All right. Thanks, everyone. I'm Jay Gelb from Barclays. The format of this track will be a fireside chat with Doug Elliot, President of The Hartford, and Beth Bombara, Chief Financial Officer of The Hartford. The Hartford is among the largest property casualty insurers in the U.S., with a strong presence in both commercial and personal lines. It also has a presence in group benefits and mutual funds. Its business includes a Talcott unit, which largely consists of a runoff annuity business. Doug joined The Hartford in 2011 and has nearly 30 years experience in the insurance industry. Beth was promoted to Chief Financial Officer two years ago. Her previous roles at The Hartford include President of its Talcott unit and controller of the company. Beth joined The Hartford in 2004. Doug and Beth, thanks for joining us today.
Thank you.
Thank you.
Let's start off with addressing some of the issues that came up during the second quarter for The Hartford. 2Q results were negatively impacted by personal auto insurance claims and also legacy asbestos losses. What were the causes of those challenges in 2Q, and how confident are you that these issues are fully addressed?
You want to start with personal?
Sure. Let me start with personal lines. Beth will cover A&A. We did have a challenging, disappointing second quarter with personal lines. Over the last couple of years now, we've talked about what we've seen in terms of our trends. 2015, on the back half of the year, we had seen a pickup of frequency. Really what developed over the first six months of 2016 was a severity element, both in terms of injured parties in those accidents and also the severity of the number of people inside those accidents. Very disappointed that both in the first quarter and second quarter, we had to go back and adjust our year-end reserves. Primarily in the second quarter, it was a 2015 accident year reserve development.
A lot of levers and a lot of actions have been underway for a period of time, Jay, to address our personal lines book. I don't think many of the loss trends are just Hartford Insurance Group loss trends. I think many of our competitors are feeling a difference in behavior in the street. We feel like speeds are up, gasoline prices are down. We're seeing more accidents. That's evidenced by even national statistics. I think I just saw last week, the 2015 data from the highway and safety group said that highway fatalities are up 7% 2015. They're up 9% for six months of 2016. We are actively working both underwriting and pricing to address the issues in our personal lines book and feel confident that we're on a road to recovery.
Great. Beth, what are your thoughts on the asbestos environmental front?
Yes. Every second quarter, we do a ground-up analysis of both our asbestos and our environmental exposures. We were obviously disappointed to have to increase those reserves again this year. We've had to do that the last several years. What we've been seeing, and we continue to see in this year's study, is that the vast majority of our accounts when we do this ground-up analysis actually behave the way we anticipated with the number of filings leveling off or decreasing. For a handful of accounts, and it's been different accounts each year, so it's not as if it's the same accounts each year. We see a situation where the filings are increasing, or they don't decrease as much as we had anticipated. Because we project our asbestos reserves out to 2059, small movements like that can have a big impact.
It is hard to predict because when we do our study, we're looking at account-by-account analysis, making our estimations as to what we expect for filings and what we expect the exposure could be. Then we have to put it through what our coverage profile looks like for that insured because not every insured is the same. Where do we fall within their limits? Are they primary limits? Are they excess limits? It's a very complicated study. It's why we do it once a year. It does take a significant amount of time. The question on asbestos is, okay, will this continue? That's hard to predict.
If you go back and you look at what third parties have said about the incidence of mesothelioma and the fact that we should be seeing declining rates and not seeing that in all situations, there's a lot of theories behind that, some of which point to the fact that in general, as people live longer, we see the phenomena of people sort of living into the disease. We do see that a bit in that the average age that we see of these incidents is increasing. It is hard to predict. We'll continue to do our annual studies, and we've also talked about that we do continue to look at are there other solutions potentially. Vis-a-vis some third party taking on some of the exposures for us.
We just recently reached an agreement to sell our U.K. subsidiary, and one of the reasons why we liked that transaction so much is that it does have some asbestos and environmental reserves in it. Because we were able to put all the exposures into one legal entity and can sell the legal entity, when we sell it, we're done. There's no pushback to The Hartford in any way. When we look at our U.S. book, one, it's much larger, the size of the reserve. Second, we don't have a legal entity that just has asbestos environmental reserves in it. Any third-party solution really looks like it would have to be some form of reinsurance.
With reinsurance comes the complication that very unlikely to find someone that would take unlimited coverage. Most of the deals that you see done in the marketplace have some sort of cap, which means at some point, if reserves were to continue to develop and you exhausted that cap, the exposure comes back to you. Second, we really value our claims process and the experts that we have in-house that adjudicate these claims. If you give that up, you're sort of losing that expertise if you ever need to take some of these exposures back. Finally, price is always going to be a consideration as well. To some extent, dealing with a potential solution for A&E really involves how much capital do you want to put to such a solution, because it's not inexpensive.
Again, you have to kind of weigh the downside protection that you get versus the potential where these claims could level off. That's a long way of saying there continues to be uncertainty there, and we'll continue to monitor the book very closely, and we're open to other solutions to the extent that they make economic sense for us.
Plenty of other primary companies with legacy asbestos environmental exposure have addressed this largely through retroactive reinsurance. If we look at the impact of The Hartford on a pre-tax basis over the past two and a half years or so, it's been around $725 million. When you think about that relative to the other factors you also addressed, does it make you lean a bit more towards trying to find a reinsurance solution?
I would say that we definitely internally spend more time today than we did three years ago in looking at other solutions. Hindsight is always perfect. Obviously once you've made a decision in years past, you can see and make some determinations as to would it have been good to have entered into something in the past. Using the information that we have today, we'll continue to evaluate if there are solutions. Again, they're not inexpensive because obviously any third party is taking on that risk as well and looking to earn a return. You have to balance that as we evaluate, again, what's the best use of our capital.
Thank you. Let's shift gears to the commercial property casualty business. Doug, where are we in the pricing cycle, and do you feel it could get worse from here potentially?
Jay, I guess I would start the answer with identifying different market segments because I think there are different parts of the cycle. When I listen and read most of the comment, I think people are talking about the middle part of the cycle. I'll just start with a comment on small commercial. Very pleased with our performance in small commercial. Over time, it has been a steadier, not fully subject to the wide vagaries of the pricing cycles of middle, and we still feel very good about where we are today. Lots of competition, but we like our value prop in the marketplace and had a good two quarters starting out 2016. In middle market, we are seeing more competition. I commented on that in the second quarter. It has ramped up over the first couple of quarters of the year. You've seen our new business levels.
We share them transparently in our supplement. We are trying hard to be a disciplined, steady player in the middle, and in the second quarter, that translated into writing less business than we had in the second quarter of 2015. To me, the marketplace, I've had a full look at July. There are a lot of similarities between July and the second quarter. As we go back and close up August and finish the quarter, we'll get a better sense. We are seeing more aggressive competition, and we'll try to do the best we can to be disciplined, thoughtful, and see through the cycle, Jay.
Okay. Can the commercial insurance underlying combined ratio, which excludes the impact of catastrophes and prior year loss development, can that underlying combined ratio remain stable or perhaps improve given that we are in a highly competitive part of the cycle for areas like middle market?
On one side, we have pricing, on the other side, you have underwriting actions. We work those levers every day. We're working on our pricing curves and underwriting activities in small, in national accounts, and in middle. We're fighting a bit of an uphill battle in the aggregate across the industries as prices have come down. Again, our loss trends last several quarters have been very moderate, very pleased with our loss performance. Therefore our pricing trends are pretty consistent with those patterns. They are better than our longer-term loss trends. I think that is to see the answer to that question, we've got to play this out a couple of years. We are doing everything in our being to pull the right levers to combat what we see as more aggressive pricing, softer conditions in the marketplace.
Broadly, where do you see the growth opportunities in commercial P&C?
We still feel like small commercial is a growth opportunity for us. We have invested heavily over the past 10 years. We'll continue to invest in our digital capabilities, et cetera. We look to small. Most recently, we acquired a company in Georgia by the name of Maxum. Maxum will be a terrific acquisition, was a terrific acquisition for us. It'll allow us to think more broadly about appetite. This was an appetite expansion idea. What's neat about Maxum is that the leadership of Maxum had worked at The Hartford, so we know some of the players, they know us. Leaning into the middle market, very excited about some new capabilities. On top of basic organic opportunities, we're building out a new vertical in energy. Energy is a sector in the country that we've really not historically been in.
We maybe have been in isolated ways. We hired an executive in Texas to build out a vertical. We most recently hired an individual in the accident and health space to look at what we may be able to do between our group and our commercial businesses. Most recently, we just announced a relationship with AXA to extend our international capabilities, which are so important primarily to that middle-market customer who is having more and more of either their activities, their manufacturing, or their sales activities abroad. We're looking both inside and outside and excited about the last 100 days.
Excellent. Let's turn to personal auto. The Hartford's had a unique partnership with AARP for many years. Could you describe that program and how it's performed over time to compare to some of the more recent quarters?
Yeah. Our AARP relationship spans, I think, three decades. It has deep history. Just last week, I was looking at our 20-year partnership together with AARP, the financial performance across that period of time has been excellent in the aggregate. We are feeling a bit of stress. The stress we saw in the second quarter was not just in our agency book. Those frequency numbers and some of the severity are also in the AARP book as well. What's new in the last couple of years, traditionally, that has been a direct book of business.
Four or five years ago, six years ago, we started now building an AARP product that we distribute through our independent agents. That has been a multi-year effort, one that we are aggressively fine-tuning, that has been a growth component of our story the last couple of years, but also one that Ray Sprague, who runs that organization, and myself are working inside trying to get our rate program, called Open Road, where it needs to be as we've rolled out a new class plan in the last three years.
There's been some underwriting profitability issues in the AARP agency?
Our total agency book, including AARP, certainly has some challenges, which we're addressing. Also, the frequency and severities are in our direct book as well. We're addressing across, but I would say in our agency book, a bit more rate is driving into that book because of the need.
Okay. The Hartford's underlying auto insurance combined ratio has exceeded 100%, representing an underwriting loss in three of the four past quarters. What do you view as the causes there, and how long could it take to fix?
Let's go back to the end of 2014. Several of our competitors and the industry started seeing increased frequency at the end of 2014. We continued to look at our book and did not see the same frequency patterns. Those patterns did start manifesting in the middle of the summer of 2015. Actually, you could strike a line somewhere around July 1. Our frequency started emerging aggressively second half of the year, not all months, but most of the months. We think about patterns of driver. We think about economic activity, mileage, gas mileage, et cetera. A lot of things impacting us second half of the year. As we look back on our predictions, 2015, our loss picks, our loss trend picks for frequency, they're still very steady and actually have held very well the last seven months.
What has not held as well, which has been the reason we've gone back and had to restate both 2014 and in the second quarter 2015 as well, are our severity picks. The bodily injury predictions, Jay, inside our personal lines auto book, a little bit in 2014, but more aggressively in 2015, are the reasons we've had to do some of that.
Okay. What's the expectation in terms of when that gets back to target profitability?
We are working our rate plans like never before. I think I've shared on prior calls, our number of filings are up appreciably throughout 2016. It will take time for these rate changes to roll in, to talk to you in the past, this is more than rate change. This is a matter of rate change and also the underwriting levers that we're pulling. We're aggressively working our agency book. We have terminated relationships that have been outliers on the negative performance side. We have pulled back AARP contracted agents that had the authority to write our AARP book that were not doing it within our scope of partnership. We've made a number of changes in our class plan. Early in 2014, we started rolling out an Open Road class plan. It was the new iteration of our pricing vehicle.
As we were rolling that out state by state, the last state to go at the end of the summer of 2015, we have now iterated several new versions of the class plan, I think, pretty aggressively. In fact, version three is going in in the fourth quarter. We expect to see change positively in 2017, but it's going to take us a couple of years, Jay. We're going to need to work our way into 2018 for all these actions to earn their way into the book of business.
The other thing I know we've commented on before, just as a reminder, the majority of our policies are one-year policies. They're not six-month policies. That's also part of the reason why it can take time to earn in. I'd say sometimes we get the question, why one year? Should you move to six months? It is something that we find that the AARP member values, right? They get their coverage in place. They have it for a year. They're not getting a renewal midway through. Through the long term, that's been very beneficial to the member. Obviously, in a period like this, when you're trying to get a lot of rate into the book, it takes a little bit longer. We do think that it kind of adds to the value proposition for the AARP member.
Okay. What type of overall target combined ratio should we be thinking about for personal auto?
Our goal over the next two years is to get our ex-cat auto ratio into that 96%-96.5% range. We feel that that would be on top of our cost of capital at that level. That's not the ending point we want to seek eventually, but we think five, six points of change over this two-year window is achievable, and we're working incredibly hard to make that happen.
The underlying combined ratio should track reasonably closely to the calendar year, right? Given the low-
Correct
catastrophe impact-
Yeah
I think in an auto reserve.
That is a point or so probably on an overall annual expected basis. Obviously without prior period reserve, that would be a straight up number then.
All right. Thank you. Okay, Beth Bombara, turning to Talcott. Can you tell us the company's progress in winding down this $50 billion legacy asset under management business?
Yeah. Talcott has been in runoff since 2012. I think if you look over the last four years and what we've been able to do is significantly shrunk some of the more significant exposures that we had in there, namely the sale of our Japan business. Right now what we have is the U.S. business. It's the variable annuity, fixed annuity, and then we also have the structured settlements and terminal funding, which would obviously not be in the $50 billion number I think that you were referring to. We've been doing a lot of actions over that time to reduce the size and also the risk associated with the books. For us, it's not just looking at the absolute size, but how do we address some of the risk in that book. We've obviously been doing that through our hedge program.
We've also done some targeted buyouts, to policyholders, really targeting what we've viewed as the more risky part of the book. We didn't think that it made sense to do sort of broad brush buyout offers across the whole book. Again, to remind you, half of the variable annuity book is just a death benefit. It doesn't have the withdrawal benefit. We'll continue to look at things like that that can help accelerate the reduction of that account value. We've also done other things from the risk side. One thing that we did a couple of years ago was we started to enforce some of the investment restrictions that we have in some of our policies. That had the effect of, again, putting the profile to what we wanted as far as the asset classes that they were in.
That also resulted in some people deciding to drop their product as well.
What was very interesting about that program is we also saw the situation where some people who had the withdrawal benefit on their death benefit contract, when we put the investment restrictions in place, they didn't want to adhere to them. We were able to drop the withdrawal benefit rider, they still wanted the death benefit. Our policies never allowed you to make that kind of choice. Your choice was to surrender the whole policy. We look at things like that, where, again, if you can reduce the withdrawal benefit, we see that as having more risk. As we've always said, we're always open to looking at other ways to move the risk to someone else. Ideally, we would like to find a solution where we are able to transfer all of the liabilities and the legal entities, not do it via reinsurance.
When you do pieces, reinsurance gets a little difficult. One, because you're still on the hook as the primary writer of the policies. There's a certain administration and just process that still takes place that you have to deal with.
Again, I think as the book gets smaller, I think there's probably more potential for us to evaluate solutions. Right now, the way we see it, Talcott has been generating earnings. We've taken out $750 million in dividends this year. We took out $1 billion last year. We anticipate in 2017 of taking out something in the $500 million range. It is generating capital that's being sent to the holding company.
In a potential sales scenario of Talcott, how should we think about statutory capital as a reference point, as well as kind of where we are in the interest rate environment, how that could influence the outcome?
Yeah. I think typically when people look to kind of the valuation, so to speak, of Talcott, they tend to start with statutory. Obviously, a much more conservative regime in what it counts as assets. Interest rates do put pressure on statutory capital. Interest rates being low does put the need for additional reserves to be posted. We talked about on our second quarter call that as we look towards the end of 2016 and where interest rates are, that we would anticipate needing to book some capital testing reserves, which would most likely result in, for the year 2016, when you think of statutory surplus generation, again, not to be confused with the capital we send to the holding company, but just stat surplus before any dividends, that we wouldn't expect that to grow this year.
That even though the variable annuity business is generating fees and generating statutory surplus, it probably will be offset by the need to post additional reserves. The longer you stay in a low interest rate environment, you can continue to see that need. Again, when you're setting those reserves, you're looking at today's interest rate environment, but there is some mean reversion that comes into effect. The longer rates stay lower, the more the need is for additional reserves.
What does that mean in terms of Talcott's ability to generate deployable cash in a persistent low rate environment? I know that the outlook, as you just mentioned, $500 million is probably less than what we'd see in 2016.
Right
Which is around $750 million.
$750 million. Yeah.
What if rates kind of stay at this level for the next few years?
Yeah. I think we'll continue to see some pressure on those outer years. I don't want to get into too much of a prediction of 2018 and 2019 because it's always going to be based on facts and circumstances. There still is excess capital in Talcott. The way I think about the dividends that we send to the holding company is, one, you look at what statutory surplus did you generate in a year. Is that surplus able to be sent to the holding company? Then we just look at the absolute capital base, and as the book continues to run off, as we think about the buffers that we need for stressed environments, we share that usually annually. We show what our stress scenarios look like and what that excess capital is we have. That's sort of the second piece.
When you find yourself in a low-rate environment and you're doing stresses, that obviously can have an impact on that second piece. The bigger impact on the low interest rates really is from that first piece, which is about the actual surplus you're generating in a given year.
Okay. From a core earnings standpoint, from your GAAP earnings perspective, what do you feel the earnings profile of Talcott could be over the next several years compared to where we are now?
Yeah. A significant portion of the earnings that we generate does come from the fees that we earn on the variable annuity block. As that block continues to run off, you would expect to see a similar reduction in the amount of income. Interestingly, if you go back and you look at what's happened over the last several years, even though the variable annuity business has been surrendering, we've also had equity markets increasing. That counter impact, when you think about the fees that you're generating, has probably made the earnings decline not as significant as otherwise would have been. A big predictor of what the earnings profile will be like in the future is what happens to markets because of that aspect.
The other piece that Talcott has benefited from over the last several years has been the large returns that we've had in our limited partnership portfolio. Over the last couple of quarters, starting in the third quarter of 2015, we saw some pressure on those returns, and so that obviously impacted our earnings. Based on what partnership returns do, and a big portion of our portfolio is allocated to Talcott, it does support those long-dated liabilities we have in the institutional book, the structured settlements. That's a factor as well. The simple answer is the earnings will decline as the book continues to run off.
Okay. Does that mean we generally expect, all else being equal, 2017, 2018 Talcott earnings to be lower?
Lower
Okay. More broadly, if we think about the overall company on the return on equity perspective, the trailing four-quarter return on equity for The Hartford was 7.4%, and if we exclude Talcott, it was 9%. What do you think a reasonable expectation is for, say, 2017 or the long-term goal?
Yeah. A couple of things. Again, that was a rolling 12 months at the end of June. Obviously, those returns were impacted by the development we saw in personal lines and our A&E charge. If you exclude those two things, just the prior year development on personal lines, don't even try and adjust anything for the current accident year exposure we saw, and you adjust for the A&E charge, those returns for the total company would've been about 8.7%, and for the ex Talcott would've been about 10.9, almost 11%. Obviously those two factors impacted the return significantly. When we think about our returns going forward, we really do more and more look at it from those two perspectives. Obviously, the full company ROE is very important and we look at that very closely, but it is impacted by Talcott.
There's really not a lot that can be done relative to fixing something there because the earnings are the earnings, and the capital is the capital. As long as Talcott is part of The Hartford, that is going to be a reality that we have to look at. We really more and more focus on the ex Talcott ROE, and that's why we started disclosing it in our investor supplements, both total company ex Talcott, and then we show it for P&C, and we show it for group benefits. That's where we really want to make sure that we are focused on generating appropriate returns.
Again, if you think about backing out the prior year personal lines development, backing out the A&E, and the interest rate environment is the interest rate environment, not making any changes for that, I think being close to 11%, sort of on an ex Talcott basis, given all of the things that we're dealing with in the environment around us, are healthy returns and well above the cost of capital.
Okay. Last one before we get to the audience response system. Looking at share buybacks, the company, I believe, plans $1.3 billion of share buybacks for 2016. How should investors think about the sources of funds available for buybacks in 2017 and beyond?
Okay. Similar to what we see this year, all of our businesses are providing capital to the holding company that supports the share buyback plan that we have this year and in general provides us flexibility as we think about excess capital. What are those sources? I've already talked about Talcott. Again, we took $750 million this year. Anticipate somewhere in the $500 million range next year. If you look at this year, the other businesses and what they provided to the holding company, P&C is about $800 million of dividends. I'm not going to get into kind of what our exact prediction or estimate would be for next year.
I think that seeing it around $800 million, where we've been in the past, is a good start, and there's always the potential for that to be a little higher based on how the business performs. Group benefits was about $240 million of dividends that are going up to the holding company, and in mutual funds, it's about $70 million. Those businesses are all still part of The Hartford and would anticipate that they would be generating cash flows next year as well that would support the holding company.
Thank you for that. Let's go to the audience response questions. Okay, the first question for the audience is, you can start the clock, if you currently don't own shares of The Hartford or are underweight, what would cause you to change your mind? We still have the Jeopardy music going.
I was going to say, yeah. There's something going.
Otherwise it's just dead air, right?
Yeah.
All right. A couple seconds left. Okay. The outcome here, 50% saying they'd like to see a higher return on equity, 25% looking for improved property casualty underwriting results, and then smaller percentages, tighter P&C market or more aggressive share buybacks. Any thoughts on that?
Return on equity is obviously something that we are very focused on, and I just commented on my thoughts relative to that and where we're headed. I don't find that result surprising.
Okay. Next question. Oh, sorry, go ahead.
The only thing I would say is that obviously we can't control column one, but we are working hard to improve number 2, which is all about, at the moment, personal lines, because I really do feel like our other businesses are in pretty good shape. Always can be better, but fairly healthy start to 2016 on the commercial side.
Okay. Next question is, my confidence in The Hartford's ability to fix its personal auto insurance underwriting results by mid-2017 is You can start that countdown. All right. Just winding up here. The view is kind of straddling between high confidence and neutral confidence, but pretty evenly split. We'll look at the outcome there. Okay, next question, please. We can start the countdown. Should Hartford maintain its current business mix? Yes, it should remain in its current form. No, Hartford should explore options to divest, that being the second option. The audience is saying 50% saying yes, 50% saying no. I don't really know what to do with that, but
I think we'll just go to the next question.
Good idea. All right, next question, please. My expectation for Hartford's share buybacks in 2017. I'll give you my estimates just for point of reference. I'm assuming $1.1 billion in 2017. On buybacks, the audience is 50% saying between $1 billion and $1.3 billion, then sort of straddling either less than $1 billion or between $1.3 billion and $1.5 billion. That's what the data says.
Yeah, that's what the data says.
You like that?
All right. We'll leave it at that.
Are there any more ARS? All right. We're all set on ARS. I believe we have just a few minutes left for any questions from the audience. Any questions?
There's one way back there.
Can I ask about digital binding? Just, it seems like a lot of carriers have the digital binding presence, but you can't actually do the digital bind. You have to make a phone call at the end to talk to someone. Can you just talk about the pros and cons of that and why we haven't seen more of that in the industry overall?
Digital buying. We'll start with agreeing that to us, it's hard getting data, but it does not look like a lot of folks are end-to-end throughput. When people call us, they do need to talk to somebody. I'm assuming you're talking commercial, Small Commercial. Over time, I think we're all working on data. We're continuing to work on our algorithms. We've been engaged with agents and they drive our franchise, we understand that our customers over time will want to be serviced differently than they are serviced today and will have expectations to purchase over time. We're trying to make that more seamless. We want to be available over time to purchase our product where they want to purchase, how they want to purchase, et cetera.
Right now, our core theme is built around digitizing for that agent who is driving so much of the Small Commercial business.
Any other questions?
In light of the discussion around your ROE, can you talk about what you think of as your cost of capital, how you get there, and how that influences decisions at the company?
Yeah. Cost of capital is something we obviously pay a lot of attention to. There's different measures that people use depending on how you're looking at your beta and what your view is, obviously, on a risk premium. I would say when we look at it today, we would see it somewhere in that 8%-9% range, probably ticking a little bit closer to nine. Again, there's usually a range around that. It's something that we are very focused on. It's something that we look at very closely. It's something that our board also is focused on in understanding that relationship.
Excellent. Well, please join me in thanking The Hartford for joining us today.
Thanks, Jay. We appreciate it.
Thanks.