Good day, ladies and gentlemen, and welcome to the American Financial Group 2017 third quarter results conference call. At this time, all participants are in listen only mode. Later, we'll conduct a question and answer session and instructions will follow at that time. If anyone should require operator assistance during the call, please press star then zero on your touchtone telephone. As a reminder, this call is being recorded. I would now like to introduce your host for today's conference, Ms. Diane Weidner, Assistant Vice President, Investor Relations. Ma'am, we may begin.
Good morning, and thank you. Welcome to American Financial Group's third quarter 2017 earnings results conference call. I'm joined this morning by Carl Lindner III and Craig Lindner, Co-CEOs of American Financial Group, and Jeff Consolino, AFG's CFO. Our press release, investor supplement, and webcast presentation are posted on AFG's website. These materials will be referenced during portions of the call. Before I turn the discussion over to Carl, I would like to draw your attention to the notes on slide two of our webcast. Certain statements made during the call may be considered forward-looking, as defined under the Private Securities Litigation Reform Act of 1995. These statements are not guarantees of future performance.
A detailed description of these risks and uncertainties can be found in AFG's filings with the Securities and Exchange Commission, which are also available on our website. We may include references to core net operating earnings, a non-GAAP financial measure, in our remarks or in responses to questions. A reconciliation of net earnings attributable to shareholders to core net operating earnings is included in our earnings release. If you are reading a transcript of this call, please note that it may not be authorized or reviewed for accuracy. Thus, it may contain factual or transcription errors that could materially alter the intent or meaning of our statement. Now, I'm pleased to turn the call over to Carl Lindner III to discuss our results.
Good morning. As we begin our remarks this morning, it's important to acknowledge that our thoughts and prayers remain with those who have been impacted by the catastrophic events these past few months. We are grateful to our claims professionals and insurance specialists who are helping our policyholders recover, restore their businesses, and rebuild their communities. We released our 2017 third quarter results yesterday afternoon. If you'd please turn to slide three of the webcast slides for an overview. Natural catastrophes and low interest rates notwithstanding, we reported core earnings per share of $1.06 per share. These results include $0.95 per share in previously announced catastrophe losses, the primary reason for the lower year-over-year operating earnings in our P&C insurance operations. Our third quarter cat losses did not change from what was pre-announced a few weeks ago, and while significant, were within our risk tolerances.
Third quarter annualized core operating return on equity was 8.1% for 2017, compared to 12.2% in 2016. Craig and I are pleased to report meaningful operating earnings amid an unprecedented level of natural disasters in a continued low interest rate environment, especially when many peers have reported net losses and/or declines in book value this quarter. Circumstances like these demonstrate the solid fundamentals underlying our business and the value within our diversified specialty insurance franchise. Net earnings per diluted share were $0.13. These results include non-core charges of $0.82 per share to strengthen our A&E reserves, also net realized losses on securities of $0.08 per share. As announced last quarter, $0.03 per share related to the early retirement of debt.
Based on results through the first nine months of 2017, we revised AFG's 2017 core operating earnings guidance to a range of $5.90 to $6.20 per share, a decrease from the range of $6.40 to $6.90 per share announced previously. This revised range includes our results through the first nine months of the year and our expectations for fourth quarter catastrophe losses, including the California wildfires. Craig and I will discuss our guidance for each segment of our business later in the call. We continue to be encouraged by reports on tax reform and the lower corporate tax rate that both the Trump administration and congressional Republicans are advocating. This work is essential to enabling U.S. businesses to remain competitive. Beyond that, this work is essential also to ending a continuing pattern of U.S. businesses losing market share to offshore peers.
We believe it's incumbent upon Congress to close a loophole provided by affiliate reinsurance that creates an unlevel playing field to benefit foreign competitors in both the property and casualty and annuity markets and deprives the U.S. Treasury of billions of dollars in tax revenue. I'd like to turn our focus to our property and casualty operations. If you'd please turn to slides four and five of the webcast, which include an overview of third quarter results. As I noted earlier, it's been a challenging few months for the industry overall, yet I'm pleased with the strong growth across our portfolio of P&C businesses and otherwise strong specialty property and casualty underwriting profitability reported during the quarter. On slide four, you'll see that gross and net written premiums increased 11% and 13%, respectively, in the 2017 third quarter compared to the same quarter a year earlier.
Property and casualty operating earnings were 38% lower year-over-year, with specialty property and casualty underwriting profit down significantly due to higher catastrophe losses, primarily from Harvey, Irma, Maria, and two earthquakes in Mexico. The third quarter 2017 specialty property and casualty combined ratio of 99.3% was 6.1 points higher than last year's third quarter, and included 8.4 points in catastrophe losses and 2.9 points of favorable prior year reserve development. Overall renewal pricing in our specialty P&C group was up 1% during the third quarter. I'd like to turn to slide five to review a few highlights from each of our specialty property and casualty business groups. Our property and transportation group reported third quarter underwriting profitability of $6 million compared to $44 million in the prior year period.
Lower underwriting profits in our crop, property and marine, and ocean marine businesses were the primary drivers of these results. Note, though, the comparable 2016 quarter included very strong profitability in our crop business. Catastrophe losses for this group were $25 million in the third quarter of 2017, compared to $7 million in the comparable prior year period. At this point in the year, we have a more complete view of our expected crop results, which are shaping up to be a bit better than we'd planned. We expect to have a slightly above average crop year. Fall harvest prices were established at the close of business yesterday, with corn futures at $3.49, down 11.9% from spring discovery prices, and soybeans at $9.75, down 4.3%. Both are well within acceptable ranges from spring discovery prices.
Corn and soybean yields are above the five-year trend line averages, and we're pleased that our initial concerns with regards to drought conditions in the Dakotas and Montana didn't materialize to the extent that we'd anticipated. For property and transportation, third quarter gross and net written premiums were 8% and 7% higher, respectively, than the comparable 2016 period. The increase was largely the result of higher year-over-year premiums in our agricultural and transportation businesses. This growth was partially offset by lower premiums resulting from an exit from the custom bonds business, which was part of our ocean marine operations. Overall renewal rates in this group increased 2% on average for the third quarter of 2017, with most of our price increase coming from commercial auto and non-crop agribusiness operations.
Specialty casualty group reported third quarter underwriting profitability of $2 million compared to $13 million in the prior year period. Higher underwriting profitability in our excess and surplus lines, targeted markets, workers' compensation, and professional liability businesses were more than offset by lower underwriting profitability within Neon, primarily the result of third quarter catastrophe losses. Neon's book includes property cat exposed business, which is consistent with its Lloyd's peers. Catastrophe losses for this group were $56 million and $2 million, respectively, in the third quarters of 2017 and 2016, respectively. Gross and net written premiums increased 18% and 24%, respectively, for the third quarter of 2017 when compared to the same prior year period. New accounts written in our targeted markets businesses were the primary driver of the increase.
Additionally, higher premiums in our workers' comp businesses, primarily the result of rate increases in Florida, coupled with the growth in our executive liability, excess and surplus businesses in Neon, contributed to the year-over-year growth. In addition, net written premiums were higher as a result of timing of reinsurance placements within Neon. Renewal pricing for this group increased by 1% in the third quarter. Pricing decreases that continued in our California workers' comp operations were more than offset by pricing increases in our other workers' comp businesses and within our excess and surplus lines businesses. Our specialty financial group reported an underwriting loss of $3 million in the third quarter compared to an underwriting profit of $19 million in the third quarter of 2016. The decrease was due primarily to catastrophe losses in the lender-placed mortgage property book.
Most of the other businesses in this group continued to achieve strong underwriting margins. Catastrophe losses for this group were $31 million and $5 million, respectively, in the third quarters of 2017 and 2016, respectively. Gross written premiums decreased by 3%, and net written premiums increased 1% in the third quarter when compared to the same 2016 period. Lower premiums in our financial institutions business were partially offset by higher premiums in our surety business. Renewal pricing in this group decreased by 1% for the quarter, driven by a decrease in our lender-placed mortgage property insurance book. Please turn to slide six for a summary view of our 2017 outlook for the Specialty Property and Casualty operations.
Based on results through the first nine months of 2017, as well as our expectations for fourth quarter cat losses, we estimate a combined ratio between 94% and 95%, which is slightly higher than the range of 92% to 94% estimated before a very active third quarter for catastrophes. We now expect growth in net written premiums to be in the range of 6% to 9%, which is up from our previous estimate of 3% to 7%. We've adjusted assumptions within each of our Specialty Property and Casualty groups slightly. Our revised P&C earnings guidance includes an initial estimate of expected losses from the California wildfires in October.
Based on the information available at this time, we estimate a pre-tax loss from these events, net of reinsurance and inclusive of reinstatement premiums, in the range of $20 million to $25 million, the midpoint of which is approximately $0.18 per share. We now estimate a combined ratio in the range of 92% to 94% in our Property and Transportation Group, narrowed a bit from the range of 91% to 95% estimated previously. Growth in net written premiums is now expected to be in the range of 3% to 6%, a change from the 2% to 6% estimated previously. Our Specialty Casualty Group is now expected to produce a combined ratio in the range of 96% to 98%, two points higher than the range of 94% to 96% estimated previously.
Growth in net written premiums is now expected to be in the range of 10% to 13%, up from our previous estimate of 7% to 11%. The combined ratio in our Specialty Financial Group is now expected to be in the range of 88% to 90%, revised upward from the range of 84% to 88% previously announced. Additionally, we expect net written premiums in this group to be up 2% to 5%, slightly higher than our previous range of flat to 4%. We continue to expect overall property and casualty renewal pricing in 2017 to be flat to up 1%. I surely expect there to be implications from the magnitude of the recent catastrophe losses as insurers, reinsurers, and capital providers take a hard look at the erosion of surplus and what it'll take to achieve appropriate risk-adjusted returns.
We intend to be out front on improved underwriting selection and pricing for property business, and especially cat-exposed property business, in anticipation of price increases in the reinsurance markets. Our strategy is to continue to have a lower relative net catastrophe exposure when compared to industry peers. We'll give you an update on our outlook for 2018 pricing when we release our fourth quarter results. Finally, we expect 2017 property and casualty investment income to grow between 4% and 6% year-over-year. Now, I'll turn the discussion over to Craig to review the results in our Annuity segment and AFG's investment performance.
Thank you, Carl. I'll start with a review of our Annuity results for the third quarter beginning on slide seven. The Annuity segment reported $102 million in pre-tax operating earnings in the 2017 third quarter, compared to $107 million in the third quarter of 2016. Under GAAP rules, a portion of reserves for FIAs is considered an embedded derivative and is recorded at fair value using assumptions for items such as projected interest rates, option costs, and surrenders. Variances from those assumptions, as well as changes in the stock market and interest expense related to the embedded derivative reserve, are recognized through AFG's reported core earnings. Many of these adjustments are not economic in nature, but rather impact the timing of reported results.
In the third quarter of 2017, the benefit of a higher stock market was more than offset by lower interest rates, resulting in a $4 million unfavorable impact to annuity operating earnings. In the third quarter of 2016, the impact from changes in the stock market and interest rates was a positive $1 million. Annuity earnings before the impact of fair value accounting were $106 million in both the third quarters of 2017 and 2016. As you'll see on slide eight, AFG's quarterly average annuity investments and reserves grew by approximately 11% and 12% respectively year-over-year. The benefit of this growth was offset by the runoff of higher-yielding investments. Both quarterly periods included a positive impact from a strong stock market and higher than expected income from certain investments required to be marked to market through earnings.
AFG's Annuity segment reported statutory premiums of $876 million in the third quarter of 2017, compared to $941 million in the third quarter of 2016. This decrease resulted from AFG's adherence to pricing discipline at a relatively low and decreasing interest rate environment during the year, as well as from aggressive pricing by certain of our competitors. Additional information can be found in AFG's quarterly investor supplement posted on our website. Please turn to slide nine for a summary of the 2017 outlook for the Annuity segment. Based on results through the first nine months of 2017, we now expect 2017 earnings before fair value accounting for Fixed Indexed Annuities to be in the range of $395 million to $410 million, an increase from the $385 million to $405 million previously estimated. This is the third increase this year of our pre-fair value guidance.
Our forecast assumes modest increases in interest rates and the stock market in the fourth quarter, as well as a more normalized expectation of earnings from certain investments required to be marked to market through earnings. We continue to expect our 2017 pre-tax annuity operating earnings to be in the range of $370 million-$390 million. Fluctuations in the returns on investments that are required to be marked to market through earnings or large changes in interest rates and/or the stock market as compared to our expectations could lead to additional positive or negative impacts on the annuity segment's results. These earnings expectations do not reflect any potential impact from our fourth-quarter review or unlocking of the major actuarial assumptions in our fixed annuity business.
Based on premiums through the first nine months of the year and our recent level of sales, we now expect 2017 annuity premiums to be slightly lower than the $4.4 billion reported in 2016, a slight decrease from our previous estimate that premiums would remain flat year-over-year. The U.S. Department of Labor Fiduciary Rule became effective on June 9, 2017, although the DOL delayed certain requirements until January 1, 2018. As a result, insurance-only agents are able to continue selling Fixed Indexed Annuities through the end of 2017, provided the agent acts in the customer's best interests, makes no misleading statements, and receives only reasonable compensation. The DOL recently released a proposal to delay full implementation of the rule until July 1, 2019. There's uncertainty as to whether the rule will take effect in its current form at that date.
We continue to believe that full implementation is likely to cause short-term disruption in annuity premiums. Nonetheless, we do not believe the new rule will have a material impact on AFG's results of operations. We believe that our business model, which we adopted many years ago, positions us well in a changing regulatory environment. Please turn to slide 11 for a few highlights regarding our $45 billion investment portfolio. AFG's third quarter 2017 net realized losses on securities of $8 million after tax and after deferred acquisition costs, compared to net realized gains on securities of $1 million in the comparable prior year period. As of September 30, 2017, unrealized gains on fixed maturities were $533 million after tax, after DAC, and unrealized gains on equities were $173 million after tax.
As you'll see on slide 12, our portfolio continues to be high quality, with 90% of our fixed maturity portfolio rated investment-grade and 98% with an NAIC designation of 1 or 2, its highest two categories. We've provided additional detailed information on the various segments of our investment portfolio in the quarterly investor supplement on our website. I will now turn the discussion over to Jeff, who will wrap up our comments with an overview of our consolidated third quarter 2017 results and share a few comments about capital and liquidity.
Thank you, Craig, and good morning, everyone. We are pleased to report $1.06 in core EPS in a particularly challenging quarter for our industry. Slide 13 summarizes the operating earnings results that Carl and Craig have covered. The $1.06 is based on core net operating earnings in the quarter of $95 million. You will see a more detailed view of the components on page four of our quarterly investor supplement. Property and casualty pre-tax operating earnings were down by $58 million when compared to the 2016 third quarter. Significantly higher catastrophe losses were the story in the quarter for our P&C segment and for the P&C industry. We had pre-announced pre-tax catastrophe losses of $105 million for the quarter on October 3rd. These cat losses were $91 million higher year-over-year when you compare the $105 million in Q3 2017 to the $14 million in last year's third quarter.
One year ago, we were preparing to close the merger with National Interstate, which was completed in November 2016. The elimination of the non-controlling interest, which was booked in P&C other expense, and the double taxation related to National Interstate, coupled with improvements in underwriting profitability, resulted in National Interstate contributing $0.17 per share to AFG's third quarter 2017 results. Compared to only $0.03 per share in the previous year's third quarter. As Craig noted earlier, annuity earnings before fair value accounting were flat year-over-year. Pre-tax annuity operating earnings decreased by $5 million due to fair value accounting adjustments. Parent company interest expense increased by $2 million. Other expenses were $9 million lower year-over-year, primarily the result of lower holding company expenses. We had an extraordinarily high effective tax rate on core operating earnings during the third quarter of 40%.
This was driven by catastrophe losses in Neon, for which no tax benefit can be recognized. AFG's effective tax rate on core earnings through the first nine months of the year was 33%. Our full year earnings guidance through 2017 now assumes an effective tax rate of approximately 32% on core pre-tax operating earnings. Slide 14 provides a reconciliation of core net operating earnings to net earnings. In addition to realized losses on securities and a loss on the retirement of AFG's 5.75% senior notes, net earnings were reduced by an A&E reserve strengthening of $74 million, or $0.82 per share. The detail of our special A&E charge is presented at the bottom of slide 14. According to data provided by S&P Global Market Intelligence, industry three-year survival ratios for asbestos and environmental reserves were 6.4 times paid losses as of year-end 2016.
The three-year survival ratio for AFG's P&C insurance businesses now stands at 14.6 times paid losses. As indicated on slide 15, AFG's adjusted book value per share was $55.08 as of September 30th, 2017. Note that we have paid dividends of $2.44 per share year-to-date. The result is growth in book value, plus dividends was 8.3% through the first nine months of 2017. Adjusted tangible book value per share was $52.50 at September 30th, 2017. At the end of the quarter, parent cash was $435 million. We maintained sufficient capital in our insurance businesses to meet our commitments to the rating agencies, and our overall excess capital stood at approximately $1.1 billion at September 30th, 2017. This is substantially equivalent to the balance at the outset of the quarter.
The payment of the second quarter 2017 special dividend and the cat losses of the third quarter will not preclude our consideration of a special dividend later in the year. We returned $27 million to our shareholders through dividends during the quarter. We announced in August that our board had approved an increase in the company's regular annual dividend from $1.25 per share to $1.40. That was a 12% increase, and it commenced with the dividend we paid in October. The company has increased its dividend in each of the last 12 years. Approximately 4.1 million shares remain under our repurchase authorization as of November 1st. We plan to hold approximately $200 million-$300 million as dry powder to maintain flexibility for opportunities that may arise. We review all opportunities for the deployment of capital on a regular basis.
In closing, page 16 presents a single-page summary of our 2017 core earnings guidance. AFG's expected 2017 core operating results exclude non-core items such as realized gains and losses and other significant items that may not be indicative of ongoing operations. We'd like to open the line for any questions.
Ladies and gentlemen, if you have a question at this time, please press the star then the number one key on your touchtone telephone. If your question has been answered, or you wish to remove yourself from the queue, please press the pound key. Our first question comes from Amit Kumar with Buckingham Research. Your line's now open.
Thanks, good morning. Just a few couple of questions. The first question goes back to the discussion on guidance. You mentioned the California wildfire number. Can you sort of help us, maybe just state what's the dollar amount of the number that you're factoring in, just so that we're on the same page?
A range of $20 million to $25 million pre-tax after SE reinstatements and that. I think we said $0.18 per share is what we're figuring.
Okay. I missed that. Apologies for that.
That's all right. I think probably, a lot of that will fall in probably specialty casualty related to winery losses.
Got it. That's helpful. The second question I had was on the underlying loss ratio in property and transportation as well as specialty financial. Did you guys talk about why the underlying loss ratio was higher in those two segments compared to Q3 2016?
Amit, this is Jeff. For Property and Transportation, I know that you're well acquainted with our crop business. Last year the crop business performed at a very high level for us.
I'm looking at our investor supplement. When you look at the underlying combined ratio, it's moved up by 5.4 points quarter-over-quarter. If you go back and look at where we were in the third quarter of 2015, which was more of a normal crop year.
It's substantially equivalent in the third quarter of 2017 than it was in the third quarter of 2015. What you're seeing there is the impact that a very good crop year could have on the Property and Transportation segment.
Yep.
Carl previously in the call referred to the 2017 crop year as slightly above average. We'll certainly update you on that when we get to the fourth quarter results of operations. In terms of the Specialty Financial, it's our smallest P&C sub-segment.
It has a unique aggregation of businesses, many of which aren't correlated with the overall P&C markets. We've always encouraged people not to look at the loss ratio there, but to look at the combined ratio. When you look at the underlying combined ratio, that's moved by 2.2 points down during the course of the quarter. A lot of times when you have movements up in the loss ratio, there are offsets in the expense ratio that would affect that. I would continue to encourage you to view that segment on that basis rather than just taking out the loss ratio data.
Got it. The only other question I had for now was, the overall expense ratio was lower. Is that due to comp accrual numbers, or was there anything else why the overall expense ratio in P&C was lower?
That would be one element. When you look at our release as it relates to the third quarter catastrophe events, this ties into what we were just talking about, one of the offsets to the catastrophe losses after reinsurance and after reinstatement premiums was release or reversal of previously accrued profit commissions, particularly in the specialty financial sub-segment.
Okay.
We quantified that as $8 million. That also would contribute to a lower overall P&C expense ratio.
Got it. Okay. That's all I have for now. I will re-queue. Thanks for the answers and good luck for the future.
Thank you.
Our next question comes from Paul Newsome with Sandler O'Neill. Your line's now open.
Good morning. Thanks for the call. I wanted to ask first a broad-based question about possible changes, if any, in specifically how you're underwriting post the numerous hurricanes. Is it just you're going to be a little bit more tougher or even welcome higher prices? Are you actually changing specifically how you're underwriting, maybe getting out of businesses or restrictions of any kind on your underwriting perspectively post the third quarter?
I think it's some of all the above, Paul. This is Carl. Generally when catastrophes like the string of catastrophes happen, the first thing I do is ask our operating guys two questions. What did you learn? What adjustments are you going to make? Just like a lot of other companies, we're in the middle of lessons learned and also how we're going to adjust our approach to things. Every event always has some unexpected twists and turns. Harvey, I think, surprised the industry with the size of the flood component there. With the storm, just rain settling in there for days at a time and that. I'm not sure the models really assume that, for instance. The California wildfires, I think generally the industry probably thought the communities themselves would be protected, and you couldn't have the size of losses as you had in Santa Rosa.
We're right in the middle of the lessons learned part, and we're also right in the middle of adopting what our strategy's going to be, and it could be different for different parts of the business in that. Obviously, in Neon's property cat book, prices are going up, and in the D&F book. We're determining what those are. In our wildfire exposed business, we're in the middle of also reviewing that. I'd say generally, as I mentioned in our call, we intend on moving forward and adjusting prices and terms and underwriting appetite as we feel is necessary to achieve the right returns long term. That's our approach to the property business right now.
Could you also talk about the asbestos charge in the third quarter? It seemed to me quite large in size, and I would've expected maybe some asbestos charge, but seeing how it doesn't seem to ever end. If anything, it looks like it's getting worse, not better in an environment where theoretically people exposed to asbestos should have been no longer with us as we all age. I'm surprised. Should we just continue to think that asbestos reserves just need to be pumped up by somewhere in that same range every quarter, in every third quarter?
You kind of had me when you said every quarter. This is Jeff Consolino, Paul.
Yeah, that would scare us all, I suppose.
Let me start with the fact that, as you say, the third quarter of every year is when we take a more in-depth look at these legacy exposures, asbestos and environmental. We alternate years. This year is part of the every other year cycle where we have an outside expert or a group of outside experts come in and help us look at our exposures. That would include an internationally recognized expert actuarial firm, as well as other subject matter experts and expert counsel. Last year would've been the off year where we would've looked internally. When you go through and look at the history of our third quarter movements on A&E, I wouldn't necessarily compare the third quarter of 2017 to the third quarter of 2016 because the reason we hire outside experts is to augment the expertise we have with a outside or fresh perspective.
Also, since the third quarter evaluation of 2016, you would've seen that AM Best published in November of 2016 their updated look at these exposures for the industry based on their review of the statutory annual statement, footnote 33 data. The result there was that AM Best did not change their ultimate loss expectation for environmental, but moved up their asbestos ultimate expectation for the industry from $85 billion to $100 billion, which is about a 18% increase there.
In that, on the asbestos side, they took note of the fact that for workers who were in their peak working years when asbestos use peaked out, they're well on in age, as you referred to, but life expectancy. As a result, Paul, what you have is people living longer and having a longer gestation period under which they might develop a concern over some kind of illness that they would associate with asbestos exposure. AM Best and other outside experts are looking at that and saying the industry needs more pure IBNR and also needs to look at patterns for asbestos. On the environmental side, for us, it's all very site specific, and so we look at the sites. Certainly, as we said, the length of time that we're seeing to resolve these specific sites and payments for defense and evaluation are increasing somewhat.
That's what led to that move. Taking a step back, we're never going to promise that it's all behind us. These are complex exposures. We're doing our best to reserve at an appropriate level across the business, including with this, and that's the number we came up with this quarter. I wouldn't read that into anything for future quarters or future years.
Great. Thank you very much.
Our next question comes from Greg Peters with Raymond James. Your line's now open.
Good morning. Carl and Jeff, could you talk a little bit about the reinsurance structure for your crop business and possibly how it might change next year considering this year's results? Maybe broaden it out to potential changes in reinsurance costs for other parts of your business going into next year.
Thanks, Greg. Starting with crop, we make effective use in our crop business of both quota share reinsurance and also stop loss reinsurance, and we think that we've got an excellent set of reinsurers and strong reinsurance partners involved in that. Our quota share is a multi-year treaty, although we have the ability to flex the amount ceded. That would not be something that we'd be making wholesale changes to at this point. More broadly, in the reinsurance ceded part of our business, we have concluded some treaties that cover some among our 35 specialty P&C businesses in the last couple weeks and months. For those businesses which are generally not property exposed, we've continued to get very favorable terms, including favorable ceding commissions, so we're pleased with that.
We do renew our corporate property catastrophe and property per-risk treaties at January 1st, and Neon also renews its treaties at the outset of the year. Certainly, we're aware of the discussion and rhetoric about increased reinsurance costs. Notwithstanding this year, we feel like Great American, for example, has been an excellent business to reinsure over a long period of time, given our approach. We'll go into the market and work with our partners to work out whatever the best and most appropriate deal is when it comes to that renewal. I guess I don't want to really prejudice those discussions with a lot of extraneous commentary at this point, but we'll certainly let you know how it went when we get back together for the fourth quarter call.
Thanks for the answer. On the catastrophe loss in the specialty financial book, I know you talked a little bit about it in your prepared comments and in one of your answers. I was kind of surprised by the loss, and I'm curious whether it's something we should start to anticipate in our projections going forward, that there will be periodic catastrophe losses of that size in that business.
I think it's basically a property book. It's a large book. It's some $300 million or so. The book does have catastrophic exposures. In this book, we've tried to manage, as in some of our other businesses, to lean towards accounts that aren't as heavily coastal property exposed. When you're writing accounts that have broad geographic mixes, you're bound to have some coastal property in it. This was an unusual event or an unusual quarter when you had the number of sizable events that there was. I think that's the unusual thing when you look at the whole industry's numbers, company by company. They're larger than usual. Those would be my thoughts.
Greg, if I could pile on. This is Jeff Consolino. What we saw in the third quarter was a frequency of severity. Multiple events striking the U.S. and also other parts of the world. One observation we made in our pre-announcement in early October was that we have a comprehensive reinsurance program which provides us a lot of vertical cover against extreme events. When you have multiple events, we are going to be retaining a net loss before the reinsurance towers kick in. You asked about expectations going forward. Year to date, our cat loss as a percent of earned premium is below 4%, in the high threes. If you look over a 15-year period for our company, a year would typically run at about one and a half points.
Echo Carl's thought that this is an unusual quarter, anytime you have a lot of cat activity, lessons are learned, and different things happen. Years where we've had cat losses exceeding 2% of earned premium, you'd have to go back to years like 2008 with Ike and Gustav and other U.S. windstorms, 2005, which was the KRW year, and 2004 with Charley, Frances, Ivan, and Jeanne. The nature of catastrophes is they're lumpy, they're infrequent. I just gave you a number that spans a 15-year period as well as some points that exceed the 15-year average. How that plays out in any one year, it's up to providence to tell us with retrospect. What doesn't change is we are going to maintain a prudently cautious approach to accepting catastrophe risk, and we'll maintain a comprehensive outward set of reinsurance protections to manage against the extreme events.
Maybe that's one way to think about it and one way to frame the question you asked, that's what I would offer at this point.
Thanks for the color. Craig, if I could just ask one question to you, or actually in two parts. Can you update us on where American Financial is in the cycle of reduction of crediting rates and if there's any room further for reduction or if there's any plans? On the annuity sales by channel, I know you provided some color in your opening comments, I are hoping that you could provide us some additional color on sales by the bank channel, by the independent agent channel, et cetera.
Okay. Let me address your first question. We have the ability to lower credited rates on $24.4 billion of reserves by another 88 basis points. A significant amount of room. Now, how much of that we are going to implement really will be a function of the level of interest rates. Because of our model, generally lower cost of putting business on the books, we have been able to maintain credited rates longer than many of our competitors, even at a declining interest rate environment. It is nice to have the cushion that we have in the event that interest rates would fall from here, which we don't expect, but it's certainly a possibility, that we have a tremendous amount of room to continue to lower credited rates if we would need to maintain spreads and profitability. Second question, hang on one minute.
Let me get the breakdown on premiums.
I was looking at slide 15 in your supplement, I believe, where you provide statutory annuity premiums, Craig.
Okay. Let me go to that, Craig.
Our next question comes from Jay Cohen with Bank of America Merrill Lynch. Your line's now open.
I can be patient while you get that question.
I think we're going to go back to Mr. Peters' question on the annuity section for just one moment, and then we'll address Jay's question. Thank you.
Your question on the premiums, it shows us the breakdown here between retail and financial institutions. In the quarter, financial institutions, the premiums did slow a bit as we can see in the numbers. But what was your specific question on the breakdown of premiums, Greg? Yeah. Greg, we think that longer term, the financial institutions side of the business will continue to be the growth engine for us. Although, we were happy to see the retail premiums pick up a bit. Our hope is we're going to be able to, over time, continue to grow both sides of the business.
Can you guys hear me? It's Jay Cohen.
Jay, if you could go ahead and continue. I think you're next in the queue. Thank you.
Great. I'm sure Greg was very satisfied with those answers. If he wasn't, I'm sure he'll buzz in. On the property transportation, specifically, I guess it's the crop business. I probably didn't appreciate the seasonality in that business. It does now that I look back, it appears that the third quarter loss ratio tends to be higher than the first half. First, is that accurate? Secondly, why is that?
Jay, this is Jeff. You definitely are onto something there. Agriculture itself is a seasonal business, and as a result, our crop is seasonal. In general, when you look at our premiums in the property and transportation segments, and looking at slide eight in our deck, you can see a significant amount of premium coming through in the third quarter, both in 2016 and in 2017, compared to the other quarters. This is the quarter where we recognize a lot of the premium from the crop business. I would distinguish that from profit, where generally we wait to know how the crop year's turned out. We tend to have the bulk of our profit come through in the fourth quarter, once we've gotten through the averaging period that Carl gave the statistics on. That's how the crop business works.
For better or for worse, we tend to book only modest or no underwriting profit leading up to the time that we're through the averaging period, just so you may recall in past years, some other companies that didn't have the same approach would go ahead and book profits, and then if you had either a fall in price or a decline in yield or drought or other conditions, they'd have to turn around and unbook previously booked profits that they shouldn't have booked in the first place. We want to make sure that we're appropriately recognizing the profit once all the unknowns have been resolved.
Jay, the first quarter sometimes can be a true up. There are some crops like citrus and things you don't know what the answer is, or in years where there are lots and numbers of claims, it takes a while. You may not have a complete picture until you get through the first quarter. If the profit ends up being larger in those years, some of that gets booked in the first quarter of the following year.
Got it. I've only covered you guys for 20 years. I probably should have picked up on this earlier, but glad I did now, so thank you for that. Second question, on the casualty business, you mentioned there was some benefit from timing. Can you quantify what the growth rate would have been excluding that timing impact?
Jay, this is Jeff. I'm not going to quantify it right now because I want to be precise and not give you numbers. Directionally, the timing difference was with Neon. The new management team conducted their strategic review in the first part of 2016, and as a result, had pushed a lot of their reinsurance decisions until after they made a decision on what the business is going to be. That resulted in a lot of outwards reinsurance being placed for Neon in the third quarter of 2016. Once that project was completed, Neon reverted to more of a traditional January centric purchase of reinsurance. You've got apples and oranges in terms of Neon ceded premium between Q3 of 2016 and Q3 of 2017.
Got it. That means it's really just a net to gross then. Going forward, it really shouldn't have much of an impact.
That's correct. There's a lot else going on in specialty casualty premiums this quarter with growth in our targeted markets business and other lines. Neon's contribution is timing, not anything other than that.
That is helpful. Last question, Carl, you may have touched on this during your commentary, for some reason, the sound quality when you were speaking was not very good. The question is on workers' compensation. Florida arguably should be getting better given the price increases, certainly we hear about competitive conditions outside of Florida, where prices generally are going down. Net net, does workers' comp get better as you move into 2018, or is there some downward pressure on the margins?
Are we talking specifically about Summit? Is that what your focus is on?
I was thinking comp more broadly. Obviously, Summit's a big piece of it.
Yeah. I think you almost have to kind of look at the pieces in our case. I think Summit, which is Florida and Southeast, we've been real pleased with their results and by the caliber of the management team we have there. We've had good underwriting profitability since we acquired it. Still with the Florida rate change this year and the predictive analytics being used there, we feel good about this year being able to be an overall profitable year from an underwriting standpoint at Summit. One thing you may be referring to is there's a rate decrease in the state of Florida, which has not been approved yet, which is around 9% that's being proposed. I think in the case of Florida's underwriting profit, if that rate decrease is taken, will be smaller underwriting profit.
I think with that, we figure we'll still be able to make maybe a very small underwriting profit. Overall, though, Summit should make a solid underwriting profit. There are some pricing adjustments in some of the other southeastern states that has had an impact there. Generally, when you see the price decreases, it's a response towards historical accident year profitability being better. It's kind of good news, bad news. Your premium's impacted, but when we look at our overall reserve position in Summit, it's very solid. That's my perspective. For continued overall, I think in the Summit business, we'll continue to make an overall solid underwriting profit. Florida, if that rate's approved, maybe that piece still make an underwriting profit, but a small one. In California, another state that's significant for us, our outlook for this year is for a solid accident year underwriting profit.
Again, similar type of situation. A rate decline of about 10% this year, there's a small January 1, 2018 decrease of maybe somewhere in the 3%-5%, probably for us going forward in 2018 that's projected. On an accident year basis, I think we'll continue to make a small accident year underwriting profit next year, although that's still at a double digit return on equity in that. Again, because of the improving industry results and the prior accident years being better, we have what I consider to be a strong reserve position in our California comp company. We feel real positive about the claims environment, reform is holding in California. We don't see really any change in that claims environment happening over the next couple of years.
The loss cost trends and loss ratio trends seem to be very mild. We're implementing predictive analytics starting this past September in both underwriting pricing and claims there. I think that will also offset some of the impact on the rate. Those are our two biggest pockets. National Interstate writes some workers' comp. I think rates aren't really moving much either way there. That business has been profitable. That'd be my perspective on our business.
That's a great update. Thanks, Carl.
Jay, this is Jeff. Carl had encouraged me to be a little bit more precise in my answer to the specialty casualty growth. When you look at the third quarter, our third quarter net written premium in that segment is $624 million. That's up $120 million from the $504 a year ago. Out of that $120, approximately $20 million is attributable to the timing of the Neon reinsurance purchases. Point out also that Neon is growing on a gross basis beyond that and contributes to the overall growth, probably at the same level that a Summit would or something. The biggest component of the growth, when you take out the timing, is the targeted markets business, as we had indicated.
Got it. Thanks for the precision.
Again, if you have a question, please press star and then one. Our next question comes from Larry Greenberg with Janney Montgomery. Your line's now open.
Thank you. Good afternoon. Just three quickies, I think, all related to the California fires. The first, you mentioned that it was going to be mostly in specialty casualty. I assume that's Neon. I assume that would suggest maybe a slightly elevated tax rate again for the fourth quarter. Are these going to be treated as one or multiple events? Finally, just when you talk about the cat load for this, would you just be adding that to your normal fourth quarter cat load? When you think about guidance, is it netted against a piece of what would be the normal cat load? Thank you.
Yeah, maybe I'll address the question about Neon and with the wildfires. Actually, Neon is really not that big of a piece of the wildfires. It's really mainly within our targeted markets business within our specialty casualty area written by Great American.
Larry, just to continue on with that. That's why we had made the comment about at the midpoint, at $0.18 a share. When you work through that math, that would not therefore distort our expected tax rate in the fourth quarter. When I had gone through what we expected our tax rate to be for the full year, that would be considered in there. As for whether this is an event or a series of events, I think for a company in the abstract, they would have to go back and review the terms of their outwards reinsurance. Many reinsurance contracts are subject to things like hours clauses or things like contiguous or co-located events within a certain radius or diameter of a footprint.
We believe that the California Wildfires are a single event, even though they have multiple PCS numbers as you apply that to our reinsurance.
Thanks. Just the final one, would you just add what you expect from the California Wildfires to your normal cat load, or do you net it against a piece of that?
Yeah. I think, Larry, we're probably closer in our guidance to adding on top of a normal cat load. We might have made a tweak here or there on top of what we would have expected for the quarter to reflect the fact that we're into the quarter and have some events. We have additional assumptions about further cat losses, i.e., a cat load in the fourth quarter that would be greater than what we have talked about here for the California Wildfires. There's still some cat buffer, if you will, in our thinking.
Great. Thank you very much.
At this time, I'm showing no further questions. I would like to turn the call back over to Ms. Diane Weidner for closing remarks.
Great. Thank you. Thank you all for joining us this morning. We look forward to talking with you again at the end of next quarter.
Ladies and gentlemen, thank you for your participation in today's conference. This does conclude the program. You may now disconnect. Everyone have a great day.