Thank you.
Good day, everyone. Welcome to Selective Insurance Group's third quarter 2017 earnings call. At this time, for opening remarks and introductions, I would like to turn the call over to Senior Vice President, Investor Relations and Treasurer, Rohan Pai.
Good morning. Welcome to Selective Insurance Group's earnings conference call. My name is Rohan Pai, Senior Vice President, Investor Relations and Treasurer. This call is being simulcast on our website, and the replay will be available through November 26th, 2017. A supplemental investor package, which includes GAAP reconciliations of non-GAAP financial measures referred to on this call, is available on the investor's page of our website, www.selective.com. Certain GAAP financial measures will be stated during the prepared remarks that are also included in our previously filed annual report on Form 10-K and quarterly Form 10-Q reports. To analyze trends in our operations, we use operating income, which is a non-GAAP measure. Operating income is net income excluding the after-tax impact of both net realized investment gains or losses and discontinued operations. We believe that providing this non-GAAP measure makes it easier for investors to evaluate our insurance business.
As a reminder, some of the statements and projections made during this call are forward-looking statements as defined by the Private Securities Litigation Reform Act of 1995. Forward-looking statements are not guarantees of future performance and are subject to risk and uncertainties. We refer you to Selective's annual report on Form 10-K and any Form 10-Qs filed with the U.S. Securities and Exchange Commission for a detailed discussion of these risks and uncertainties. Please note that Selective undertakes no obligation to update or revise any forward-looking statements. Joining today on the call are the following members of Selective's executive management team. Now I'll turn the call over to Greg.
Thank you, Rohan, and good morning. First, I'll make some introductory remarks and then focus on some high-level themes for the quarter. Mark will then focus on our financial results, and John will review the insurance operations in more detail, providing additional color on key underwriting initiatives. The big story in the quarter was the industry's record level of insured catastrophe losses, with some estimates for the three hurricanes, Harvey, Irma, and Maria, exceeding $100 billion. Selective's third quarter results remained solid despite the higher catastrophe losses, benefiting from our disciplined underwriting, product risk appetite, and geographic concentration. We continue to execute on our strategy of maintaining underwriting discipline while finding opportunities to drive top-line growth. Our statutory consolidated combined ratio for the first nine months of the year was an excellent 92.2, an impressive result in the context of a projected underwriting loss for the industry in 2017.
Following the third quarter catastrophe losses, on an underlying basis, or after adjusting for catastrophe losses and reserve development, our year-to-date combined ratio was an extremely strong 90.7. The standard commercial lines and personal lines segments produced excellent overall results. The E&S segment generated a loss for the third quarter, and we remain focused on addressing loss development and improving underlying profitability. Our annualized operating return on equity was 11% for the first nine months of the year, in line with our long-term goal of achieving 300 basis points above our weighted average cost of capital. On a rolling one-year basis, our weighted average cost of capital is approximately 8.5%. The combined ratio for the first nine months demonstrates our strong commitment to maintaining underwriting discipline in a market that is very competitive.
Our ability to effectively balance the objectives around profitability and growth reflect, one, strong franchise model with Ivy League distribution partners. Two, sophisticated underwriting tools and processes. Three, efforts to enhance the overall customer experience in an omnichannel environment, as customer centricity is at the heart of Selective. The investments in building out these strategic competitive advantages will enable us to continue to generate financial outperformance over time. We maintained top-line growth in the commercial lines and personal lines segments, although premium volume declined in the E&S segment. The lower level of E&S business reflected efforts to improve profitability through price increases and business mix shifts. Despite overall competitive market conditions, we've been focused on obtaining renewal price increases that match or exceed expected claim inflation.
We obtained targeted renewal pure price increases in our standard commercial lines business, averaging 2.7% for the third quarter and 2.9% for the nine months. This compares favorably to the 0.6 point increase for the Willis Towers Watson commercial lines, or CLIPS survey for the first six months of 2017. We have, in fact, outperformed the CLIPS index in terms of price increases for the past 33 consecutive quarters ending June 30th, 2017, and expect to outperform in the third quarter as well. In addition, we view property pricing, which has been under pressure in recent years, as another area to drive overall pricing higher. Our margins in the third quarter remained solid. The various catastrophe events did not meaningfully impact our 24 footprint state for standard commercial and personal lines, although we did experience some losses in the E&S business, which we write across the country.
The losses from Hurricane Irma and Hurricane Harvey totaled a manageable $14 million on a pre-tax basis net of reinsurance. The industry catastrophe losses, however, do have larger implications. First, they highlight the continued coverage for GAAP for perils such as flood. Second, the losses serve as a reminder that the industry needs to anticipate and price for extreme and tail events. We take a conservative approach to managing catastrophe loss volatility by purchasing a reinsurance program that has a low retention of $40 million per event and limits our exposure at the one in 250 year return period or 0.4% event set to 3% of stockholders' equity. Our retention is $5 million per event for the two new states, Arizona and New Hampshire, and for our non-footprint states.
Turning to our 2017 guidance, we, one, are maintaining our underlying statutory combined ratio, excluding catastrophe losses of 89.5 and 3.5 points of catastrophe losses. This forecast incorporates favorable reserve development for the first nine months of the year but assumes no additional prior year reserve in the fourth quarter. Two, increasing our forecast for after-tax investment income to $115 million. And three, maintaining our estimate for weighted average shares outstanding of 59.2 million on a diluted basis. From an overall industry perspective, the return on equity will most likely be zero for the full year 2017. As we think about 2018, you've all heard me say many times that, "Arithmetic has no mercy," when it comes to improving overall results. I believe there are four key metrics that highlight industry performance as follows. One, earned renewal pure price changes. Two, expected claim inflation trends.
Three, expected after-tax new money rates. And four, the effective duration of the fixed income security portfolio. I've read a lot of commentary about projected 2018 industry underwriting performance. However, most of the opportunity to impact the 2018 underwriting performance has already passed. The industry has already recorded much of the written renewal pure price on its policy inventory that will be earned next year. You have to understand that for Selective, commercial lines renewal, pure price increases are at an industry-leading 2.9 points versus an expected claim inflation trend of approximately three points. Our year-to-date after-tax new money rate is 220 basis points or 20 basis points higher than the rate on the sales and maturities. The effective duration of our overall portfolio, including short-term investments, is only 3.6 years.
When coupled with excellent operating cash flow, running at approximately 17% of net premiums written, after-tax investment income should increase. Before turning the call over to Mark, I wanted to quickly remind you that we will be hosting an investor day in New York City on Thursday, November 9th. You should have already received invitations. If not, please contact our investor relations team, and we'll make sure that you get the details. Now I'll turn the call over to Mark to review the results for the quarter.
Thank you, Greg. I'll discuss our financial results, beginning with some key metrics and trends for the company as a whole, then we'll also touch on our segments. For the quarter, we reported $0.79 of fully diluted earnings per share and $0.72 of operating earnings per share, which is up 16% from 2016. We generated an annualized operating ROE of 10.1%. Through the first nine months of 2017, we have generated operating income of $133.7 million, which is up 14% from 2016, and an operating ROE of 11%. This is an excellent result thus far in a year characterized by significant catastrophe loss activity, continued low interest rates, and a competitive pricing environment. For the third quarter, GAAP underwriting income totaled $21 million after tax and generated five points of operating ROE.
In addition, the investment portfolio generated after-tax net investment income of $30 million, which coupled with our ratio of invested assets to equity at 3.4 times, generated 7.1 percentage points of operating ROE. Consolidated net premiums were up 4% for the third quarter and are up 5% for the year-to-date period. The consolidated combined ratio was 94.3% in the third quarter on a GAAP basis and 93.7% on a statutory basis. On an underlying basis or prior to catastrophe losses and prior year reserve development, our statutory combined ratio was 91.3% for the quarter and 90.7% year-to-date.
During the third quarter, we increased our loss pick for the commercial auto line of business for the 2017 accident year, which increased our third quarter underlying combined ratio by 1.1 percentage points and our year-to-date combined ratio by about 40 basis points, resulting in some modest deterioration in our underlying statutory combined ratio from last quarter. That said, our underlying combined ratio is still well ahead from 2016 for the quarter and year to date. After a long stretch of low to moderate catastrophe loss activity for the industry, the third quarter of 2017 was record-setting according to some estimates. Our results this quarter include $24 million of net pre-tax catastrophe losses, which added 4.1 points to the combined ratio and includes $14.4 million of net losses from Hurricane Harvey and Irma.
Of this amount, $6.2 million relates to Hurricane Harvey and resulted from losses incurred primarily in our E&S segment. Hurricane Irma resulted in $8.2 million of losses, primarily in our standard commercial line segment and principally from losses in Georgia and South Carolina. The remaining catastrophe losses relate to other events. We incurred no losses from Hurricane Maria. During the quarter, we experienced $9.9 million of net favorable casualty reserve development, which reduced the quarter's combined ratio by 1.7 percentage points. Our practice of setting loss and expense reserves is disciplined and for the past 11 years has produced favorable casualty reserve development. Continued favorable loss reserve development in our workers' compensation and general liability lines were partially offset by adverse development in our commercial auto line of business and in our E&S segment. Moving to the segments.
Our standard commercial lines segment net premiums written were up 5% for the third quarter and 6% for the first nine months of the year. The quarter's GAAP combined ratio for commercial lines was 92.1%. Results benefited from $19.9 million or 4.5 points of net favorable prior period casualty reserve development. The commercial lines segment also experienced $14 million of catastrophe losses, which added 3.2 points from the combined ratio, including $7 million from Hurricane Irma. In our standard personal lines segment, net premiums written were up 7% for the third quarter and are up 5% year to date, benefiting from increased opportunities in personal auto. Personal auto net premiums written were up 13% for the third quarter of 2017. The personal lines GAAP combined ratio of 88.7% was impacted by three points of catastrophe losses.
We are pleased with our personal lines book, with homeowners operating close to our long-term targets on an underlying basis and auto positioned for a favorable pricing environment. Our flood operation contributed $2.2 million of claims handling fees during the quarter, with $1.2 million related to Hurricanes Harvey and Irma, with Harvey driving the majority of these fees. In our E&S segment, net premiums written declined 4% for the third quarter but are up 2% year to date. The GAAP combined ratio of 120.4% included $10 million or 18.4 points of adverse prior year casualty reserve development and $7.3 million or 13.5 points of catastrophe losses. The prior year reserve additions principally relate to higher than expected loss severities for accident years 2015 and prior.
While we are disappointed with the reserve strengthening, we believe the steps we have been taking to increase pricing and improve the business mix will result in improved margins over time. The catastrophe losses this quarter included $5.3 million from Hurricane Harvey. Moving on to expenses. Our overall GAAP expense ratio was 33.9% for the third quarter, which was down 180 basis points from the comparative quarter. On a statutory basis, the expense ratio was down 130 basis points from the year ago period to 33.2%. We continue to seek out areas of efficiency and cost savings while continuing to invest in our employees and key initiatives around geographic expansion, enhancing our underwriting tools and the overall customer experience. For the first nine months, the statutory expense ratio was down 100 basis points compared to the prior year period, reflecting our disciplined approach to expense management.
Other expenses, which are principally comprised of holding company costs and long-term compensation, were also down relative to the comparative quarter, but elevated relative to our expectations. This is primarily due to higher costs related to long-term stock compensation arising from the strong appreciation in our share price through September 30th. As we highlighted last quarter, we continue to expect expense savings in this line item due to the restructuring of a long-term stock compensation plan we completed earlier in the year, although there is some volatility in this line item given the variable accounting treatment for a portion of the plan. Turning to investments for the quarter, after-tax net investment income was 19% from a year ago. Fixed income securities, which represent 91% of our portfolio, experienced an increase in after-tax net investment income, resulting mostly from a higher book yield.
Our fixed income portfolio is highly rated with an average credit rating of double A minus and a 3.6 year effective duration, including short-term investments. The after-tax yield on the fixed income portfolio averaged 2.2% during the quarter, compared with 2% a year ago. The new money after-tax yield on the fixed income portfolio during the third quarter was 2.3%. Alternative investments, which report on a one-quarter lag, reported a strong pre-tax gain of $2.7 million in the quarter. The results were driven by solid performance in our private equity portfolio. Risk assets, which principally include high yield fixed income securities, equities, and alternative portfolio, are up modestly to 7.6% of total invested assets from 7.1% at year-end. We've been gradually diversifying our investment portfolio and will likely continue to modestly increase our risk asset allocation over time, depending on market opportunities.
Our balance sheet remains strong with $1.7 billion of GAAP equity at September 30th. Book value per share increased 3% for the quarter and is up 10% for the first nine months of the year, benefiting from strong earnings and net unrealized investment gains. We have increased our quarterly dividend by 13% to $0.18 per share after taking into account our strong financial and liquidity position. We are adequately capitalized to support our expected growth and continue to target a premiums to surplus ratio of approximately 1.4 times, which is about twice the industry average. We continue to adopt a conservative stance with respect to managing our underwriting risk appetite, investment portfolio, reserving processes, reinsurance buying, and catastrophe risk management. This allows us to maintain higher operating leverage than the industry as a whole, with each combined ratio point equating to about a point of operating ROE.
This model positions us well to generate superior returns in today's low interest rate environment. With that, I'll turn the call over to John to discuss our insurance operations.
Thanks, Mark, and good morning. Our focus remains on finding opportunities for profitable top-line growth by leveraging our franchise agency relationships, unique field underwriting model, and sophisticated underwriting tools, which allow us to segment and price risks on a granular basis. We've often talked about our long-term growth strategy within our footprint states for commercial lines. We seek to appoint new agents in our current markets to represent 25% of available commercial lines premium and growing to a 12% share of wallet with our appointed agents. Together, these targets represent an approximately 3% commercial lines market share and an additional premium opportunity of more than $2.5 billion in our 24-state footprint. So far this year, we've appointed 73 new agents in our footprint, excluding our new states of Arizona and New Hampshire, as we continue to drive our market share higher.
Effective July 1st, we opened two new states under our geographic expansion program with the launches of our product in Arizona and New Hampshire. We opened Arizona with 16 agency appointments and New Hampshire with 10. In each case, these agencies control about 25% of the available commercial lines premium. We are pleased with the receptivity and performance in these two new markets and expect to write approximately $9 million of net premiums written for the year. Our regional office in Arizona will serve as the underwriting hub for our Southwest expansion, which also includes Colorado in early 2018 and New Mexico and Utah by the end of next year. We continue to invest in tools and technologies that strengthen our underwriting capabilities and allow us to obtain the right price for a given risk.
Our new underwriting insights tool, which we deployed to our new business underwriters earlier this year, provides real-time insights into how each prospective account compares to similar accounts already in the portfolio. This tool provides our field underwriters with high-quality information at the time of decision and should improve overall risk selection and operational efficiency. Turning to our operations, our standard commercial lines segment, which represented 79% of total year-to-date net premiums written, again produced extremely strong results. Solid net premiums written growth for the quarter was driven by stable retention of 85%, new business growth of 9%, and renewal pure price increases averaging 2.7%. Overall competition remains fairly aggressive, especially for larger size new business accounts and low hazard workers' compensation. We constantly monitor our renewal portfolio on a granular level based on profitability expectations using our Dynamic Portfolio Manager underwriting tool.
For the highest quality standard commercial lines accounts, which represented 49% of premium in the quarter, we achieved renewal pure rate of 1.2%. On the lower quality accounts, which represented 11% of premium, we achieved pure rate of 6.5%. This granular approach to administering our renewal pricing strategy allows us to achieve additional loss ratio improvement through mix of business changes while continuing to deliver pure rate increases that equal expected claims inflation. Drilling down to the by line results for commercial lines, our largest line of business, general liability, reported $11 million of favorable prior year casualty reserve development for the quarter, which related primarily to lower than anticipated claim frequencies and severities for accident years 2015 and prior. We achieved renewal pure price increases of approximately 1% so far this year.
Our workers' compensation line experienced $14 million of favorable prior year casualty reserve development for the quarter as a result of lower than expected severities for accident years 2016 and prior. Workers' compensation pricing has been flat with a 0.3% average price increase so far this year as loss cost filings by NCCI and other individual state bureaus have been trending negative overall. Commercial auto remains an area of focus as we take steps to improve the overall financial results. For the third quarter, commercial auto experienced $5 million of unfavorable prior year casualty reserve development, primarily relating to higher claim frequencies and to a lesser extent, severities for the 2015 and '16 accident years. In addition, we strengthened current accident year reserve for this line by $6.5 million, reflecting higher than expected loss trends for recent quarters.
We've been actively implementing price increases, which average 6.7% so far this year, and believe the elevated loss trends support the need for additional rate going forward. In addition to price increases, we've also been actively managing the new and renewal books in targeted industry segments, reducing exposures and increasing price on higher hazard classes. In personal lines, which represented 12% of total year-to-date net premiums written, margins overall have been profitable for the quarter and first nine months. The homeowners line experienced strong profitability in the quarter, with an 80.3 statutory combined ratio. We continue to target a 90 combined ratio in a normal catastrophe year, or one that has approximately 14 points of catastrophe losses on an annual basis. Renewal pure price increases across our homeowners book averaged 2% for the first nine months of the year.
In personal auto, we've seen greater top-line growth opportunities as competitors have been increasing prices to reflect higher loss trends. A sharp increase in new business volume has resulted in a pickup in premium growth for the personal auto line in recent quarters, after years of flat or declining premium volume. Renewal pure price increases on our book averaged 3.5% for personal auto liability and 3.6% for physical damage during the first nine months of the year. Our 2017 personal auto rate filings include a fully earned impact of approximately 6%. Our E&S segment, which represented 9% of total year-to-date net premiums written, generated a statutory combined ratio of 120.1 for the quarter. Results in this segment were impacted by elevated catastrophe losses, primarily from Hurricane Harvey, as well as $10 million of adverse reserve development relating to accident years 2015 and prior.
We've been taking deliberate steps to push pricing where appropriate and let go of business that does not meet our profit targets. Overall price increases averaged 5.5% so far this year, with significantly higher rate increases in the casualty lines of business. We closely monitor pricing by segment and remain comfortable with rate levels on the new and renewal inventories across the book. We also expect changes to our claims operation to drive improvement in both loss and expense outcomes going forward. Our strategy has been to drive profit improvement and allow the top line to float down if the market does not support our pricing stance. That dynamic was clearly present in the quarter. Small, low hazard E&S remains an attractive market to complement our core commercial line segment. We continue to invest in our underwriting, pricing, and claims sophistication in order to deliver consistent profitability in the future.
As we look out to 2018 and beyond, we maintain our keen focus on balancing our objectives around growth with our longer-term profit targets. We continue to invest in strengthening our franchise agency relationships, enhancing our sophisticated underwriting tools, and building out our customer relationship capabilities, which together underlie our high-tech, high-touch operating model. With that, we will open the call up for questions. Operator?
Thank you. We will now have the question and answer session. If you would like to ask a question, please press star followed by the number 1. Please unmute your phone and record your name so that I can introduce you. If you wish to recall your question, please press star followed by the number 2. One moment, please, for our first question. Our first question is from Arash Soleimani with KBW. Please go ahead with your question.
Thanks. Just had a question on the E&S development. Can you just provide a bit more detail around what exactly drove that? I know you said it was severities in accident year 2015 and prior, it was a pretty large amount, just want to get more color in terms of what that amount consisted of.
Arash, thank you for the question. This is John. I will start, and then certainly Greg or Mark could jump in as well. There is not a lot more to tell you other than what you have just read back from our prepared comments, which it was in fact 2015 accident year and prior, and it was severity driven. I think from our perspective, frequencies have held pretty steady with what our expectation would have been for that accident year, and the severity is what is driving the movement in the quarter. I will also reiterate, we have said this in the past relative to our reserving philosophy overall, which is we do have, and knowingly have, a tendency to react more quickly to what may be negative news in the reserve portfolio, and that philosophy continues to apply to E&S as well.
Beyond that, let me just make a couple of other comments relative to how we think about this business and how we think about our profitability in the current year and how we think about it going forward. A couple of points just to reinforce what we have talked about in the past, and again, tie this back to the fact that the development came from 2015 and prior. Starting in January 2016 is the point at which we made all the changes in our claims handling process to merge this operation under the management of our overall claims organization. That is an important point because at that point in January of 2016 through the current accident year, is when our claims organization has had their hands on files since day 1, as opposed to 2015 and prior, which were transition files.
Not that you can't improve the outcomes from both a loss and an expense perspective on transferred files. It's a little harder to do than it is when you manage the files from day one. I think that's important to understand when you think about what provides us context and informs our view relative to how we think about the more current accident years and how we think about the future. The other point I'll make relative to the more recent accident years, and this was also referenced in the prepared comments, is we very carefully and in a very targeted way, measure pricing on new business and the renewal portfolio by geography and by segment.
As we've started to talk about dating back to middle part of 2016, we have been achieving our target pricing levels pretty much across the book for both the new and renewal inventory. We've also said very clearly in this segment, which we've said over time, we would like to be in that 10%-15% kind of range for the overall company. We were willing to take some top-line volatility in order to manage bottom-line performance on a more consistent basis. We think we're on track to moving towards that profitability target, and what you saw in the quarter, and have seen on a year-to-date basis, is that top-line pressure coming through, and that continues to be a trade that we're willing to make. Not a lot more specifics relative to the development, but this is fairly straightforward. It's general liability.
It's low limits profile and some very lower hazard segments in this industry, and that's where the development was.
I don't really have anything to add to that other than the fact that what John touched on it, and that's really about this was part of the transition, and I think the transition reflected the fact that we wanted to make sure we addressed the reserve position for the prior years. Nothing's ever final, but we wanted to make sure that as we move into the fourth quarter and into next year, we really felt solid about the positioning of the overall book, which hopefully you got from John's comments. We do feel positive about the overall book of business.
No, thanks. That's very helpful. Just to extend the question a little bit, when I look at the 3Q17 core loss ratio for E&S, backing out the reserves and the cats, and also adjusting for, I think you guys had 840 basis points decline in non-cat property losses in E&S. When I back that out of 3Q16, it implies still about 230 basis points of core loss ratio improvement in the E&S segment. My question there is that improvement just a function of what you said in terms of January 2016 and forward, the changes you've made? I guess what I'm trying to ask is you're not seeing any signs that would imply that loss picks should be higher in E&S going forward.
It seems like this was really just all legacy stuff that doesn't really have any bearing on trends you're seeing on business you're writing today. Does that make sense?
Clearly, what you're seeing, you pointed out too, is a big part of that is improvement relative to non-cat property. The other point that's driving that, and this kind of refers back to Greg's overall comments in the beginning, which is when you think about performance, you need to think about earn rate level relative to loss trend. We have been earning rate in this book of business, and that will have positive impact year-over-year on the underlying loss ratios.
I would focus more, when you start to look at underlying performance, I would have a tendency to focus more on the year-to-date number. You're going to get lumpiness quarter-to-quarter relative to property activity, both favorably and unfavorably, and I don't view that necessarily as an indication of either a positive move or a negative move. Again, when you look at this book, the underlying combined ratio is about a 91 in the overall book, I will tell you that's what I would have a little bit more of a focus on going forward rather than just a quarter.
Okay. I guess to ask it differently, the stuff that caused you to record the adverse development, that doesn't imply any trends that would make you think the initial loss pick should be higher than you may have thought a few months ago.
We would have adjusted. You would have seen the changes then in the 2016, 2017 year, and this was, as John mentioned, 2015 and prior.
Yeah, just to add to that, Arash, if you go through your math, you're right. If you back out, if you look at it on an accident year basis, excluding CATs and non-CAT property in the quarter, the underlying so to speak, loss ratio is about 230 basis points better than the year ago period and about 170 basis points on a year to date basis. That, as John mentioned, and Greg as well, really reflects the earn rate coming through the book of business. Which would reflect an improved profitability on the underlying book of business. The reality is the combined ratio in the quarter was 120, which is not something we're happy with.
From an underlying trend perspective, the earn rate is coming through the book, and that combined with the focus on the underwriting mix improvement and the claims handling outcomes, should help drive our profitability in the future periods. Unlike commercial auto, where we increased the current accident year, the development related to E&S was 2015 and prior. It didn't impact 2016 or 2017.
The only point I'll add, Arash, this is John.
Sure.
We've talked about this in the past. Mark referenced mix improvement. Just remember, this is a business where you could drive mix improvement more quickly because the typical retention in this segment is in the mid to high 50% range. You have the opportunity to turn over your renewal inventory more quickly, and that helps drive mix improvement when you target certain classes of business, either for a heightened rate or in certain cases, active non-renewal.
Thanks. That makes sense. What's a fair target combined ratio, did you say, to think about for this book? I don't know if you've put one out there, but just.
What we've generally said is we look at this segment running a few points better than our overall commercial lines, but that would take us a little bit of time to get there. We also told you that this is a business that we're going to manage more by ROE, and as John indicated in his comments, let the top line flex with market conditions. In favorable market conditions, you'll see this unit grow more rapidly. In highly competitive markets, you'll see it shrink.
In terms of the competitive conditions, I know you also mentioned in the earnings release that it is highly competitive right now. Does that imply the next few quarters that if that kind of trend persists in terms of the competition, the book should continue to shrink near-term?
I would say that looking out, we don't want to get too far out. I would say there's obviously some segments that we're reviewing today. Generally, I would say this market condition is pretty aggressive right now. It's pretty competitive when I say aggressive. Pretty competitive right now.
The primary pressure point for us on the top line in E&S has been new business, year-over-year has been down significantly, and I would attribute that more to our unwillingness to flex on our pricing stance. Whereas the renewal inventory has held pretty steady even with some additional rate going through it.
Thanks. The comment you made about the book should run a few points better than the commercial book. What's the kind of rough timeline for that? Is that something that you expect next?
I would say that you've got to give us a little bit of time for our claim changes to manifest themselves and our picks. These things that we are changing, that fundamentally we think reduce our pure premium costs going forward, have time to really bake themselves in. I think from an expense scale standpoint, we're pretty much there today where we thought the book needed to run. I would say it's a combination of continuing to drive rate, looking at a couple of segmentations in there that we want to make sure that we feel we can hit our long-term targets in. The overall part then would be just in terms of ongoing new business and new business growth rates.
Okay, thanks. That's helpful. I just wanted to touch on the guidance real quick. I know on the combined ratio guidance, you kept that intact. Last quarter on the call, you talked about how you took it from a 90.5 to an 89.5 to give credit for the one point of favorable development. Then if you back out that point of favorable, you get to a 90.5 core. Should we still think of the core as a 90.5, or is there any update to that based on the favorable development you had in 3Q? Just want to make sure I'm interpreting the guidance.
Yeah, no, we haven't changed anything. Just remember, when you look at always the movement from nine months to 12 months, there's always a spike of about 50 basis points in our expense ratio as a result of how premium volume comes in in the fourth quarter. It's our lightest premium volume quarter. It probably represents a little under 23% of the premium that we would normally. If you look at a quarter at 25%, the premium volume in the fourth quarter is lower. You always get a little bit of an increase of about 50 basis points from nine months to 12 months in the statutory expense ratio. I think going forward, and I know Mark would love to see us do this, is just convert all of our guidance to GAAP, that's something that I know we would be looking at.
Since the guidance was out in a form of statutory, we've left it that way for the year, that would avoid some of the confusion around the combined ratio of movements.
Okay. Yeah, I was going to ask the same thing. I was going to say, going forward, it would be, I think, very helpful, yeah, if it was on a-
You're just building up the feathers in his bed for this one.
Well, thank you very much for all the answers.
Thank you.
Thank you.
Thank you. Our next question is from Mark Dwelle with RBC Capital Markets. Please go ahead with your question.
Yeah, I'd say that last round of questions pretty much exhausted most of what I had.
Okay, go ahead. Fire away, Mark. How are you today, by the way?
Doing fine.
I love your pieces, by the way. Read them religiously.
I guess since we covered a lot of ground in the last set of questions, I guess the one item that I feel like still remains out there is, you commented correctly that a lot of the action that impacts the 2018 numbers has kind of already been taken from a written standpoint. How are you approaching the opportunity to try to seek broader rate improvements? Is Selective's position such that there is more rate to get, or you kind of kept up a good rate increase clip right along the way, maybe you don't have as much upside to grab as some of your less diligent competitors?
Yeah, Mark, this is John. I'll take a stab at that, and others can fill in as well. I would say if you look back to starting in 2009, not that we've ignored the marketplace that we operate in, but we've really focused on our target combined ratios and the pricing of our inventory in a very granular way and tried to manage it in that way. If you look at our performance, and we constantly highlight our performance relative to the CIAB commercial lines pricing survey, it's been very strong, and we've kept retentions high throughout that period of time. Our philosophy's not going to change. We're going to continue to manage the business that way.
We're going to continue to manage it to our target combined ratios, and more aggressively manage that 10% or so of our renewal inventory that we think is really driving overall performance in a negative way. At the same time, make sure we're creating the oxygen in our pricing strategy to protect that 50% of the business every year that actually exceeds our margin targets. That'll be our philosophy. That said, we highlighted a couple things in the course of the prepared comments. Number one would be workers' comp pricing continues to be very pressured, and it's also a very competitive segment for new business.
That's a little bit harder to control company by company, because in many places you're dependent upon loss cost filings from either NCCI or the individual state bureaus, and it's harder to overcome what may be a decline in a loss cost filing, and you're seeing a lot of those come through. I think that's going to be a negative for the entire market going forward. Pricing on the other hand, where you've seen our property rather, where you've seen a lot more discussion following the catastrophe activity, has been a line that is in need of more price for the industry.
What you may have seen in the third quarter is a wake-up call for that particular, we're getting a little over six and a half points of rate on that line, but we expect that those trends will support the need for market pricing to continue to move for the commercial auto line of business. When you put it all together, our philosophy and our approach to managing pricing will continue to be the same.
Mark, here Greg, I would just say there's no company that you're going to speak with that has such a tight vertical lens into pricing. Let me just say, on the ground level, how it really happens is we have our inside folks that work with our agencies 60-90 days out. They've got the inventory, as John articulated. They've got it by cohort. Their ability to dial up or dial down pricing is very agile, and that agility comes through every single month. We can see that. We can see it by region. We can see it by agency. We can see it by inside underwriter. I would just answer your question as what I hear is a lot more tailwind than headwind right now. Let's be realistic.
I mean, the ROE at zero this year for the industry, that's nothing for anybody to be proud of. Irrespective of these hurricanes, it wasn't going to be much better than 500 or 600 basis points without this large activity, which is nothing to be proud of either. Thinking about cost of capitals in the 850 range and other things, the industry needs to get a certain amount of discipline. I think this whole tail event scared a lot of people as the mega event that could've come up the spine of Florida could've been extremely catastrophic. I think this has people thinking of reinsurance programs where they protect themselves. I think people are thinking about counterparty differently, retro programs differently. These are all things that they should've been thinking about way before any of this had happened.
My guess is that, I want to make this point, if you don't get overall rate, then you got no rate. I don't want to hear about commercial auto rates at 10 and giving it all back in GL and comp. Overall rate, if you're not getting overall rate, then you're getting no rate. When you think about claim inflation in this business, it's all predicated on medical. Medical permeates everything that we do, medical trends, if anything, are edging higher. They're not edging lower. We sit there and go property and everything else, this should provide a lot more tailwind for us. You got to remember, we're very dialed in on the amount of inventory we really are touching. Again, if any company's got agile information, sophistication to do it, and a deployment into a highly franchised model, that's us.
Okay. I very much appreciate the comments. Thanks. That's all my questions.
Thanks, Mark.
Thank you. As a reminder, if you would like to ask a question, please press star followed by the number one. Our next question is from Paul Newsome with Sandler O'Neill. Please proceed with your question.
Good morning, Paul.
Good morning. I thought you might want to weigh in a little bit more on the impact of these catastrophe events that we've had kind of broadly. My question really is this. In your opinion, has there been a lesson learned from these events? I think back of another hard market, 9/11 taught us about clash risk. The hard market in the '80s taught us about the potential for very large corporate lawsuits on the liability side. There seemed to be sort of a specific lesson that told underwriters they need to do something different. I wonder if there's a lesson here to be learned, in your opinion, from an industry's perspective and from an underwriter's perspective.
Yeah, this is Mark here. Why don't I jump in and then Greg and John can follow up as well on the topic.
Greg talked a little bit about the cat loss activity this quarter, and the need to price for it and get adequately compensated for the risk the company's taking. I think from our perspective, when you look at our business model and the way the quarter shook out for us
Whether the event that took place happened exactly as it did, which for us hit two big E&S states, Texas and Florida. It was a good test case of the relatively new E&S segment for us, the reinsurance programs we have in place, and worked out well for us in terms of a relatively modest impact, two and a half points on the combined ratio. If the storm track had been different, if Irma had that sort of really significant impact or if, in particular with Maria, if the Bermuda High hadn't shifted to the east and Maria had hit the East Coast with a CAT 3, really significant CAT event.
I think again, for Selective, with the reinsurance protection that we buy and the way we think about catastrophe risk management, I think it really showed the strength of the franchise and the business model that we have to have a sort of a lower volatility underwriting result and good reinsurance program in place. From a lessons learned perspective, I think there's always lessons that are learned from each new catastrophe. I think time will tell what lessons will come out of this. I do think staring down that Thursday forecast, the really bad Hurricane Irma forecast, really caused people to sit back and really think about the business model, really think about the value of reinsurance and how important that is as part of an overall primary operation, and as Greg mentioned, the counterparty credit risk. From a claims handling perspective, we'll see what happens.
We saw after Gustav and Ike in 2008, a lot of sort of latent claims development as the billboards went up in Texas and litigation reopened claims. We got the AOB issue in Florida. I think that'll be exacerbated with Hurricane Irma. Time will tell what lessons will be learned. From our perspective, we feel pretty good about the business model, the underwriting risk appetite that we have, and again, the reinsurance we have to protect the balance sheet from significant events like this.
The only other point I would add, Paul, this is John, is relative to flood exposure and aggregation of flood exposures. Prior to this quarter, there was a lot of talk about how the private market was going to really step in and provide capacity for a flood, either in an excess or a wrap or in certain cases, a complete takeout of business that's currently in the Write Your Own program through the NFIP. Based on this experience, I think understanding the aggregation of flood exposure in addition to the wind exposure in some of these geographies that you saw impacted may have some companies rethinking their appetite for that business. Now, we're a significant player in the Write Your Own program. We think that program can be shored up in certain ways, especially relative to pricing per unit of exposure, true exposure.
I think you may see some companies reexamine their appetite relative to providing private flood insurance, especially in higher hazard zones.
Paul, Greg, just again, I think you got a pretty good sense from everybody. To me, it's always been about collateral counterparty retro. I will tell you, our folks have been dialed in on that for years. We've been pushing on disclosures, on trying to get deeper into retro programs. I would say our folks have been very good at capturing collateral where the opportunities have been out there in the marketplace to better collateralize your reinsurance programs to the extent possible. I would bet you there's many companies now that are sitting there looking at attachment points and return periods and questioning whether or not their return periods were adequate relative to an event set that large. It's not only a matter of the property side.
As reinsurers have become more diversified, you got to cross issues relative to the fallout into the casualty programs as well. As companies stretch to get diversified and they write more liability as a CAT writer, that's also another long-term exposure that's not easily truncated.
Great. Thank you very much.
Thank you. At this point, there are no further questions on the queue.
Well, I thank you all for attending. I know Rohan is excited about the investor day that's coming up. As I've indicated to you many times, we are really looking for a great day and hopefully everyone will be there in attendance. Thank you very much. Any follow-up, please reach out to Rohan and his team, make sure that you get on the invite list, and thank you very much for participating in today's call.
That concludes Selective Insurance Group's third quarter 2017 earnings call. Thank you for your participation.