation, we're looking at diagnostics that will try to get our arms around those claims as quickly as possible. I gave an example on the call moments ago. Sometimes we don't get these till late into their life, but we are looking at our claim practices and wondering whether we can make adjustments there. More importantly, inside our underwriting, we are buckling down and making sure that we've got all the disciplines around driver behavior, past driver behavior, and we're looking at new class plan tweaks, et cetera. We're working both ends of this.
Okay.
I feel like we have put our arms around the issue, and it's something that, although has been frustrating, I feel like we're on top of.
You're not seeing any increase in frequency in the commercial auto space necessarily?
No, this has not been a frequency issue. This is more on the severity side for sure.
Great.
Hey, Brian?
Yeah.
Again, just to context, this is primarily a middle-market phenomena. Our middle-market premium is a little over $200 million on an annualized basis. It's irritant as we feel, but we're trying to work through it the best we can, as Doug said. I think the other context, too, is just the larger balance sheet. You know we're one of the few companies that has disclosed our carried reserves in excess of our actual point estimates. At year-end 2014, that was about 3.5%. We continue to grow that here in 2015. This little minor irritant in relation to the larger reserve positions that we have for adverse deviations in the future, we're very satisfied with, particularly us continuing to grow those positions here in 2015.
Brian, I guess the last point I'll just add over the top is, as you know, we continue to drive rate into that book of business, right? I've given you underwriting claim actions, but we're not forgetting about the need for a rate to deal with that increased severity.
Great. Thank you.
Operator, we're ready for the next question.
Your next question comes from Michael Nannizzi. I apologize, from Gary Ransom. Michael Nannizzi with Goldman Sachs, your line is open.
Thank you. Can you hear me?
Yes, we can, Michael.
Oh, great. Doug, can we dig a little bit more into the auto underlying combined deterioration year-over-year? Can you quantify, it looks like the expense ratio was higher, we can't really see that breakdown at the product level. Can you just help us understand what the breakup between the expense ratio and this frequency deterioration was?
Sure, Mike. Let me see if I can maybe walk third quarter 2014 to third quarter 2015 for you, just to take you through that. If we look at an ex-cat basis, I'm going to walk from 90.9% to 95.6%.
The frequency dynamic in our auto book, both PD and liability, roughly 2.5 points of that 4.5 point change. Frequency driving more than half of that change.
There's another point coming from ex-cat property. I talked about the fact that we've seen a few more fires and water losses in the quarter, there's a point of that change coming from property. The other point or so is coming from expenses. That's how I think about the 4.5 point move.
Okay. What about just in auto?
In auto, you'd have to re-weight that without the property premium.
The auto news has nothing to do with the homeowner change, right? The two and a half points weights up to within auto, probably 3.7-ish or so. The other point I would make when you think about the roll forward with auto, we had a very favorable frequency quarter in third quarter 2014. We're comparing a little bit of a move, a 3% frequency move third quarter 2015 to a very favorable quarter last year. Makes the compare more challenging, but nonetheless, our pick is where it needs to be for third quarter 2015.
Okay. If I look at the underlying combined and auto deteriorated 460 basis points year-over-year. Can you tell us of that 460, how much of that was the higher expenses? Looked like expenses were about one point higher across personal lines. I'm just trying to figure out how much of that increase is expenses and how much of it is due to frequency.
100 to 110 basis points of the 450 would be the expense move.
Okay.
The frequency I think you have it, is 250.
Okay. The rest is frequency. Okay. Can you talk about the rate action that you've taken so far and when do you expect to see this trend normalize?
I guess a few things. As I commented, this did emerge on us over the course of the summer. September reasonably performed, not near the patterns we saw in July and August. We are, though, thinking we're in a different frequency environment. In fact, we had planned for that, right? Some of our plans in 2015 suggested that we were not going to see some of the favorability we had seen in 2014 and prior. As we move forward, we obviously are watching carefully, right? Whether we have a new norm of a frequency trend or not, I think it's early to call. I would not suggest that we think we're in a 3% go forward world, but we are contemplating whether we need to move and how aggressively state by state.
Lastly, our indications, and maybe I didn't say this before, as we think about our book of business, our frequency change does not look to be because of the new business we've written over the past year to 18 months. It looks to be across our book. Yes, there's a state profile. We're spending time with all the large states and also the small states. We're looking across all of our profiles. We're spending a lot of time on our renewal book, and we'll be dealing with and are dealing with rate actions necessary to counteract what we see pressure inside the frequency end.
Okay. Just on the middle market book, you mentioned a large property loss. Can you tell us, as far as a component of the commercial business, how many points of the combined ratio that represented?
A large property loss.
I think you mentioned in middle market, right?
I think I said GL. We had an old claim case in our primary liability book many years back, where as we worked our way through the court settlement process with this customer, we just decided it was a time we had to change our estimates, and so we did that, and that was a prior development move.
Okay. I thought I'd seen in the presentation that you mentioned that you had a middle market property loss. I'll go ahead and take a look at that. I'll follow up offline. Thanks.
Okay. You got it.
Your next question comes from Gary Ransom with Dowling & Partners. Your line is open.
Good morning. A while back, you talked about a new small commercial initiative through the AARP channel. I was wondering if you could give us an update on where that stands.
Sure, Gary, this is Doug. We did launch an AARP initiative last summer of 2014 with AARP. It has been slow to develop, but we've learned a lot. We continue to work that effort. The aggregate sum of the premiums is not over the top of $10 million, so this is still small dollars to us. But working with AARP and leveraging some of the things we've done in personal lines, I think we've learned a lot, and we continue to adjust and shift as we go forward and expect to continue to write customers that are a big part of the AARP program.
Do you think this can be something quite a bit bigger over the coming years?
I think it's early to call that, Gary. I'd rather give it a little bit more time. I think we probably, on both sides, expected a little bit more traction in the first year or so, I'm not deterred by that. I still think there are terrific customers that will become The Hartford customers over time. We just have not been able to generate the traction that we expected yet.
Okay. On the higher AARP marketing costs, is there any early read on what that has generated in terms of new sales as we've come after that?
It's a bit seasonal, Gary. As you know, we ramp up those efforts second half of the year, particularly leading into the January 2016 quarter. We do not have the success yet that those marketing efforts are geared at, I think we will over the coming months, obviously they'll be geared to geographies and our customer segments as well.
Okay. Thank you very much.
Thank you, Gary.
Your next question comes from Randy Binner with FBR Capital Markets. Your line is open, Randy.
Hey, thank you. Couple follow-ups on the frequency and then the commercial auto. On frequency in personal auto, I just want to get a simple point right, is that I think the narrative here is that the older drivers in this AARP heavy book are still different. Is that right, that they're a safer driver, they're just getting caught up in collateral damage out there out on the roads? The other piece of the AARP question is, can you talk a little bit about your pricing power to push that rate against the frequency experience in that book versus what we might see in a more wholesale channel?
I think your first point is well taken. Clearly, history plays that out. Our driving experience and our AARP financial experience has been very solid over a longer period of time. Yes, we believe exactly what you shared. Secondly, on the pricing piece, we're just going to have to work at this, right? This is a state-by-state, month-by-month effort. Obviously, the premiums don't all earn in day one. We're going to have to chip away over the coming quarters. It's not a three or six-month process, but it's one that has already begun and one that will accelerate as we look at these patterns that are coming at us today.
Yeah, I guess what I'm thinking is you should have better pricing power with an affinity channel, right? Can you share how past pricing increases have gone in this channel? How good the retention or reception is?
Well, we share retention with you in our sup. You've seen very steady performance in our retention area. I would expect that to continue. Consistency is a big part of that performance, right? We're very aware of how consistent we need to perform, both on a state and a product basis. I would say that across not only personal lines, but small commercial as well. We intend to address the signals we're seeing here. We're going to watch carefully, I think that our customer base is very well informed, and I expect that our retentions will remain strong moving through time.
Randy, it's Chris.
Okay.
I think the only other observation, too, is there's probably a little bit of industry tailwind helping all of us, given others reacting to the even higher frequency and moving rate actions in various states. Most of our policies are on a 12-month basis, given the more preferred marketplace. We don't have a lot of six-month policy phenomenon. The actions that others have taken, I think will lay a path that we could draft off of.
We also do this in conjunction with our partner, AARP, right? I mean, it's their members. We want to be thoughtful. We want to be consistent. We don't necessarily want to shock the system here. As Doug said, this is a three-month phenomenon for us, and we're not sure exactly where it's going to top out at. We know how to manage the AARP relationship. We know how to manage our 50 state regulators. And I think our past performance has indicated that fairly well.
Okay, great. Just jumping over to commercial auto. I think Doug had mentioned that you're getting high single-digit price increases there, and maybe that was common for the industry as well. I guess the simple question is, how do we get a sense that that's enough? The commercial autos, it was a problem during the financial crisis years, and now it's a financial recovery problem, if you will, with the 10 to 13 accident years. It seems like these are probably litigated claims, I'm guessing, where the medical is building up as well. Correct me if I'm wrong there, but I mean, is high single digit enough? I mean, should we be going for double-digit price increases here? I'm kind of interested in reflection on that.
Randy, a few points for you. One is that our high single digit is approaching double. It has been approaching double over the last several quarters. Yes, we are pushing the curve. Secondly, we are seeing the benefit inside the early looks of our 2014 and 2015 accident years. I feel better about progress inside those years. Again, it's a liability line that takes a while to play out. I want some maturity before we make those calls. I'm encouraged by progress and also the underwriting efforts underneath that are tweaking and adjusting our book of business, both new and renewal, continue to have their impact.
All right. I'll leave it there. Thanks a lot.
Operator, can I just make a comment before we take our next question? I think Mike from Goldman Sachs asked a question about a property loss, and I just want to come back to it. In the financial package, we did share that there was a property loss. It impacted our Middle Market book of business. We did have a $10 million net fire in that book of business. I think it was a point in change inside the combined ratio, so we commented on it in our disclosures. In the Midwest, I think it was a one-off, but that is the answer to the fire question.
Your next question comes from John Nadel with Piper Jaffray. Your line is open.
Hi, good morning, everybody. Doug, maybe to beat the dead horse of personal lines for a moment. If I look at the year-to-date underlying combined ratio, so excluding prior year, excluding cats, I think you're running through the first nine months at about 91.5% if my math is right. Your guide for 2015 was a range of 89%-91%. It doesn't seem like you'll get back to that unless you get some favorable weather or something else coming through in 4Q. I guess the question is, as we look out to 2016, if you saw the trends that developed here in the third quarter continue, what kind of range, relative to that 89%-91%, would you think would be reasonable looking out?
I'll take that, John. It's Beth Bombara. I don't think we really want to get into providing 2016 guidance at this point. I mean, you're right. When we look at 2015 and the combines that we provided at the beginning of the year, the 89%-91%, we're probably running a bit above that by 1 point. As we look into preparing our views into 2016, taking into consideration some of the actions that Doug Elliot is talking about, we'll firm that up. Sitting here today, I think we would expect to be kind of on the higher end of that.
Got it. Okay. That's helpful. I guess I have a question about the level of earnings for Talcott this quarter and thoughts on statutory capital from Talcott as well. Talcott's core earnings in the quarter, and if I have it right, $107 million. Now, you had the negative impact of lower alternatives returns, but I guess that was offset or maybe partially offset by better bond prepayment income. I just sort of want to get a sense if you had a 6% limited partnership or alternatives return there and more normal prepayments, what that segment would have looked like. Given you took a $500 million dividend out of Talcott in July, is it simply a matter of hedge gains given the negative market performance that drove statutory capital to be relatively flat quarter-over-quarter?
Yeah. A couple things. First on the NII piece, you're right in that the underperformance that we saw on the limited partnerships was really made up by the outperformance that we saw on these non-routine items. I mean, just to give you a sense for the quarter for Talcott, probably around $30 million pretax of these non-routine items. I kind of think of those a little bit as a wash. On the surplus side-
Okay
We did see a sizable increase in statutory surplus for Talcott once you take out the effect of the dividend. A couple things. One, you're right in that because of the market performance during the quarter, we did have hedge gains, and because our targets are more economic than stat, the reserves did not move to the same degree. We had a benefit there. The other item that impacted the quarter as well was just the recognition of deferred tax assets. That number does tend to bounce around a lot quarter-to-quarter just because of the way the recognition rules work. That was also a significant piece of that. When I think about Talcott for the year, because again, if you back out the impact of dividends, we see statutory surplus up almost $450 million.
I still go back to that $200 million-$300 million range we've talked about before. I think at the beginning of the year, I was feeling more like we'd be at the lower end of that range to maybe slightly below. Sitting here today, I kind of see us now at the higher end and potentially maybe a little bit above that. It really is going to depend on where market conditions end for the quarter.
Just a real quick follow-up on that, Beth. That $200 million-$300 million, if we're at the upper end of the range, given where things end, that's a 2016 dividend out of Talcott, correct?
Potentially. I think what we've said is that we want to look and see how the year ends, and some of that also, when you think about dividends, you got to think about where is the statutory surplus generation coming from. If it's coming from recognition of deferred tax assets, those are hard to dividend out. They're not cash yet.
Understood. Yep.
As we've said, we'll end the year, we'll assess where we are. We do still anticipate taking out the $500 million in early 2016, we'll evaluate what other capacity there could be depending on how we end the year.
Thank you so much.
Ian, we're ready for the next question.
Your next question comes from Meyer Shields with KBW. Your line is open.
Thanks. Two quick questions, if I can. First, can you give us a sense of the size of the renewal rights book that you bought from AIG?
Meyer, it's relatively small. Think of it in terms of around $30 million.
Okay. Second, I think you mentioned that you were growing in specialty large commercial property, and I was hoping you could talk about that because that seems to be one of the areas where pricing is particularly weak.
Meyer, this is Doug. We have been building capability in the public entity area for the last couple of years. That area has quite a bit of activity in the third quarter because July 1 tends to be a big renewal crossover for that book of business. A couple of underwriters in that area. We've had nice success. It's really had some traction and very pleased about how that's gotten off the ground in the last couple of years.
Okay. Thank you.
Your next question comes from Jay Cohen with Bank of America Merrill Lynch. Your line is open.
Yeah, sure. One more on the auto frequency issue, that is there seems to be a variance among companies as far as how they're experiencing this, where clearly Allstate and GEICO have seen a tick-up in frequency. Now you have, whereas seemingly Travelers and Progressive don't seem to be bothered nearly as much by this. I know you don't know these companies like you know your own company, but I'm wondering if you can reflect on some of the potential explanations for the differences we're seeing, because we're all a little confused by it ourselves.
Well, let me share some thoughts and maybe some of the thoughts I've just shared. I guess as you look at miles driven, just at the core, Jay, we're seeing gas prices remain at fairly low levels and the miles driven are up. As you think about parts of the country, we saw that over the middle part of the year, and I think we just have more drivers, and we have generally a better business climate. We have both commercial and personal line drivers on the roads. It's hard to avoid accidents as they continue to occur. Again, our book has performed well over time, and we're very pleased with that. The fact that they're on the roads today, they're just more accidents, and they're a part of those accidents.
As we think about kind of further dissecting, there is a statewide mix, as I mentioned earlier, and I think that will continue to play out. I don't want to make projections of 2016 and 2017, but I think the state does matter, congestion, et cetera. It's early. I don't want to take two months and make too much of it, but I also want to give you an indication that we're taking it very seriously, disappointed in our quarter, and are going to work hard to make sure we're connecting the pricing and our frequency discussions.
Jay, the only other observation I would have is we've said it before. Our mature book is different than other aspects of others. We don't know other companies' books as well as they know it, but all we know is that the mature driver does exhibit different driving patterns and different levels of frequency. Not only other industry observation and others have talked about it, but I've been more attuned to it, just particularly with teenage and young adult drivers, is there's too many devices in cars these days that are potentially creating a unfocused driver situation. I think it's real, and I think we all need to take personal responsibility to continue to put these devices down and focus on the road because there's just too many distractions. That's all I'll say on that.
Okay. Thanks for the thoughts, guys.
Your next question comes from Jay Gelb with Barclays. Your line is open.
Thanks. My first question's on property casualty investment income. In the third quarter, it was $267 million, and that included only a $5 million contribution from the limited partnerships and other alternative investments. Beth, I'm trying to square your comments in terms of what that means for 4Q. Should we expect the total net investment income for P&C to be below that $267 million in 4Q, if I understand the message you're trying to deliver? When you look at the compare from third quarter to fourth quarter, and we think about, again, the limited partnership returns, we are expecting to see those be below what we saw in the third quarter, just based on what we see today. That would be a negative.
As I commented, P&C doesn't really seem to benefit as much as the Talcott portfolio from some of these non-routine items that also impacted the compare.
More broadly, with Hartford generating around the 9% return on equity for the trailing 12 months, I think there's some concern now whether the company has the ability to get that higher over the next year or so. What are your thoughts on that?
Jay, I think we feel good about the improvements obviously we've made over the last couple of years. I know you know this team has worked hard to deliver that 9.1 trailing where we are today, and we're going to finish the year strong. We'll talk about our guidance for 2016 and beyond in February. I think what we've been trying to signal, particularly with my comments, Doug's comments, Beth's comments, particularly on low interest rates, there are serious competitive pressures. There's serious low interest rate pressures that are affecting the industry's book of business, and we're not going to be immune to it. I would still say that we think we have levers to continue to manage and expand our ROEs going forward. Clearly, it's not at the rate and pace that we've been able to deliver in the past.
Understanding the guidance won't come until early next year. Qualitatively, can you give us some insight on those levers?
Well, Doug's talked about it here, just in the personal lines book. It's underwriting actions, it's pricing action. It's a mix of business action. We're trying to be a more efficient organization while we invest in some of our new capabilities, particularly technology. As I mentioned, coming out of CIAB, I think we have the opportunity to expand our market share, not competing on price, but by delivering more products and services to our existing customer base, and they would do more business with us. Those are just some of the top of mind thoughts.
Excellent. Hopefully, capital management as well.
Clearly. I'm focused on the numerator, as you know. We have our plan through 2016, which I think you know is meaningful and will contribute to our performance.
Thanks for the answers.
Thank you, Jay.
Your next question comes from Tom Gallagher with Credit Suisse. Your line is open.
Good morning. One follow-up on the prior year development in commercial auto. Understanding the higher medical costs that you saw this quarter, Doug, is that what you've been assuming from the standpoint of 2014 and later initial loss picks? Or is that if medical costs remain at that level, you'd also have an issue there?
Tom, we are expecting some of that pattern that we've now seen emerge in 2010 through 2013 to also emerge in 2014 and beyond. Yes, we've become a bit more conservative in our development factors in those outer months in the most recent accident years.
In other words, if those elevated medical costs remain continue as is, you shouldn't have adverse development on 2014 and later.
Correct. That was our goal, and that's why we made the move.
Okay. The other question I had is, I guess for Beth, there was a $200 million capital contribution to a property casualty U.K. runoff entity. Can you just provide a little color what's going on there? Is that a business you're planning on selling? Any more capital going to be needed there?
Sure. We have been in the process of consolidating our U.K. P&C runoff businesses into a single legal entity. What we were disclosing was some of the activity that happened in October as this process does require court approval. We received it on October 13th, the capital moved around within the entities. This really is consolidating these businesses into one legal entity has a couple advantage. One, it is more efficient from a capital perspective when you take into consideration the capital standards in the U.K. It also allows us to consolidate operations, which gives us a little bit of operational efficiency. I'd say the second point is it does provide us greater flexibility to potentially act on more permanent solutions to dealing with these types of exposures.
Think about this group of business as having reserves about $600 million, and I'd say a little less than 60% of that is asbestos exposure. I'm not going to comment on the likelihood that we actually would be able to find such a solution.
Okay. If I could just sneak in one last one, Beth. For the 4Q actuarial review or balance sheet review that we should expect for Talcott
When I consider your long-term separate account return assumptions, which are still north of 8%, is that a meaningful risk as we think about current interest rate level and potentially changing that, as it would relate to a DAC impairment?
Yes. Correct in that we do look at the DAC assumptions and other reserve assumptions in Talcott in the fourth quarter. Sitting here today, I think a couple things to keep in mind, you're right that our long-term assumption is in that 8% range. We've benefited from many quarters of outperformance to that, which actually, when you think about how the DAC calculation works, gives you a little bit of buffer on that. Sitting here today, I do not see any changes in long-term expectations that would impact our views of that balance as we head into the fourth quarter.
Okay, thanks.
Thanks. Ian, I think we have one more question in the queue, if we could take that.
Your next question comes from Bob Glasspiegel with Janney Capital. Your line is open.
Good morning, Hartford. I was just wondering if we could just look a little bit more carefully at the homeowners book, where you've been sort of pushing 8% rate increases and retention's been dropping a little bit. The margins, you aren't making a lot of progress year-over-year, which you highlighted sort of the fluky fire losses this quarter. Have you ruled out any adverse selection going on in that book? What is your overall pricing versus exposure growth doing today?
Bob, you're right. We have had some up-and-down behavior of that book of business that has been disappointing to us. We're spending a lot of time at a granular level inside that book of business, by state, et cetera. I think more progress to come, not satisfied at all about our results. Leaning into the fourth quarter, the fourth quarter traditionally has been our best quarter of homeowners performance, we hope to continue that in the fourth quarter. There are things, some of which you described, that we are aggressively cutting apart as we speak right now to make sure there isn't an adverse element there. I don't believe that's the case today, right now, leaving no stone unturned.
A very thoughtful answer. Is there any difference between the agency and the AARP book? Is it a similar profile?
The AARP book has performed well across both our product lines. I would say that home has underperformed in general in both areas. We're not accepting and feeling great about our homeowners results across both channels, and looking at all.
Hey, Bob, it's Chris. All I would add in addition to Doug's thoughtful comments, as you said, is home's very strategically important to us. It's important to the AARP relationship. It's probably a product line that we've underinvested in. We're catching up, quite honestly. We're committed to catching up and providing more homeowners insurance to more AARP members, whether it be on a direct basis or through agents. No, it's got our attention and, as Doug said, we're trying to fix it as quickly as possible.
I think the industry is moving from that being an accommodation product to being a standalone P&L item, the professionalism of your competitors has certainly been enhanced. I think the opportunity to achieve that is there for sure.
Bob, I would agree with that. I think I inferred this, clearly our performance in the AARP space has been closer to our targets than our agency performance, right? We've got more work to be done in the agency area. In total, as I look across the line, still not pleased that we're not quite where we want to be.
Thank you.
Thank you. Ian, I believe I misspoke. There is one more question in the queue?
Your final question comes from Jimmy Bhullar with J.P. Morgan. Your line is open.
Hi, good morning. I had a couple of questions. First, on the personal lines business, one of the reasons for the weakness, obviously, was higher AARP direct marketing spending. Obviously, that's a controllable factor. Just wondering, what's your expectation for spending over the next few quarters? If your loss rates do remain elevated, do you intend to slow down spending a little bit to balance out profitability? Secondly, on buybacks, you spent about $300 million on buybacks in the third quarter. I realize you run a 10b5-1 program, to what extent do you have the capacity or the intent to be more proactive on buybacks, given that you've got the capital already? If the stock price drops further, would you maybe front end some of the buyback activity?
Jimmy, let me address the first part of your question, maybe Beth wants to work on the second. Obviously being part of the way through the fourth quarter, we have the ability to look at and address what we're going to spend in the next 60, 90 days, we're doing that as we speak, given the trends we're seeing. Second part of the answer I would share with you is that although I've characterized it as marketing spend, obviously part of that is the ability in our service centers to handle the demand that comes out of the marketing spend. It is really a marketing spend in the aggregate, we're making sure that if we're running ads on a weekend, we're ready to handle those requests as they come in.
It's a combination of both generating frontline response and also being able to handle the flow as it comes to our centers.
On the question on share buyback, yeah, we did do $300 million this past quarter. You're right, our practice has been to put in trading plans. For the fourth quarter, our trading plan is around $350 million. We can always be opportunistic. As you know, our practice has been to really spread that out over the period. We've seen that has worked very well for us, but it doesn't mean from time to time that we might take advantage of some opportunistic trades, but don't anticipate to deviate significantly from how we've thought about this in the past.
Okay, thanks. Maybe if I could just ask one final question. On your annuity business, your fixed annuity surrenders went up. I think that's because of the enhanced surrender value program that you initiated in June. Maybe if you could just address the scope of that, and has the benefit of that come through already, or do you expect additional or continued elevated lapses and into the fourth quarter? On the VA business, your surrenders actually went down. So how much of that is just because of the drop in account values versus maybe that in the past they were elevated just given the surrender value programs that you had before?
Yes, I'll address the VA piece first. Yes, we have seen a little bit of a decline in the VA surrender rate. You're right, in a market where we would see equity levels go down, we would typically tend to see the surrender rates act accordingly. We do believe there's just a little bit of the impact of the actions we've taken had in the past, kind of front-ended some of the surrenders. On the ISB, the program that we have in the fixed annuity book, we did benefit in the quarter from that. There is still some activity that's happening in the fourth quarter.
We might continue to see a little bit of bump from that. As we go into 2016, it remains to see what other type of initiatives we might do.
Okay, thank you.
Thank you, and thank you all for joining us today and your interest in The Hartford. Please note for your calendars that The Hartford will be presenting at the Goldman Sachs Financial Services Conference on December 9th in New York City. As always, if you have any follow-up questions, please don't hesitate to contact either Sean or myself by phone or email. Thank you for your attention, and have a good day.
This concludes today's conference call. You may now disconnect.