Good morning, ladies and gentlemen, and welcome to the first quarter results teleconference for Travelers. We ask that you hold all questions until the completion of formal remarks, at which time you will be given instructions for the question and answer session. As a reminder, this conference is being recorded on April 21st, 2015. At this time, I would like to turn the conference over to Ms. Gabriella Nawi, Senior Vice President of Investor Relations. Ms. Nawi, you may begin.
Thank you. Good morning and welcome to Travelers' discussion of our first quarter 2015 results. Hopefully, all of you have seen our press release, financial supplement, and webcast presentation released earlier this morning. All of these materials can be found on our website at www.travelers.com under the investor section. Speaking today will be Jay Fishman, Chairman and CEO, Jay Benet, Vice Chairman and Chief Financial Officer, Brian MacLean, President and Chief Operating Officer, Alan Schnitzer, Vice Chairman, Chief Executive Officer of Business and International Insurance, and Doreen Spadorcia, Vice Chairman, Chief Executive Officer of Claims, Personal Insurance, and Bond & Specialty Insurance. They will discuss the financial results of our business and the current market environment. Then we will refer to the webcast presentation as they go through prepared remarks, and we will take questions.
Before I turn it over to Jay, I would like to draw your attention to the explanatory note included at the end of the webcast. Our presentation today includes forward-looking statements. The company cautions investors that any forward-looking statement involves risks and uncertainties and is not a guarantee of future performance. Actual results may differ materially from those projected in the forward-looking statements due to a variety of factors. These factors are described in our earnings press release and in our most recent 10-Q and 10-K filed with the SEC. We do not undertake any obligation to update forward-looking statements. Also, in our remarks or responses to questions, we may mention some non-GAAP financial measures. Reconciliations are included in our recent earnings press release, financial supplement, and other materials that are available in the investor section on our website. Now, Jay Fishman.
Well, thank you, Gabby, good morning, everyone, and thank you for joining us today. We're very pleased to start 2015 by reporting another strong quarter with operating income of $827 million or $2.53 per share and an operating return on equity of 14.5%. Our underwriting results remained very strong across all of our business units, as evidenced by our combined ratio of 88.9%. Additionally, we achieved a record level of retention in domestic business insurance while simultaneously posting positive renewal rate change. With Quantum Auto 2.0 in full rollout in agency auto, we achieved policy in force growth of 1.8%, the first time that metric has been positive since 2011. We posted these results notwithstanding another severe winter and a reduction in net investment income from last year's first quarter.
Jay will speak about our investment income in just a moment, but for now, suffice it to say that nothing happened that we had not contemplated or considered a possibility. The operating and production results for this quarter remain very encouraging and consistent with our belief that the amplitude of the historical cycle has narrowed. Importantly, this stability contributes to our confidence in our ability to continue to deliver superior returns on equity. The confidence that we have is in part evidenced by our capital management strategies. Today, our board of directors announced an 11% increase in our quarterly dividend to $0.61 per share, marking the 11th consecutive year of dividend increases and bringing the compound annual growth rate on the dividend to nearly 10% over this time. In addition, we returned $850 million in capital to shareholders, bringing our total since 2006 to $31.6 billion.
We've done all of that while continuing to make significant investments in our business and our people. Canada, Brazil, and the data center in Omaha, just to name a few. Right-sizing capital and reinvesting in the business are not mutually exclusive. Both are fundamental to the way we think about the business for the long term. We intend to continue with these strategies, and in that regard, our board also authorized an additional $5 billion of share repurchases. We've made substantial progress in improving our product returns over the past five years through a thoughtful, focused, and highly analytical approach to our business that has been well executed in the marketplace. We intend to continue on this path.
In that regard, fixed income returns remain at historical lows, and as a consequence, we must continue to be mindful that this challenging environment has already lasted far longer than most would have assumed, and we will continue to factor that environment into our pricing strategies. In addition, we note that weather patterns do seem to be different. This change, when combined with increased real estate development, causes us to be very attentive to incorporating the real cost of weather uncertainty in our business. Part of the answer to this challenge above and beyond price and policy terms is helping our customers consider their risks and protect their assets in the most thoughtful ways possible, and we know that we and our independent agents are committed to helping customers do just that. As the world gets riskier, the value of thoughtful insurance advice goes up.
With that, let me turn it over to Jay.
Thanks, Jay. We're pleased with our first quarter results. Operating income of $827 million and operating return on equity of 14.5%, despite operating income being down $225 million from the very strong first quarter we had in 2014. This decrease in operating income was not driven by deteriorating underwriting results. As Jay just said, we reported a consolidated combined ratio of 88.9% this quarter. The decrease in operating income primarily resulted from three items: net investment income that was lower by $104 million after tax. The inclusion in last year's first quarter of a $49 million after-tax benefit from a change in state law that led to a reduction in our estimated liability for state assessments for workers' compensation premiums, and while still at a high level, net favorable prior year reserve development that was lower by $32 million after tax.
Net investment income this quarter is being compared to a very strong first quarter last year, both fixed and non-fixed income returns were lower than the prior year quarter. Non-fixed income NII was lower principally due to private equity returns, which were slightly positive this quarter as compared to a very strong first quarter last year. Energy-related private equity funds produced negative NII of $21 million after tax, giving up a small fraction of what had previously been earned in these funds in recent years. While we cannot predict future oil prices, even though we report private equity returns on a three-month lag, a number of factors suggest that the vast majority of the impact of the recent decline in oil prices has been reflected in the current quarter's results. Current oil prices are slightly higher than at year-end.
Share prices of public companies in our funds' portfolios are also higher than they were at year-end. Nothing we've seen has reduced our confidence in the general partners of our portfolio funds. Additionally, for non-energy-related private equity funds, which represent approximately 85% of our private equity portfolio, returns were adequate this quarter, although far below what was an unusually good first quarter in 2014. Fixed income NII was lower principally due to what we've been saying for many years now. Securities having higher book yields ran off during the past 12 months and were replaced with securities having lower yields due to the current interest rate environment. Another contributing factor was the level of short-term investments, which yield very little, being somewhat higher this quarter than in the prior year quarter.
Each of our business segments once again contributed to our net favorable prior year development, which totaled $243 million on a pre-tax basis. In Business and International Insurance, net favorable development of $77 million resulted from better-than-expected loss experience in the general liability product line for accident years 2005 and prior, along with better expected loss experience in workers' comp for accident years 2007 and prior. In Bond and Specialty Insurance, net favorable development of $35 million resulted from better-than-expected loss experience and contract surety for accident years 2010 through 2012. In Personal Insurance, net favorable development of $131 million resulted from better-than-expected loss experience in recent accident years for homeowners and auto liability coverages, as well as better-than-expected loss experience in homeowners for recent cat and non-cat weather-related losses.
On a combined stat basis for all of our U.S. subs, there were no accident years or product lines that had any meaningful unfavorable development this quarter. Operating cash flows of $199 million compared to $703 million in the prior year quarter, a $504 million decrease entirely due to the $579 million payment we made this quarter in final settlement of the asbestos direct action litigation. As this amount was fully accrued at the beginning of the quarter, this payment had no impact on the quarter's operating income or share repurchases, and it will have no impact on future share repurchases. Holding company liquidity was over $1.7 billion at the end of the quarter, and all of our capital ratios were at or better than their target levels.
Net unrealized investment gains were approximately $3.2 billion pre-tax or $2.1 billion after tax, up from $3 billion and $2 billion respectively at the beginning of the year. Book value per share was $77.96, or 1% higher than at the beginning of the year. I'd point out that book value per share growth was achieved despite the negative impact of the strengthening of the dollar on foreign currency translations. Turning to capital management, we continue to generate much more capital than we need to support our businesses, allowing us to return $850 million of excess capital to our shareholders this quarter.
We paid dividends of $178 million and repurchased $672 million of our common shares this quarter, including $600 million under our publicly announced share repurchase program, consistent with our ongoing capital management strategy, and $72 million to partially offset shares issued under employee incentive plans, mostly to cover employee withholding taxes due upon the vesting and payout of performance and restricted stock awards. As Jay said, the board raised our quarterly dividend from $0.55 to $0.61 per share and added an additional $5 billion for share repurchases. With all that, let me now turn the mic over to Brian.
Thank you, Jay. Alan and Doreen will go over the individual segment results in a moment, before they do, I want to make a few comments on the impact of weather and the operating environment in the domestic commercial marketplace. Regarding the weather, patterns continue to be unpredictable, that was clearly evidenced by another challenging quarter. Weather losses in the first quarter of 2015 were slightly higher than the first quarter of 2014, more importantly, in both years, we experienced losses above our expectations, driven by the polar vortex last year and the extreme snow and cold in the Northeast in 2015. The real story is that nearly 95 inches of this snowfall came in just a 30-day period.
To show just how unusual that was, The Washington Post cited a meteorologist who calculated that Boston should not expect to see another 30 days with that much snow for another, and I quote, "approximately 26,315 years." I'm not sure what they meant by approximately, but suffice it to say it was an extremely unusual event. As we think about the continuing unpredictability of weather patterns and our exposure to catastrophic weather losses are than they were just six or seven years ago. Importantly, historically, we would've expected to have been below budget absent a fairly significant Atlantic storm. Unfortunately, it has now become all too common for us to have significant cat losses from what we traditionally thought were the lower severity frequency events like tornado, hail, and winter storms.
Weather patterns are changing, and accordingly, we are continuing to reassess and manage our property exposure and pricing in a thoughtful way. Regarding the domestic commercial pricing environment, we feel very good about the returns we are seeing in this business, and we are encouraged by the continued stability of the market. In the fourth quarter of 2010, we embarked on a strategy to improve the returns and profitability of our commercial business by increasing rate, improving terms and conditions, and focusing risk selection. We said that we would execute on this strategy thoughtfully and in a manner that would not be disruptive to the marketplace. Accordingly, our initial goal was to achieve gradual rate increases, eventually reaching a level that would drive significant improvement on the returns of our products over time.
Once our returns had reached an appropriate level, we hoped to see rate moderate at a level that would allow us to sustain reasonable returns. Looking back, this is almost exactly what we have seen over the last four and a half years, and we couldn't be more pleased, not just in the results, but in how we have achieved them. We couldn't have scripted it better. With that, let me turn it over to Alan.
Thank you, Brian, and good morning. We feel very good about our performance in Business and International Insurance this quarter. Operating results were strong, and importantly, we were encouraged by the continued stability in the U.S. marketplace as reflected in the facts that our retention is at a record level and we're still achieving renewal rate gains. As Brian said, we could not have scripted this better. The underlying combined ratio was 92.7, an increase of 3.2 points over the prior year quarter. Eliminating the benefit from the state workers' compensation assessment in last year's first quarter, the underlying combined ratio increased by about a point, reflecting a typical level of quarterly fluctuation. Particularly in light of the severe winter weather that Brian mentioned, we feel great about the underlying profitability.
Turning to the top line, net written premiums were up about 1% from the prior year quarter, with the timing and structure of a number of reinsurance treaties and changes in foreign exchange rates having an adverse impact. Domestic net written premiums were up about 2%, but adjusting for the timing impact of the reinsurance treaties, premiums were up almost 3.5%. In our international business, net written premiums were down 7%, driven by the adverse impact of foreign exchange. The production results this quarter in our domestic business continue to be very encouraging. As we've said in the last few quarters, more of our business is achieving target returns, and our objective has been to keep more of that business through our very granular account-by-account and class-by-class execution. Record retention this quarter speaks to the success of that strategy.
Renewal premium change was down about a point from the fourth quarter of last year. Within that, pure renewal rate change was also down about a point. Notably, renewal rate change across most of our businesses has been steady for each of the three months within the current quarter. Turning to Select, retention hit a four-year high and renewal premium change was strong at 7.3%. Overall, we're pleased with the returns in this business and will continue to seek higher retentions and higher levels of new business. In middle market, retention was a very strong 87%, with over 90% retention on our best-performing segments. Renewal rate change moderated less than a point from the fourth quarter to 1.1%. In terms of new business, we're very pleased with the results in middle market this quarter.
As you've heard from us, the impact of rate gains over the past four years has resulted in more new business opportunities that met our return thresholds. The amount of new business we wrote in the quarter was impacted by the high volume of new business activity in the marketplace. January is a big month for new business submissions. It's worth noting that the prior year quarter was particularly low from a new business standpoint. That obviously contributes to the quarter-over-quarter increase. Our international business also posted solid production results. Over the past several quarters, renewal premium change has been steady and retention has trended up. In the current quarter, new business levels were lower than we would've liked, largely reflecting our disciplined response to conditions in the Lloyd's marketplace.
I'm pleased to note that as we announced last week, we agreed to increase our stake in the P&C business of our Brazilian joint venture to 95%, up from 49.5%. We will maintain our 49.5% stake in the JV surety business. Our P&C business in Brazil is small. We launched it in 2012 and wrote about $20 million of net written premium last year, but we continue to see long-term growth opportunities. The transaction is expected to close in the fourth quarter of this year, subject to regulatory approvals and customary closing conditions. To sum it up, we feel great about the quarter. We're focused on maintaining our attractive returns, and we're well-positioned to execute where we find opportunities. With that, let me turn it over to Doreen.
Thank you, Alan. Good morning, everybody. Bond & Specialty Insurance started 2015 with a strong first quarter, and we remain exceptionally pleased with the financial return in this segment. For the quarter, operating income was $124 million, a reduction from the first quarter of 2014, due primarily to a lower level of net favorable prior year reserve development and lower net investment income. Underlying underwriting results remained very strong and well within our targets. As for top line, net written premium for the quarter was essentially flat to 2014 for both our management liability and surety businesses. Across our management liability businesses, retention and new business premium were both slightly improved from recent quarters, while RPC levels remained broadly consistent. Not a lot of movement in the quarter to highlight for you, and we continue to feel great about this segment's results.
I'll turn now to Personal Insurance, where we also had a terrific first quarter, with strong underwriting results in both agency auto and agency homeowners and other. Operating income for the quarter was $252 million, down slightly from the first quarter of 2014, due primarily to lower net investment income. The underlying combined ratio was essentially flat to the prior year and at a level that we feel great about for this segment. Looking at agency auto, we once again delivered strong top and bottom-line results for the quarter and continue to be very pleased with the performance of this business. The 90.2% combined ratio for the quarter benefited from approximately three points of favorable prior year reserve development, which was driven by better than expected severity and bodily injury.
The underlying combined ratio of 93% was a slight improvement from an already strong 2014 and in a range we're comfortable with. Production results also continued to be strong, driven by Quantum Auto 2.0. New business premium was 67% higher than the first quarter of 2014, and we continue to grow policies in force, which increased 21,000 during the quarter. Net written premiums increased 4% from the prior year quarter. By all measures, great results for auto for the quarter. Turning to agency homeowners and others, we once again delivered strong financial results, and we're making progress on production as we execute local product and pricing strategies. The overall combined ratio of 74.4% was strong and benefited from favorable prior year reserve development. The underlying combined ratio of 79.5% is also well within our return expectations for the quarter.
Financial results remain strong as we continue to execute our disciplined underwriting and pricing strategies. As for agency homeowners production, we said earlier that we continue to make progress. New business premium was up almost 40% from the prior year quarter and continues to trend favorably. We feel great about our market-leading homeowners capabilities and know this product is a critical component of the customer package. Brian discussed the continued impact of weather. With these considerations in mind and our strong financial results in this line, we remain focused on improving our offering, both product features and process improvements to begin growing this business again. Based upon the new business trends we're seeing, we're encouraged with our path. Lastly, on agency homeowners and others, net written premium was down 5%, primarily driven by the impact of changes in the timing and structure of certain of the company's reinsurance treaties.
To sum up, Personal Insurance had excellent results and a great start to 2015. With that, I'll turn the call back to Gabby.
Great. Thank you, Doreen. We're now ready for Q&A, and I'd like to remind the participants to please limit yourself to one question and one follow-up. Thank you.
Thank you very much, ladies and gentlemen. If you'd like to register a question, please press the one followed by the four on your telephone keypad. You will hear a three-tone prompt to acknowledge your request. If your question has already been answered and you would like to withdraw your registration, please press the one followed by the three. If you're using a speakerphone, we ask that you please lift your handset before entering your request. One moment, please, for the first question. Our first question comes from the line of Randy Binner with FBR. Go ahead.
Hi. Good morning. Thanks. Two-part question on reserves, especially in the commercial area. I saw the comments that you took some CGL in 2005 and prior on work comp in 2007 and prior. One, was wondering about more recent years in work comp and casualty lines, if those need to season more. Two, if you could provide your initial take on the 2012 and 2013 accident years. The question there is, given that you're about 36 months out from there, and we had a kind of a mini hard market in pricing in those years, and I think evidence of loss cost trends remaining benign, wondering how the 2012 and 2013 years are looking now that you can have a better view on that.
This is Jay Benet. I'll try to wrap in some comments about all of that in some overview sentences. As it relates to the reserve development, these years do have to season a bit for us to get a good look on what the longer tail liabilities are doing. However, having said that, as you know, we have a very robust and granular process for looking at our reserves. We look at them every quarter, and if there are things that are developing, either currently or from prior years that would have a rollover effect, which we call base year movement, into the current years, we'll reflect that as appropriate. Looking at the more seasoned years at this point in time, we did take action to reduce reserves, as you pointed out, in general liability and workers' comp for 2007, 2005, and prior as we disclosed.
None of that really had a major impact on the more current years. The current years haven't developed in any meaningful way, one way or the other at this point in time in the current quarter. There have been actions in previous quarters, some of the more recent accident years where we've made some minor unfavorable development taking place in any of our products, any of our accident years. On the favorable side, we pointed out in the press release what that was.
I guess the follow-up would be just kind of honing in on 2012 and 2013 and understanding they're not fully developed, but there is some amount of time that's passed. Given the pricing that was better then, do you have evidence that loss cost trends remain benign in that period of time? Was there an elevation in loss cost trends in those years in particular? From what we're seeing, it seems like most loss cost trends for casualty have been incredibly benign, especially in those years.
Yeah. Well, as we've said in previous calls, we're not seeing anything as it relates to loss cost trends that's surprising to us or different from what our assumptions are. These are, of course, very long liabilities. You're going to react in a prudent way to any changing information that you see. As it relates to this whole, what you're referring to as benign or whatever, everything is really dependent upon what your reserving assumptions are. As long as your reserving assumptions are based on what's taking place in the marketplace, you're not making adjustments for things. Which is not to say over time, you wouldn't look further at the data and see things that would cause you to recognize favorable development or, God forbid, unfavorable development.
At this stage, our reserves, as we always do, are best estimates of what we see today, and we'll just see how they develop going forward.
It's Jay Fishman. Just let me add one point. A couple of times in your question, you referenced pricing and its impact on reserving. Pricing is not a factor as we set up reserves for our business. Our reserves are driven by costs, and the pricing, what we sell it for, is driven by the marketplace. The notion that our reserves are set differently in a different pricing environment is just wrong. It's just not correct. What Jay, I think is what, so far, so good. So far, the estimates that we established in the years of which Jay referenced are continuing to stand up to whatever the developing trends are.
Understood. Thank you.
Thank you. Next question, please.
Our next question comes in line of Kai Pan with Morgan Stanley. Please go ahead.
Good morning. Thank you so much for taking the call. First question is on the margin. Alan classified the one-point movements in term of business margin deterioration as a quarterly fluctuation. Just wonder at that one point impact by some non-cat large weather-related losses. Also going forward, if you see your pricing continue to accelerate, are we going to expect some continued margin deterioration?
Hi, it's Alan Schnitzer. Let me take that. The one point that we characterized as typical fluctuation really arises from two components. One on the loss side, there's some fungibility of losses. We're talking about relatively small numbers in our losses. What we really saw in the quarter was a very small, you can count them on one hand, a small number of large fires around the world that contributed to after tax less than $20 million. That may be where that comes from. The rest of it was on the expense side, there's always going to be some normal fluctuations in expenses from period to period.
In terms of the margin outlook, you could certainly look at what rate is doing and what loss trend is doing into a very narrow quantitative analysis and say, "Gee, margins are going to shrink." We tend to look at margins on a broader basis. There are lots of things other than the written rate and an estimated loss trend that are going to impact margins going forward. You got things like volume exposure, mix expenses, claim initiatives, weather large losses. All those things every quarter are going to have an impact on the results. At least the way we look at written rate and loss trend at the moment, it's hard for us to say necessarily what margin's going to do over the near term, and we take our best shot in giving you some perspective on that in the outlook section of the 10-Q.
I would suggest that if you haven't had a chance, you take a look at that.
This is Jay Fishman. One other observation. Your question and Alan's answer is in respect of written rate versus loss trend. Obviously, on an earned basis, as we look forward to the rest of this year, looking over to Jay Benet to see if I'm right, we anticipate that earned margin will continue to modestly expand from here. All other things being the same, and they never are. I'm simply talking about the calculation of earned rate versus loss trend modestly. Everybody's pointing modestly here. It's not a matter of this year. Now, I second the motion on Alan's comment, which is one is obviously a lower number than four, and all other things being the same, that would suggest some margin compression. Again, all other things are not the same. Exposure, a meaningful portion of exposure acts like rate.
You get volume differences, you get productivity efficiency. It's close, and I would hesitate on concluding definitively that written margins, and therefore sometime down the future, earned margins at this level will, by definition-
Reduce. They may. It just isn't that clear yet. The numbers are just not that pointedly different.
Thanks. I have follow-up on the reserve side. If you look at the short-tail line releases over the past 2 years that account for majority of the reserve releases happening in these 2 years, I just wonder in hindsight, because short-tail line, you would imagine probably a little bit easier to reserve than the long-tail lines. What have they helped you guys in the past 2 years in terms of short-tail line reserve releases and the sustainability of that going forward?
We can't talk about the sustainability of it because going back to our reserving philosophy, we're always reserving at best estimate. I can talk about some of the things that have caused the reserves to develop favorably. You say that short-tail lines are easier to reserve for, it's an interesting phenomenon that takes place if you're dealing with short-tail and particularly if you're dealing with property. You have events that take place, in many cases are taking place in a current quarter where you're coming up with the estimates in that quarter. Even though it's a short-tail business, you're dealing with not a lot of claim information at that particular time. You're looking at recent activity or historical activity, to make judgments as to how those claims are going to develop, how the storms are going to develop.
Depending upon the nature of the storm, where the storm hits, you could have situations where you're thinking that a tornado has a lot of hail associated with it. You look at the history of how that's developed and put up numbers. Then, as time goes on, we found that we overshot the mark a little bit because that particular storm didn't follow the pattern of previous storms in terms of the related hail damage. It's like everything else in reserving. You have some data upon which to make assumptions. Then you extrapolate from that, you make your best estimate, and then you're always adjusting it. When I say it can lead to higher fluctuations, what I really mean by that is, the period of time in which you find you're making the adjustments is a relatively short period of time. It's correcting quickly.
Whereas, in lines like GL or workers' comp, you can often find that you're seeing long-term trends and you're evaluating them, but the period of time in which you're making the adjustments can oftentimes be over several quarters or even several years, as you see the data getting more solid. It's an interesting aspect of reserving that I found intriguing when I first got involved in it. As it relates to the short-tail lines, you do see these fluctuations because you're dealing sometimes with very, very little data.
Well, thank you so much for all the answers.
Very good. Next question, please.
The next question comes from the line of Amit Kumar with-
Hey, it's Amit Kumar from Macquarie Capital. Just two quick questions. The first question is for Doreen. The results in auto were strong. Can you sort of expand on the loss cost comment? I think previously you've mentioned that loss trends have remained at around 3%. Is it fairly unchanged in terms of where we are in this quarter?
Yes, it is. We debated about whether to put that in and figured that you'd ask us about it anyway. Yeah, we do see the severity trend for auto to be about 3%, and that's all in. I guess, I'm just trying to think, and that includes bodily injury as well. That's a little bit higher, but when you blend everything, it's about 3%.
Got it. On the surety piece, there have been recent press reports mentioning Malucelli and Petrobras. Would it be possible to maybe talk about any go forward potential surety exposure you might have to this piece?
Yeah. Amit, it's Alan Schnitzer. Doreen and I can tag team on this one. It's way too early for us to really have any assessment on the impact on us from a surety perspective. This is just starting to unfold. What I will say is that we've got great confidence in our people down there. We've got great confidence in our underwriting. We manage our gross and net exposures pretty carefully. Together with our partner and our local team, we continue to watch it. We don't have any reason at the moment to think that this is an outsized problem for us, but I'll reiterate that it's early, and we're watching it.
The only thing I was going to add is that we remain very proud to be committed with J. Malucelli as a partner, and they're one of the finest companies in Brazil. With all of the discipline they bring and we bring, it's an unfortunate set of events, but we feel like we're in the best position that we can be with that partner.
Got it. Okay.
Our next question comes the line of Jay Gelb with Barclays. Go ahead.
Thank you, good morning. First, I just had a question on the buyback, the $5 billion authorization. Do you anticipate that taking roughly two years to complete?
This is Jay, Jay. Two Jays here. Without predicting what it is, we talk about depending upon what you and others feel our earnings look like going forward. You can do the math as to how long you think that share repurchase authorization will be out there for our utilization. Of course, that's in addition to the remainder of the previous authorization, which I think is another $884. I think between looking at what your projections are and looking at the utilization in the past, you can come up with a pretty good estimate of what you think it'll be.
Okay, that makes sense. Thank you. My follow-up question is on investment income. Given the drop-off in the other investment income and in the first quarter of highest in the first quarter. If I put that together, it looks like you could see at least a 10% decline in investment income after tax in 2015, maybe into the low teens. Is my math okay?
Well, I think there's two pieces to that. One, looking at the fixed investments. We put some disclosure in our Q as to what we thought the period-over-period decrease on a quarterly basis would be for fixed income NII. Of course, the drivers of that are twofold. One, the lower interest rate environment as items mature either in the current quarter or really in the previous nine quarters as well. They're getting reinvested at the lower rates, so that will have an impact. Also, as we pointed out, we paid the $579 million on the direct action so that those funds are no longer in the investment portfolio. When you combine those two things, you're roughly at about a $30 million after-tax decrease on the fixed income NII as it relates to-
Per quarter.
Per quarter. As it relates to the non-fixed income portion, we've provided you data on what the composition of the portfolio is. You've seen the historical yields associated with it. I don't have a crystal ball to say how the economy is going to perform and whatever. We've tried to give information as it relates to the impact of the oil price decline and the feeling like we've hopefully gotten to a point where the vast majority of whatever repricing or revaluations will take place have been captured. I think you need to make your own estimates as to what you think that part of the portfolio will do.
It's Jay Fishman. I'll give it to Bill Heyman in a second. Just two points. I'm not clear what your comment about seasonality in net investment income is.
Oh, sure. On the fixed income, investment income for the year seasonally, it's highest in the first quarter?
No.
Well.
It's declining. These are just bonds. Don't interpret it as seasonality one bit. It's just yield. That's all it is. It's the same bonds.
Okay. If I look at first quarter every year, it seems like the first quarter is the highest, and then it trails off.
There's nothing about our fixed income portfolio that has any seasonality to it whatsoever. Other than as time progresses and more of the portfolio is reinvested at lower rates, we've been saying it for years now, the net investment income will go down over time. No seasonality in that number at all. As it relates to the energy dynamic, I would point out that we put out a schedule in last quarter's webcast, which summarized our private equity investments and what we thought was the, and we defined as the energy space. My recollection is that somewhere around $760 million or so inclusive, I think. Let me go back. I'm working from memory. It's not insignificant, but hardly overwhelming. Those are the two points I'd add on investment income.
Jay, it's Bill Heyman. What you see as seasonality is really a degradation of book yield that's gone on now for three or four years point and quarter by quarter. That slope, the slope of the slope will change over time, but the direction of the slope will persist for a while. Our basic allocation here, which is 93% or 94% to fixed income is designed so that even if investments other than fixed income don't perform, which we know on occasion they won't, our interest revenues, assuming we collect them, even at these levels are enough to give the company a shot at an overall ROE in our target range. That's the policy, and it's not really an accident. In terms of the non-fixed income portfolio, I know we're making the comparison this morning to the first quarter of 2014.
I think it's instructive to look at this quarter against the fourth quarter of 2014 and then the fourth quarter against the third quarter of 2014. You can see that for non-fixed income investments as a whole in the webcast on the bottom right of page six. That includes the entire universe. In private equity, the decline over the last two quarters was pronounced. This doesn't come as too much of a surprise. Given the elevated levels of the first three quarters of 2014, you could see comparisons like this again for a couple of quarters yet.
What I think we can say is that since we receive results and report them to you on a quarterly lag, we try to go behind them and look at market conditions and determine whether there is in the portfolio some embedded but yet unreported loss that means that these marks are less accurate. We don't see that. Obviously, our vision isn't perfect, but Jay Benet alluded to developments in oil prices and in the publicly held securities of these funds. We looked at equity markets generally, but it is what it is. For what it's worth, real estate this quarter performed very well at levels equal to the first quarter of 2014. Hedge funds performed better. The shortfall was in private equity, and 50% of it came from 15% of the portfolio. That's the story.
Great.
Very helpful. Thank you.
Next question, please.
Next question comes to the line of Josh Stirling with Sanford Bernstein. Go ahead.
Hi, good morning. Jay, I wanted to ask a follow-up question on Kai's question about where we are on pricing relative to loss trends and how that applies for future margins. I guess I'm really curious, where do you think the break-even point is going to be for future margin compression? If we sort of dial back, I think a year ago, maybe 3% to 4% is what a lot of companies were saying you'd want to cover loss trend with, and now it's at 1.4% versus maybe the CPI at something approaching two. I know there's lots of moving pieces. You mentioned some of them, but what do you think the level is that we think might be margin compression kind of pricing levels? If you don't mind the question, you guys have led us through this pricing cycle so far.
Will you draw a line in the sand to preserve margins before you get to that point?
Well, first, we really don't perseverate nearly as much as perhaps you all do on whether margins are a little bit wider or a little bit smaller. Independent of whatever CPI numbers come out or producer price index, we're pretty granular about our loss trend and feel that we try hard to get it right. The reason we try and get it right is a whole lot less about reserving than it is pricing. We price our product to losses, if we don't have our losses right, we're not going to get the right price. We get very driven and very focused on that. If you wanted to do just simple arithmetic and say, unless you get four points of written rate, all other things being the same, margin would be an increase in property limits on a building.
You collect more premium, granted there are more limits at risk, but absent a total loss, that number will act in the calculation equivalent to rate. When we start trying to analyze it, the best answer that we can give you right now is, it's close. It's close. Again, it's not as if we perseverate much at all on whether it's up a little bit, well certainly not up, down a little bit or flat. We think about returns over time, importantly, not about margin in any given quarter. Returns over time. It's how we think about it. That's sort of our outlook. In terms of lines in the sand, we always draw them. There's nothing different about that. One of the things that's impressive to me, Alan shared some of this data with you a couple of quarters ago. I'm sure it's dated.
We don't even quote on round numbers 60% of the requests for quotes that we get in commercial accounts. It varies, 60%-70%, something like that. Am I close on that still? Just wondering. The line in the sand starts right there. We don't quote on more business than we do quote. Then our hit rate is a function of that smaller quote rate. We draw lines in the sand every single day at every account. It's not a philosophical line in the sand. It's an account-by-account philosophical line in the sand. That's just so important. We don't pressure underwriters for volume. We don't. If you do, you'll get volume. You won't like what you get. That's been our philosophy, but you'll get it.
We let them run their business with tremendous amounts of data, like adults making thoughtful decisions, managing it for the long term. Yeah, we draw lines in the sand all the time.
I would just add to that in terms of the line in the sand. As you've heard from us, we are return-focused, and that's our objective as we underwrite account by account or class by class. That return will take into account interest rates, competitive conditions in the marketplace, everything else. While we are drawing lines in the sand on accounts, we're not going to look at some magical return number and say, "That's it. We're stopping." That's going to change, and it's going to take into account all the facts and circumstances over time.
One of the questions that we've been asked over a long period of time, and our answer has been the same. To us, the leading indicator about lack of stability in the market and the way in which you can best assess the sort of near term is retention. It is really, to me personally, given my experience here, remarkable that we had record retention in our domestic Business Insurance business while still getting positive renewal rate. I just think that's really quite remarkable and speaks to the stability of the business. If you want to get a sense of churn, you look at retention. I'm getting a little long in the tooth, but I go back to the late '60s when retentions in the middle market were in the high 60s. 70 was considered a pretty stable month.
This quarter, in the middle market, we were like in the 87%. It just shows you the magnitude of the difference and really, I think, just reinforces the way in which we run the business.
That's, Jay, Alan, that's really helpful as we think about simplifying all this. I guess my one follow-up on that, and then I'll let you guys move on. If I integrate what you've all just said, which I think is that you're sort of near your break-even point on margin compression and you're happy with rates, but then in your outlook you say that basically that you expect pricing to be broadly consistent for the rest of the year with the level it was in the first quarter. Does that mean we should basically be interpreting that the slowing pricing trend of the past seven or eight quarters, which has been relatively stable, is going to stop here, and we're going to see you guys try to keep pricing where it is here for the foreseeable future?
Again, I've got to fall back to the account-by-account discussion. That's how we think of pricing, not as a single manifest order to the organization. If an underwriter is sitting with an agent and they're talking about an account that's had, and I use this example all the time, three consecutive years of price increases, and that account has been doing just fine, and the agent and the underwriter come to a conclusion that it's appropriate to renew that account flat, we are all for it. We are all for that. No issue at all. If the conversation is that the account absorbs a tremendous amount of cost and risk management and loss experience is kind of marginal and therefore we should be asking for a fourth consecutive increase, we will. It's by the account.
What is true overall is that more of our business has moved into what one would consider return good, return acceptable than it was two, three, four years ago. As those account-by-account decisions get made, they get made in a different framework of return to the company. We think that's great. Couldn't be happier about the way our field underwriters execute on that. It's not a matter of us sending out a message that says, "Stop here." We never would say that. Our message is, look at the load, look at the class of business. Do you want it part of your portfolio? Does the account have stability? Use your best judgment, do the right thing. We leave it to them. If rate were to flatten from here, and people were pursuing that, it'd be fine in the aggregate.
If rate were to go down another point and they were pursuing that strategy, it'd be fine. The rate in the end is the result of tens of thousands of transactions, not a single order that's given out to the field.
Josh, to that, I would just add that on our most challenged segments of business, we continue to get rate in excess of loss trend.
Yeah, that actually is so darn important because it speaks to the granularity of the strategy that where accounts are not meeting our return expectations, we already draw, and in effect, again, to come back to that line, we draw the line, and we're getting rate at that point exceeding loss trend. Important.
Thanks for spending the time on it. Thank you.
Thank you. Next question, please.
Next question comes from the line of Michael Nannizzi with Goldman Sachs. Go ahead.
Thank you. Maybe if we could talk a little bit about the personal agency channel a bit. Clearly Quantum is having a positive impact. It looks like you're winning business, you're getting rate and keeping retentions high, and the margins look good. Just trying to understand sort of the competitive environment there and what's giving you the ability to win consistently post that Quantum 2.0 change. Is it an area of focus, a demographic or a scale advantage possibly that building a risk profile of the book, something else that's giving you an opportunity to continue to generate positive new business there? Thanks.
Hi, Mike, this is Doreen. I'll bring you back just probably a year or so you know what we put in place, but we actually did some back testing on what we needed to do to be more competitive in the customer segment that we wanted. When we looked at that, we thought that our expenses needed to be adjusted. That's what we did. We took $140 million out of our operating costs as well as reduced the commission on the new product. What we told our agents, so first of all, I want to give them great credit because notwithstanding that sort of salary change, they were great partners and supported that. What we told them was we wanted them to be competitive across channels. We weren't doing this just so they could switch one company's business to us.
We wanted them to be able to compete with the captives and the direct writers. While I wouldn't say that the price points are always what those companies are, it's made us and them much more competitive in those customer segments. The whole thing, you need an organization that's willing to execute to that. You need strong agency relationships, and at the end of the day, you need a product that people want.
All right. Have competitors reacted to kind of post this success or are they kind of stuck with their? I'm sure you can see from your own sort of win rate, have others tried to replicate this success or are you still able to hold that lead?
We can see it in little pockets, Mike, but nobody has really taken the approach of adjusting their cost base. If they're doing that with pricing, we can see a little bit of that, but nothing where we've seen a major restructure so that the offering would be similar.
Got it. Then you mentioned direct really quickly. I'm just curious, there's obviously a little bit more ink recently on some new entrants into the direct channel. Just given that, maybe if we talk a little bit about the direct initiative at Travelers and how you're thinking about investments there and the potential for do any of these new market participants, do they provide an opportunity for Travelers or is this something that you're just kind of wait and see at this point? Thank you so much for all the answers.
Sure. On the direct-to-consumer initiative, I think Jay and Greg have talked about this in the past. We still feel like there's a very strong value proposition for the customers that we have and the advice that they get from agents. We just haven't seen sort of the run on the bank that maybe would have been predicted a decade ago. That being said, we also think it's very prudent to make sure that we have multiple channels of business because our goal is the customers that we like, we'd like to bring in the door any way they'd want to come in. All that being said, I know there's been some talk about some entrants in the marketplace. I think I would tell you our view is to just evaluate and wait and see what that might bring. We've made no decisions, and we're in observation mode.
It's Jay Fishman. I had a couple of points. Doreen's got it all right, which is that one of the things that is, you've said this before, nothing new in this, but increasingly it's in the data. The nature of the customer who chooses to buy directly is a different customer than that which seeks advice. It's not just the nature of the customer, but it's their risk management needs. Broadly speaking, this comes from our experience with being a GEICO partner for many, many years, the percentage of renters in amongst the GEICO customer is a multiple of what the renter population is in our agency business. The percentage of minimum and low limits buyers for auto who buy direct are a multiple of those low limits or minimum limits that buy directly. Now, that's not 100%, it's not an absolute statement.
Broadly speaking, you would say that the nature of the risk management needs, based upon data, not anecdotal, not observational, not television commercials, but based upon the data, is that they're different. Now that's in effect really good news for us because the agency business that we direct our attention to has been much more resilient and robust than people 10 years ago might have suggested.
Sure.
Maybe that changes in the future. It's possible, and it would be foolish of us to be arrogant enough to say that that customer could never be attracted. We're going to continue to develop the technology such that if that business dynamic changes, we're in a position to respond. Several quarters ago, our conclusion was, and we shared with you, that the nature of the customer that we seek is not yet buying enough through a direct channel that the business is a scalable business around that. To make it scalable, you have to be at that different segment, and that's just not our sweet spot.
I'd say, gee, it would have been nice if the direct channel would have developed more robustly, but believe me when I tell you, as I had to pick one versus the other going into this, I'm much happier that the agency channel has remained more robust than people would have suspected.
Got you.
Great. Thank you. The next question is our last question, please.
The next question comes, final question comes from the line of Jay Cohen, Bank of America. Go ahead.
Yes. Thank you. Got through a lot of good stuff. Just a couple of cleanup questions. The first is, can you discuss what the non-cat weather was in the quarter? Because we had all these winter storms, certainly in the Northeast. It was hard to tell what was cat and non-cat. Were the non-cat numbers elevated at all?
Yes, Jay, this is Jay Benet. We actually had only one storm raised to the level of a cat for us. Most of the winter storm activity that took place this quarter was non-cat weather as we describe or differentiate between cat versus non-cat. Looking at this year versus last, and I think Brian made the comment, the weather in this quarter was heightened from what we would think of as a, if there is such a thing anymore, as a normal quarter. It was heightened from that. Last year was also a heightened weather quarter, and while the data, it's voluminous data, but it's also challenging data to say exactly this is a weather loss, this isn't a weather loss. The feeling around here is that the weather losses this year were elevated from the weather losses last year, but they weren't off the charts different.
They were just a bit higher than they were last year. That would be the feeling we get from the data. Higher than normal, whatever normal is these days. A bad winter season last year, bad winter season this year. This year all in a little higher than last year. That would go for both cats as well as non-cat.
Then the second question was, you talked about the, I guess, some of the changes in reinsurance and the impact on premiums. Can you discuss those changes and what drove the premium impact?
Well, the primary one is the change in the cat program that we announced last quarter. As you recall, our cat reinsurance was always done on July 1. The new cat aggregate treaty that we put in place was a January 1 deal. The premium for that is primarily what we're seeing in this quarter through the ceded line that's impacting the quarter-over-quarter comparisons because last year there was no premium for that.
Because it all shows up in the quarter in which it's purchased.
In written. Yeah. In the written.
It's not amortized, to be clear.
On the written. All right. Whereas last year in the third quarter of 2014, we would have seen a similar payment taking place. Whereas on July 1 of this year, we will not be making that payment because we've already made it.
Yeah. The easier comparison in July should be up a little bit more than expected.
Yes.
Great.