James River Group Holdings Inc (JRVR) 2017 Q4 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Fourth Quarter 2017 James River Group Holdings Ltd. Earnings Conference Call. (Operator Instructions) As a reminder, this conference call may be recorded.

I would now like to turn the call over to Kevin Copeland, Head of Investor Relations. You may begin.

Thank you, Nicole. Good morning, everyone, and welcome to the James River Group Fourth Quarter 2017 Earnings Conference Call.

During the call, we will be making forward-looking statements. These statements are based on current beliefs, intentions, expectations and assumptions that are subject to various risks and uncertainties which may cause actual results to differ materially. For a discussion of such risks and uncertainties, please see the cautionary language regarding forward-looking statements in yesterday's earnings release and the Risk Factors section of our most recent Form 10-K, Form 10-Qs and other reports and filings we make with the Securities and Exchange Commission. We do not undertake any duty to update any forward-looking statements.

I will now turn the call over to Bob Myron, CEO of James River Group.

Thank you, Kevin. Good morning, and welcome to our fourth quarter earnings call. This is Bob Myron, and here with me are Sarah Doran, our CFO; and Kevin Copeland, our Chief Investment Officer, who also heads Investor Relations.

I'm going to jump right in. We had a poor fourth quarter because of approximately $30 million of loss reserve development in one large Commercial Auto account that became evident during our year-end actuarial studies. At this point, we've been writing this new type of Commercial Auto business for several years, and the underlying loss data is more mature. As part of our year-end reserve work, we looked at each accident year in this division, and with appropriate consideration for pricing and restructuring changes by year and, in some instance, by contract, we are comfortable with our ultimate loss pick by accident year as of December 31.

I am confident this issue is behind us. The account in question grew significantly from 2015 to 2016 in terms of miles driven as well as geographic scope. It has run much better in 2017 than in 2016 as a result of changing -- changes in pricing and terms and conditions and has been recently renewed effective March 1, 2018, with additional changes to pricing and terms and conditions. As a result of the disappointing fourth quarter, senior management bonuses were significantly reduced and mine was eliminated.

During 2017, our core book of business was strong enough to generate another year of profitable underwriting even as we put up increased loss reserves in the account that I just mentioned. We ended the year with a strong balance sheet. Our IBNR as a percent of total net reserves is 65% on a group-wide basis, which is a level that we are very comfortable with. Our held reserves continued to exceed the point estimates of our independent actuary.

I'm bullish on our 2018 prospects for several reasons. The first is the renewal of the Commercial Auto account that I just mentioned. We are pleased to have this renewal completed and expect to have a significant relationship with the insured going forward. Second, in core E&S, new business submissions were up 9% in the fourth quarter of 2017 compared to 2016. Also in core E&S, our pricing study showed an average rate increase of 6% in the fourth quarter. This was the largest quarterly increase we have seen in 10 years, and rates continued to rise significantly in January.

In our Workers' Compensation book, our loss index adjusted rates increased by 4.9% year-over-year in the quarter. In our Casualty Reinsurance segment, contract terms in our third-party quota share business improved. Moreover, consistent with our own experience in both E&S and admitted businesses, underlying rates strengthened by 3.6% in the fourth quarter on the business ceded to our reinsurance segment.

In a couple of our E&S divisions, we have seen a significant number of large accounts come into the E&S space at significantly increased pricing relative to expiring pricing in the admitted space. The first is general liability coverage for certain classes of restaurants. The second is liability coverage in our Allied Healthcare division, which is comprised of classes such as nursing homes. This business has performed poorly in the admitted market and is being non-renewed and is now ending up in the E&S market with substantially increased pricing with higher retentions, lower limits and tighter coverage forms. It is clearly a hard market in this division right now, and we are capitalizing on it.

Lastly, we have seen a steady flow of opportunities for growth in our fronting business within our specialty admitted segment. We expect growth both in the number of accounts as well as in the segment overall. Given these factors, we expect to report a 2018 combined ratio of between 94% and 97% and to earn a return on tangible equity of 12% or greater.

I'm happy to give more color on our business during the Q&A session and to answer other questions. But first, I'd like to turn the call over to Sarah Doran to provide more insights into our reported results and our plans to react to the new tax regulations in the U.S. Sarah?

Thanks, Bob. Good morning, everyone. As Bob noted, despite the fact that we raised E&S reserves substantially in the fourth quarter to address the weakness in a single E&S account, we ended the year with an underwriting profit and a well-reserved balance sheet. In 2017, we made underwriting profits of $5.8 million, generated an operating profit of $47.4 million and are reporting net income of $43.6 million.

Investment income -- investment results were very strong in 2017. Net investment income increased 16.1% to $61.1 million, and invested assets grew 8.7% to $1.4 billion, alongside our continued growth in operating cash flow. Our renewable energy partnerships and other private investments generated an exceptional return of 22.4% on the year. Our fixed income portfolio, which we report as all other investment income, generated 9.4% more income in 2017 than it did in 2016.

We did not experience any catastrophe losses this quarter, and we did experience modest favorable development on the catastrophe losses experienced during the third quarter of 2017. The approximately $3 million of takedowns were within the E&S and Casualty Reinsurance segments and related to the Florida and Texas events.

We pay a great deal of attention to our expense ratio, as you can imagine, which decreased from 31.2% in 2016 to 24.3% in 2017. The reduction reflects a few things, including lower acquisition costs on our growing Commercial Auto book and the growth of our fronting business in our specialty admitted segment that comes with fee income which is booked as an offset to expenses. We, this quarter, made a refinement to certain accruals related to the change in business mix in the E&S segment, and that resulted in a $4.5 million or 2.2 point reduction to the group's expense ratio in the fourth quarter.

Finally, as we mentioned in our press release last night, because of our disappointing underwriting results, we reduced bonus accruals by approximately $5 million. Also, as we mentioned in our press release last night, we've made some changes to our corporate structure, which we believe will minimize the impact of the new U.S. tax law on our results. The outcome of this is that we anticipate our effective tax rate in 2018 will be in line with our effective tax rates over the last 5-or-so years or, more specifically, in the low double-digit range.

Effective at 1/1, we will restructure our internal quota share to cede to a newly formed, wholly owned Bermuda Class 3A reinsurer, which we have named Carolina Re. Through the end of 2017, our internal quota share had been reinsured to our Bermuda-based reinsurance company. Instead, Carolina Re will be owned by our U.S. companies and will make a 953(d) election to be a U.S. taxpayer. Our Bermuda-based reinsurance company, through which we also write our third-party Casualty Reinsurance business, will write a stop-loss policy for Carolina Re to provide it with an additional layer of support. Carolina Re will pay the Casualty Reinsurance business a market rate premium for this cover.

We do not expect this new structure to impact the location of capital within our group. And earlier this morning, A.M. Best issued a press release confirming that our ratings are unchanged for the new structure. It's our expectation that, over a period of years, the group's tax rate will creep up.

We will continue to rate our third-party book of Casualty Reinsurance business, but premiums in the segment are likely to be scaled down as compared to 2017 as we look to optimize our return on capital through what has generally been a better-returning business in the U.S. However, if conditions change, we will be opportunistic. We expect moderate growth in gross written premiums across the group in 2018, growing in insurance and shrinking in reinsurance.

Turning back to the past year. Our tax rate for the 2017 year was 21%, considerably higher than our 5-year historical average of 10.5%. Our lower underwriting profits in our E&S and Casualty Reinsurance segments led to a higher-than-average tax rate. Because of our historical internal quota share, a portion of any loss generated onshore is actually realized in Bermuda. As a result, we earned a much larger percentage of our earnings in the U.S., and they were taxed at a higher corporate tax rate. While the fourth quarter tax rate was exceptionally high, keep in mind we are taxed on an annual basis, making the quarterly tax rate less relevant.

The Tax Cuts and Jobs Act of 2017, as everyone knows, reduced the U.S. federal corporate tax rate from 35% to 21%. In reaction to this, we reduced our deferred tax liability to reflect the lower rate, which resulted in a reduction of $3.5 million to our deferred tax liability and a commensurate increase in operating income.

We continue to enjoy strong increasing cash flow from our businesses driven by our growth. Operating cash flow for 2017 was $218 million as compared to $154 million for the prior year. We ended the quarter with tangible shareholders' equity of $474.5 million, basically unchanged from the $472.5 million we had at the end of 2016. We paid $50.6 million of dividend and special dividends during 2017. Our assets are being put to effective use as premium or operating leverage rose to 1.56x at year-end as compared to 1.09x at the end of 2016.

Bob, I think that covers everything on my list. Let me turn it back to you.

All right. Thank you, Sarah. Operator, we are ready for the Q&A session. Can you please open the line for questions?

(Operator Instructions) Our first question comes from the line of Mark Hughes of SunTrust.

Could you give a little more detail on the new terms and conditions on the renewal on the large account? And is there any change in the geographic scope of the business in 2018?

Yes. Sure, Mark. This is Bob Myron. I don't -- because of the fact that it's for an individual insured, I don't want to get into a tremendous amount of detail here in particular around pricing and the like. I can tell you from a geographic perspective that, going forward, as of March 1, 2018, we're expecting to write approximately 40 states within the U.S. We know that this is the largest insurance relationship that this client has and will be on a going-forward basis, and we expect that our premium -- gross written premium from this account is going to grow in the 2018 policy year relative to the 2017 policy year.

Could you talk about the 2017 losses or how you set your losses in 2017 for the same account? If we think about both the updated pricing for 2017 and then your loss pick, presumably your loss picks were higher in 2017 on this piece of business. Could you give us some sense of how much more careful or conservative you were in 2017 versus 2016 on this business?

Yes. I think that it's -- again, I don't want to talk about specifics of pricing changes and the like, but there were pricing changes. There were also changes to terms and conditions and how the contract -- the deal was structured. And I think we found ourselves in a -- with respect to the 2016 year, where, obviously, the -- we know now the contract did not have enough premium with respect to the ultimate underlying loss activity. But there were significant changes made, again, to pricing, terms and conditions, how it was structured and how we are booking this going forward. And I would go back to my earlier comments -- or my prepared comments, which is that we went through every year for this Commercial Auto division and, in some instances, looked at it by contract, and we feel comfortable where our ultimate picks are booked as of the end of the year. And that -- we are comfortable -- from management's review, our IBNR percentage is comfortably high, and we've got carried reserves above our third-party actuarial point estimate, which is another point of validation for us in terms of reserve adequacy.

And a final question. When you think about loss picks for 2018 in E&S, you were at about 74% excluding the past adjustment. Is that a good place to start for 2018?

I don't know that we want to guide to a specific accident year loss pick for 2018, Mark. I mean, I think we want to just stick with the guidance that we've given from a combined ratio and ROTE perspective.

Is there -- let me just ask this. Are there any factors that should -- given the -- your flow of business, given the pricing -- the updated pricing for this particular account, is there any reason it should go up?

Not that we see right now, no. I don't think -- we wouldn't be expecting a substantial change, but I think we want to avoid guiding to a specific number.

Our next question comes from the line of Matt Carletti of JPMorgan -- I'm sorry, JMP Securities.

Bob, sorry to go back to the large contract, but I think it's a really important thing to understand and for The Street to get confidence that it is behind you and then, clearly, your confidence comes through on that. Can you -- I know we don't want to touch on pricing or things like that too much, but can you go into a little bit about -- in 2016, was it the strong growth of that client in terms of whether it's miles driven and geography that -- was that the surprise and that the premium was relatively fixed compared to that? Or can you help us a little bit with -- I'm just trying to get a better feel for the specifics around 2016 that wouldn't apply to '17 and '18.

Yes. So I think it's a combination of the business is new, new to the insurance marketplace and, obviously, new to us. We had some limited loss history because of the geographic expansion and miles driven. In terms of the premium that we wrote there, it was about 300-plus percent higher than it had been in 2015, so substantial growth for us. And then I think, as the geographic expansion occurred, not all the states end up performing exactly the same way. And so we took some corrective actions in that respect in 2017, and we'll continue to refine that going forward. But that's really the reasons behind it.

Okay. And then, not on pricing but more on terms and conditions, can you give us a feel for kind of the sorts of things that might have changed there? Is it loss corridors? Is it -- what sort of things changed over the last couple of renewals to kind of help balance out the risk? I think that would be of help.

Yes. I think that it's a number of things. Without getting into exactly what they were or what they could be, it's what share of the risks that we are on, the various types of risk through the various periods in terms of the coverage that applies during the rise. It's retentions. It is state-specific. It's a number of different factors.

Okay. Right. And one more, if I could, away from that topic. Could you just -- on the tax commentary and specifically around kind of implications going forward, Sarah, I appreciate your comments on expecting Casualty Reinsurance maybe to shrink a little. Can you help us with 2 things? One is what order of magnitude maybe should we think about in Casualty Reinsurance in terms of what is more offshore versus onshore or you view as core profitable versus a little more not meeting your return hurdles?

And then kind of as we think further ahead, can you help us with why write Casualty Reinsurance -- why keep it if it's not needed for kind of offshore tax purposes? And along those lines too, longer term, what is the attraction of staying in Bermuda if it's not tax and you're primarily a U.S. player so, at least in my view, the flexibility of capital is a little less of an issue?

Sure. I'll kind of take those questions in various order, if that works, Matt, and then come back if I've not been clear. But I think on Casualty Re, this year we wrote top line of almost $240 million in that business, which was almost 30% growth over last year. I would expect us to be significantly down from that. I mean, I don't want to, obviously, avoid your questions. I just don't want to get into the game of giving top line guidance on any of the segments. But if we think that our GPW (sic) [GWP] as a whole, as a group will grow modestly this year, I would expect to see all of that growth in insurance and significant shrinkage in Casualty Re. One thing to point out though is that business earns over more than a year, so we'll see some lag in the earn coming in from the growth of 2017. So that's one point.

On the tax piece -- so there are 2 pieces to the tax piece: a, it's setting up the new Carolina Re company, which will be owned by our U.S. companies; and that quota share -- our quota share going into that company rather to the -- rather than to the JRG Re, our stand-alone Bermuda reinsurance company. But the other piece that's as important is the capital to support the stop-loss policy that JRG Re, our third-party casualty business will write for that. And that helps not only for capital. It also has -- as an aside, there is a tax benefit just given the premium that will be written out of that segment. We couldn't enjoy the capital benefit or the slight tax benefit that comes through that if we didn't have the third-party business. So that's one thing just to kind of address.

The other though is, in our Casualty Re business, over the last few years, I do think it's very fair to say that the business has improved. I think we continue to take some hits from earlier business. But I think we believe, with the team that we have, if we pare down that business, we can really optimize for better-returning business and make that make sense for our overall group return hurdles. So you're right. I mean, we -- I could not argue differently that we make better returns generally in our U.S. business, but I do think there are some well-performing pieces of our Bermuda reinsurance business, and that business will continue to add a benefit.

The last thing I'd say is that, as of today, we hold most of our capital in Bermuda. And if we -- that obviously provides a significant capital flexibility as well as an add-on tax benefit, and not having that any longer and not having the Casualty Re business and the ongoing ability to write business there would likely impact that. So to get to a low double-digit tax rate, which is a very effective and efficient structure for the group and a really solid outcome from the tax changes that came out of the U.S. at the end of December, this is the best structure for us to move that forward as a group.

And let me just chime in on that. A couple of business thoughts here. We really like the team that we have here. They've done a really much improved job for us in the few years that they've been here. And I think, for the first time, we are seeing improvement in terms and conditions, and given that it's a quota share book, this is mostly -- almost completely manifested in lower ceding commissions, right, in terms of how reinsurance contracts get priced on a quota share basis.

And we're also still continuing to see underlying rate increases. And I think the refinement of the book is not certain, but it's going to likely be moving more and more towards underlying E&S general liability type of business, which is already a material portion of the book as it is. And so -- and we've had good experience with that in this team's tenure. So that's my two cents as well as why we're still positive about the Casualty Re segment.

And our next question comes from the line of Randy Binner of B. Riley.

I had a few on reserves. So did -- and I apologize if I missed this, but did you break out what the reserve activity was in the quarter at -- in E&S outside of the Commercial Auto?

We just included them in one.

Yes, we did not.

Just total number.

Can you share that?

We did not have any takedowns in the core business this quarter, if that's where your question is coming from, Randy.

Right. Okay. So the core E&S book was just flat. Is that right?

Yes. And I guess you could have -- you can get that from the fact that the large account had $30 million of development and, basically, the segment had $30 million of development, right? Yes, so the core takedown was flat.

And so -- and then to your comment, Bob, about being above the point estimate, are you -- was that in reference to the Commercial Auto book or the E&S book overall?

It's the entire group.

Can you scope that a little more for us, kind of where you're sitting in that range and how that conversation went outside of the single account?

I'm not sure exactly what you're getting at, but...

I'm just trying to get a sense of how -- so the story with James River has been one of reserve redundancy in your core E&S area over the years. The Commercial Auto segment was already -- it was deficient last year, so there was already some indication it could be deficient. But I guess what I'm getting to is how I should think about how the rest of the business would do from a reserve perspective based on kind of what we're seeing as a result of your reserve review.

Yes. I don't think we've ever broken that down relative to where we are booked relative, for example, to the third-party actuary by segment, by division and so on and so forth. I think that what I would say is that, as you pointed out, we've had a historical track record of reserve redundancies. That in part has been framed by, in prior periods, we have general -- management has generally sought to be above the carried loss pick of the independent actuary, and this is a large, internationally well-known actuarial and consulting firm. And so that gives us some -- that makes us feel good about where management has selected its reserves. And I think that, you know what, really, while we could provide a breakdown and we haven't, the overall is what matters.

But what I would say though is I think -- not to get into the details, the geography of it hasn't changed materially year-over-year, Randy, where those pieces are coming from.

Yes, that's right. Yes, I guess we're comfortable disclosing that. I mean we're not seeing any substantial swing in where -- when you look at by segment or division, where the deficiency or redundancy and so on and so forth. It's not materially changed from where we've been historically.

Okay, that's helpful. And then just you mentioned in your opening comments that some restaurant and Allied Healthcare liabilities were shifting to the non-admitted market. Can you kind of share with us just briefly kind of what went wrong with those items -- or those risks, rather? And what kind of term and pricing flex do you get on those when you get a fresh look at them?

Yes. I think what went wrong is they've been in the admitted space for some period of time. I think that it has been either between loss cost inflation or just underlying performance. And I think, in particular, nursing homes, some of these accounts had been in E&S. They went -- had been in E&S several years ago, went to the admitted market and are now coming back into E&S. And I think it's just a pretty simple situation of being pretty well -- by going to the admitted market, they got substantially lower pricing and much better terms and conditions, much better rate and form and broader coverage.

And 4 to 5 years later, as the business has not performed well, it's getting non-renewed in the admitted space, and then it's coming to us without a lot of options, so to speak, and coming through late and it needs to get placed. And therefore, there is significant price flexibility on our behalf and be able to set not only appropriate pricing but terms and conditions and things like that, I mean, like what's the size of the deductible and so on and so forth that really make a big difference. And so those are 2 areas where we're seeing strong flow coming out of the admitted space at substantially increased pricing, and it's really just -- it's because of performance in the admitted market.

One more on that. Are these like litigated claims? Is that the problem? Or is it more of just a general underpricing problem relative to....

I think it's general underpricing. I think the loss ratios are just well in excess of what are -- were sustainable based upon the pricing that was being received for them.

Our next question comes from the line of Meyer Shields of KBW.

Just a couple of small questions. One, Bob, you talked about having 40 states with this large corporate account in 2018. How does that compare to 2017?

Yes. It's -- 2017 was 49. We had all but Texas previously. But I think the important thing to note here is that we are still the largest insurance relationship, and we are -- we expect the account to grow from a premium perspective because of miles driven, pricing and so on and so forth.

Okay. No, that's helpful. That makes sense. In the past, I guess, using the quota share arrangement, there was at least an understood minimum premium requirement in terms of actual third-party premium volumes. Does that concept still matter based on the new operating structure?

You know what, that's always been, I would say, an indirect kind of frame of reference, Meyer. But it does matter, obviously, to be able to write the stop-loss.

Yes, it still matters.

Okay. No, that's helpful. And 2 other questions. One, you've talked about the price increases in core E&S. Can you give us a sense in terms of how much of the Excess and Surplus Lines segments you're considering to be core E&S and what's not included in that?

Oh, sorry, yes. Maybe that's just our defined term. We're defining core as everything that's not Commercial Auto, so this is all our legacy business before we started doing Commercial Auto. So it's manufacturers and contractors and general casualty, GL slip and falls for condominiums, it's professional liability, it's the Allied Healthcare book, energy, all of that stuff, everything, all the other divisions.

Okay. No, that's very helpful and clarifying. And the last question, I apologize for trying to -- for my clumsiness in phrasing it. But I'm assuming that the premium volumes that are going from U.S. taxable entities to nontax entities are going to go down on a year-over-year basis. Does that mean that the mix of underwriting profits and investment income is going to shift to what it had been? In other words, since more premiums will stay on paper tax under the U.S., that the net tax rate on investment income will be high -- yes, the net tax rate on investment income will be higher and underwriting income will be lower. Is that a reasonable way of thinking of it?

Yes. It's really just a stop-loss where the delta is right now, Meyer, because that's the only piece that's changing. And that -- I think what you're saying at the latter part of it is, over time, I think it's fairly likely that we'll have more premium in the U.S. and -- relative to what we have in Bermuda. So that's where I talk about the tax rate kind of creeping up over time.

And investment income.

Which will drive investment income as well, just to kind of draw that line. So that is not an overnight feature, but that's likely the dynamic over multiple years.

Our next question comes from the line of Brian Meredith of UBS.

A couple of quick questions here. First, sorry to beat on this large contract, but a couple of other questions on that one. I guess the first question is, did the experience you had in this contract make you -- or cause you to make any changes with the rest of your kind of Commercial Auto, rideshare-type business, kind of recognized something that maybe you should be doing?

No, it has not. It has not. I think a lot of those accounts are smaller; in some respect, more regional, not as geographically spread. They can be structured very differently. So -- but it's a good question, but it really hasn't caused us to change how we thought about anything else. And I -- that's an opportunity to say -- for me to say our performance on the rest of the Commercial Auto book has -- other than this 1 year from this one insured, has been very good. This is really -- so the years prior to '16 and subsequent to for the insured have been fine, and all the rest of the Commercial Auto has performed very well.

Got you. And then just curious, the reduction in, call it, 9 states, was that by kind of your own doing, saying, "Listen, these are states that we're uncomfortable being in?" Or was it just it ended up going to the bid market or different carrier with just a more competitive quote?

I just think it's natural from a diversification standpoint and given how large the overall account is. And so we probably don't want to say any more than that.

So I mean, as you can imagine, it was negotiations. But I think we can say that they all stayed in the E&S market just given the nature of the risk, I would imagine. We don't know for sure.

Got you, okay. And then, Bob, can you talk a little bit -- aside from what happened with this contract, just what are you seeing with respect to loss trend right now on your E&S book? And then maybe also, kind of what's happening with Workers' Compensation insurance right now?

Sure, yes. So what was -- the first question was loss trends on?

Just loss trend on your E&S business ex this whole Commercial Auto thing. Kind of what are you seeing right now? I mean, it has been fairly benign, I think, you've all talked about. Are you seeing any pickup?

I would say, overall, not a lot. I think it's still reasonably benign, and I think we're happy with loss emergence. On Workers' Compensation, what we have -- we saw was we've had several really strong years from a low reported loss ratio perspective. The first 9 months of the year were -- would have been higher than what the run rate has been for 5 years, and then it moderated pretty significantly in the fourth quarter. And so all of that, when put together because it's sort of ends up being a multiyear analysis, but the trends on underlying loss cost continues to go down, and that is impacting pricing. And therefore, that's why we like to look at pricing on a loss cost index adjusted basis and try and look at it and say, okay, how much margin do we think -- where is margin expansion going relative to where loss cost is going. And that's what I've cited earlier. That's what I was talking about with, I don't know, the 4.5% or 5% net increase that we saw.

So I think Workers' Compensation is in a fine spot from a margin perspective. We made a decision to go out and place 50% quota share of that business effective October 1. I think that, that was a bit of an opportunistic play because we had the ability to place a quota share at -- with an attractive ceding commission relative to what our own acquisition costs were. And furthermore, we've got some -- on this individually written Workers' Compensation book that we had, we had some states expand -- we're having some modest state expansion into some of the neighboring states that we have been in. And when you do that, it's never a bad time to buy a little bit more reinsurance. And so we've got some attractive pricing, and we thought we could take a little of the potential volatility about moving into new places. And that's what we decided to do.

Great, great. And then just curious, on that quota share, was there an unearned premium that went with it? Or was it just quota share just on go-forward business written?

There was not.

(Operator Instructions) Our next question comes from the line of Mark Hughes of SunTrust.

In looking at your current accident year loss picks in E&S, you started at about 69% in early 2016; and then in the fourth quarter, that stepped up to 74%; and in the subsequent quarters, sort of 74%, 75%. Were you starting to see this already in the fourth quarter of '16? And could we assume the sort of stepped-up loss ratio through late 2016, early 2017, is that one of the reasons you're a little more confident in the 2017 reserves?

So I think the movement of the accident year loss ratio, Mark, is really the mix shift with the growth of Commercial Auto. And I think that you saw that pick up a fair amount, obviously, over the last 18 months, and we're likely at a point of stabilization there so I think that's really what that's reflecting throughout the whole book as well as on mix.

Yes. I'm just thinking you're putting the 2016 business on the books. Your earned premium growth really accelerated in Q2 and Q3, but your loss pick was still 69%, 70%. Then it was only in the fourth quarter where you had a similar rate of growth that it stepped up to 74%. So it seems like the mix was already shifting earlier in the year, and then it was later in the year that we saw the loss pick go up. So I hear what you're saying, that the mix shift and timing may be influencing all this, but it seems like perhaps there was some recognition that you needed to put some extra reserves against that business and that carried over into 2017.

I'm just looking through the loss picks in the E&S business and -- or the accident year loss ratio, and they're pretty consistent at this kind of low to mid-70s numbers, with the exception of the third quarter, which obviously had the cat events in it.

Yes. No, that's right. I'm going back to 2016.

Yes. Well, I mean, the growth in this account was material towards the end of 2016, which is obviously what we're reacting to now.

Yes. And I guess what you're getting at is, yes, I mean, we have -- we do -- as we've mentioned before, we do book this -- that division at a higher loss ratio because it has a lower expense ratio than the rest of the E&S business. So that unquestionably is a consideration when we look through the reserve adequacy at the end of the year. Obviously, where we had, had those [cede] at the end of the year, obviously -- where we had had those years booked at those higher loss ratios that are weighting in was a consideration there, right? No question.

And then the 6% increase in pricing, is that mix driven or some mix impact in that? If you look at renewing accounts, sort of similar risk, similar end markets, is pricing up that much or a little bit less?

Are you asking if it was -- let me answer the question this way. I would say that it was pretty broad across the divisions in terms of where are we getting rate increases. And part of this is what we know -- what the market will bear and also because we're out there seeking those rate increases as well, right? Now I think that further to the comments that I had, right now, we're seeing the largest increases if it's a renewal in the Allied Healthcare area, but there are several divisions that are driving -- a significant number of the divisions that -- of the core divisions that are driving that overall rate increase without Allied Healthcare leading the way. But it's not -- that's -- it's not like 1 or 2 divisions weighting that to that higher number, if that's what you're getting at.

And I'm showing no further questions at this time. I'd hand the call back over to Bob Myron for any closing remarks.

Thank you, operator. Thanks to everyone for your participate -- participation on this call, and we look forward to talking to you again next quarter.

Ladies and gentlemen, thank you for participating in today's conference. That does conclude today's program. You may all disconnect. Everyone, have a great day.