United Insurance Holdings Corp (NASDAQ:UIHC) Q3 2018 Results Earnings Conference Call November 1, 2018 5:00 PM ET
Adam Prior - IR
John Forney - CEO
Brad Martz - CFO
Greg Peters - Raymond James
Elyse Greenspan - Wells Fargo
Christopher Campbell - KBW
Greetings and welcome to the United Insurance Holdings Corp. Third Quarter 2018 Conference Call. At this time, all participants are in a listen-only mode. A question-and-answer session will follow the formal presentation. [Operator Instructions] As a reminder, this conference is being recorded.
I would now like to turn the conference over to your host, Adam Prior of the Equity Group. Please go ahead, Adam.
Thank you, Kevin and good afternoon everyone. Thank you for joining us. You can find copies of UPC's earnings release today at www.upcinsurance.com in the Investor Relations section. You're also welcome to contact our office at 212-836-9606 and we'd be happy to send you a copy.
In addition, UPC Insurance has made this broadcast available on its website. Before we get started, I'd like to read the following statement on behalf of the company. Except with respect to historical information, statements made in this conference call constitute forward-looking statements within the meaning of the federal securities laws, including statements relating to trends in the company's operations and financial results in the business and the products of the company and subsidiaries.
Actual results from UPC may differ materially from those anticipated in these forward-looking statements as a result of risks and uncertainties, including those described from time-to-time in UPC's filings with the Securities and Exchange Commission. UPC specifically disclaims any obligation to update or revise any forward-looking statements, whether as a result of new information, future developments, or otherwise.
With that, I'd now like to turn the call over to Mr. John Forney, UPC's Chief Executive Officer. Please go ahead John.
Thanks Adam. This is John Forney, President and CEO of UPC Insurance. With me today is Brad Martz, our Chief Financial Officer. On behalf of everyone at UPC, we appreciate your taking time to join us on the call.
We probably should have had this call yesterday on Halloween since our quarter had both treats and tricks in it, but here we go. On the positive side, we continue to produce strong organic growth. Our earned premium was up over 13% year-over-year. Both personal and commercial lines have contributed to that growth.
On the personal lines side, the growth has been spread across our geographic footprint with the Northeast being particularly strong. At the end of Q3, over 60% of our personal lines policies were outside of Florida.
On the commercial lines side, premium growth in the quarter was over 10% with no degradation in rate. That growth trend in commercial has been accelerating, which is testimony to the strength and perseverance of the AmRisc team that underwrites and distributes our American Coastal products.
Another positive; our hurricane response, both operationally and financially, has been outstanding. Operationally, according to data provided to us by Verisk Analytics, our response time on Florence was 30% faster than the industry. And on Michael, it was over 50% faster than the industry. It's no wonder that this time around, I have been seeing a stream of e-mails from our agents and policyholders praising the service they've received from Scott St. John and his world-class cat team, led by Tim Cotton.
Financially, our reinsurance program and our reinsurance partners have excelled. We buy more first-event reinsurance limits than any other Florida-domiciled company, even those with far more exposure in the state than we have. So, program exhaustion is not something we worry about.
We're one of the very few companies where Irma never got into a single-digit rate online reinsurance layer. And since we don't buy reinstated limit, we never have to worry about surprise reinstatement premiums impacting our results.
At the bottom of our program, which is where all the action has been over the past couple of years, we bought an ex-Florida retention buy-down that helped limit our retained Florence losses. We appreciate the amazing support we've received from our reinsurance partners and we're committed to a long-term win-win partnership with them.
The final and most significant positive note for the quarter was our launch of Journey Insurance Company. We have been on a journey to create the premier specialized property cat insurer in the country and we're making good progress. From a $200 million Florida homeowners carrier six years ago, we have become a $1.2 billion geographically and product-diversified firm with business in 12 states.
We are the 19th largest homeowners' writer in the country and the number one insurer of commercial habitational property in Florida. Best of all, we've achieved that growth while producing all-in cat and non-cat loss ratios better than the industry average in almost all of our states.
So far, we've been able to accomplish all that with just the Demotech rating. But we are already the largest Demotech carrier in the United States and we have long felt that we needed to have a complement of A.M. Best-rated products in our offering portfolio as we continue on our journey.
The formation of Journey Insurance Company, which marks the fifth insurance company in the UPC family, achieves that goal. Journey is also significant because we formed it in partnership with Tokio Marine Kiln, a company for whom we have great respect and with whom we look forward to growing a long-term strategic partnership.
We will use Journey, which received an A- rating from A.M. Best, to write both personal lines and commercial lines business. Initially, our focus will be on Florida, Texas, and South Carolina, three states we know quite well and have a long track record in. But that will be just the beginning.
We have already filed our first product and hope to begin writing business before the end of 2018. At the same time, we will continue to grow our Demotech companies in their states and product areas.
On the negative side of the ledger, we had unusually high non-cat losses during the quarter, especially in August, and to a lesser degree, in July. While Q3 typically has higher non-cat weather losses than other quarters and that was true for our Q3, the real driver of our elevated non-cat losses was the disproportionately high number of claims and large losses unrelated to weather.
For example, we had 11 liability large losses in the quarter versus two in Q2 and none in Q1. Large loss incurred overall for the quarter was $10 million higher than last year's Q3, which had a more typical experience. We also received 19% more non-cat claims in August than last year, more than double the average increase for other months this year.
The good news is that these results do not appear to represent a trend. After August bad experience, September dropped back to normal levels. And October was even better than that. It's important to note that none of the loss experience is related to adverse development as our non-cat reserves were slightly redundant for the quarter.
Still, we remain vigilant in our underwriting and rate actions to make sure we can stay ahead of loss cost in our portfolio. We implemented double-digit rate increases in three of our states this year, single-digit in a few others and we are continuously doing rate indications and filings as needed.
At this point, I'd like to turn it over to Brad for his remarks.
Thank you, John. And hello, this is Brad Martz, the CFO of UPC Insurance and I'm pleased to review the financial highlights. But before we get to those, I would like to remind and encourage everyone to review our press release and Form 10-Q for more information regarding our results.
Highlights of UPC's third quarter 2018 included gross premiums earned of roughly $304 million, an increase of over 13%; a GAAP net loss of $11.7 million or $0.27 a share; non-GAAP core loss of $14.9 million or $0.35 a share.
Both bottom-line measures were driven by net retained cat losses of $34.6 million or $0.60 a share, which added over 20 points to the combined ratio. As John mentioned, it was certainly a difficult quarter from a loss perspective, so I'll discuss that in more detail along with our initial guidance for Hurricane Michael in a moment.
Some additional insight into UPC's revenue for the quarter includes direct premiums written were a mix of 80% personal lines and 20% commercial lines. Commercial lines grew approximately 13% year-over-year, slightly faster than personal lines at just over 10%. Roughly 53% of our growth in direct written premium came from Florida. And the Northeast region was our fastest-growing, up 16% year-over-year led by New York.
Assumed commercial excess and surplus lines premiums grew nearly 80% to $17.2 million in the third quarter. Our investment income increased 41% year-over-year and other revenue decreased $10 million or 73% due to the change in our presentation of ceding commissions earned that we implemented and disclosed last quarter.
UPC's third quarter losses decreased approximately 16% from $143.1 million last year to $120.6 million this year due to the full -- two full retentions incurred from Hurricanes Harvey and Irma in 2017. This produced a 39.7% gross loss ratio, down over 13 points from 53.4% last year.
Net retained cat losses in the current quarter from Hurricane Florence totaled $25 million, and the remaining $9.6 million was due primarily to the increased retention under the company's aggregate reinsurance program, which was expected.
UPC did see continued favorable reserve development on prior accident years and recognized roughly $2.7 million during the current calendar quarter. Excluding the impact of net retained catastrophe losses and favorable reserve development, UPC's gross and net underlying loss ratios were up over six and 11 points respectively compared to a year ago.
The increase in underlying non-cat losses is a challenging comparison to the prior year for two primary reason reasons: First, UPC's non-cat losses were abnormally low last year due to hurricanes Harvey and Irma impacting UPC's two largest states, Florida and Texas.
Second, we saw frequency and severity of large losses, both in personal lines and commercial lines above normal during the current quarter. Of the $27 million increase in underlying non-cat losses, approximately $12 million or 44% came from personal lines, fire, and liability-related losses in multiple states and the other $15 million or 56% was driven by commercial property losses in Florida.
Hurricane Michael, our second hurricane event of the year, made landfall in Florida during October and is not included in our loss results for the third quarter. UPC expects to retain between $25 million and $35 million of losses in the fourth quarter of 2018 related to Michael. We are providing a range because our net retention for Michael will depend on the gross loss for Hurricane Florence, which is currently estimated at $70 million.
If our ultimate gross loss from Florence is less than $70 million, UPC's Michael retained loss could be toward the higher end of the range since our retention for the second event is dependent on how much reinsurance is eroded from the first event.
UPC's non-loss operating expenses increased $7.8 million or approximately 11% year-over-year during the current quarter, driven by a $7.7 million or 16% increase in policy acquisition costs, which were consistent with premium growth. All other operating expenses combined were flat year-over-year and resulted in slight improvements in our gross and net expense ratios.
On an underlying basis, operating cost increased $17.9 million year-over-year, excluding ceding commissions in the same period a year ago. The underlying expense ratios are not a perfect comparison between the current and prior period given the current year bifurcation of ceding commissions between policy acquisition cost and ceded earned premiums.
UPC's policy acquisition cost, as a percentage of gross and net earned premiums, did rise year-over-year due to premium growth in commercial lines and personal lines outside of Florida, where our acquisition expenses are generally higher than the business overall.
On the balance sheet, UPC ended the quarter with total assets of over $2.3 billion, including nearly $1.2 billion of cash and invested assets. Unrestricted liquidity at the holding company decreased to approximately $72 million at the end of the current quarter, primarily as a result of the $40 million contribution made to form Journey Insurance Company.
Non-controlling interest is a new item for the company and represents Tokio Marine Kiln's equity investment in Journey. Shareholders' equity attributable to UIHC shareholders decreased just under $529 million with a book value per share of $12.33. Our combined statutory surplus for the group increased to $484 million at the end of the third quarter, including Journey's statutory capital and surplus.
I'd now like to reintroduce John for some closing remarks.
Thank you, Brad. One day after the end of the quarter, we welcomed Chris Griffith as our new Chief Information Officer. Chris joined us from Safety National, and he has had a lot of relevant insurance industry experience. Chris brings great technical and leadership skills to UPC and we look forward to his contributions as we continue to drive efforts to use technology as a competitive advantage for our company.
With that, we will conclude our prepared remarks and we are happy to answer any questions that anyone has.
Thank you. We'll be now be conducting a question-and-answer session. [Operator Instructions]
Our first question today is coming from Greg Peters from Raymond James. Your line is now live.
Good afternoon. Thank you for the call. I guess just starting off, you talked about your new entity, a rated entity, Journey, and your relationship with Tokio Kiln, and I'm curious if that is affected at all by the RenRe acquisition of their reinsurance platform?
Thank you, Greg. This is John and that's actually a very good and timely question. The answer is no, it's not impacted at all by their sale of Tokio Millennium Re to RenRe. This is a completely separate entity that we're dealing with and that transaction has no impact whatsoever on our relationship with Tokio Marine on Journey.
Perfect. And then just I know, Brad, you talked about your guess around what the retention might be for Michael. I know its early days. Do you have an idea on the gross loss?
And then piggybacking on that, can you talk a little bit about your expectations for reinsurance pricing, especially in these lower layers? It looked like they're going to get hit for the second year in a row.
Actually, the third year in a row. We've had five major events.
But who's counting?
In 25 months, it's not impressive. Gross losses over $1 billion; net retained, almost $170 million in 25 months, a pretty remarkable run. The modeled expected loss for Michael varies anywhere from $50 million to $120 million. We'll probably record the initial gross loss towards probably above the top of that range given some of the development we saw in the previous events in Florida, just to be conservative.
And then just on the outlook, I lost count obviously of the cat losses you guys have reported, but some perspective on what your reinsurance partners are talking about in terms of price for next year. And I know it's early still, but I'm sure there's noise in the marketplace on this already.
I'll take a stab at that, Greg, it's John. Of course, there's noise in the marketplace. But noise isn't the same thing as information and so it's hard to decipher all of that. I'll just say we truly have partnerships with our reinsurers. We buy over $3 billion of property cat, which we've seen statistics that say that makes us one of the 10 largest buyers of U.S. property cat reinsurance in the world. And the other nine on that list are AIG and Allstate and State Farm and Liberty Mutual and all the big guys, and then there's us.
So, we're very important partner to the reinsurance community and they to us, obviously. So, we don't -- we're not transactional. We're strategic with them. And if you look at our reinsurance panel over the last five years, the average return, even with the three years of cat losses that our reinsurers have earned, it's between 6% and 9%, a positive depending on the reinsurer.
And so they've earned decent returns and almost all of that is collateralized, so those are the kind of returns they're looking for even including all the cat losses. So, we have a win-win relationship with our reinsurers. We never try to be the lowest price program out there because we know we want to be in it for the long term.
So, we're not -- we're going to do what makes sense for us and for them, but we're both looking at it as a long-term relationship and not, let's see if we can, either on our side to try to see if we can get a few points off on the upper layers because the upper layers haven't been hit, by the way, on our program in any of those three years.
Or -- and I know they're not thinking, well, we're going to try to get UPC for everything we've got on the lower layer, which has been hit, as Brad said, three years in a row. We'll do something fair. It's really impossible to say what that might be right now, but I'm confident we'll end up with a deal that makes sense for us and for them.
Thank you for those answers. And just the one final question on the underlying loss in LAE ratio, which, as you have pointed out, is ticking up. As you're probably aware, other homeowners' insurance companies have talked about some erosion in the underlying profitability of their business.
And I'm curious, one, if you feel like you're getting selected against in your -- with the risk that you are taking on. And two, are you doing anything on the right front? And that's my last series of questions.
Okay. Well, I would tell you we look very closely at our results state-by-state. And as I mentioned on the call, our loss ratios, direct loss ratios, including cat and non-cat are below industry averages in -- of the nine states that comprise 95% of our business.
For the last three years, each year, below industry averages usually significantly in seven of the nine states. The only two states where there is slightly above industry averages are Rhode Island, which is a very small state for us, and ironically, Florida. And the reason they're higher in Florida then the industry average is because we have the ability to retain more cat losses than many of the smaller companies in Florida and so we do. And that's what causes our direct loss ratio in those three years to be slightly higher. Everywhere else, we're below what's happening in that state for other companies. And so we don't see any evidence of being adversely selected against.
As I mentioned in my remarks at the beginning, we have taken a lot of rate action. And I saw comments on another call recently from someone; I think it might have been the Chubb folks that said rate isn't necessarily the answer to everything. We couldn't agree with that more.
We're taking rate where we need it. But we're also using technology to help improve our underwriting. We're rolling out a new proprietary predictive analytics tool, that we've been working on for year and a half this quarter that we've been back-testing against 10 years of data that we think is going to be quite powerful in helping us improve our underwriting results everywhere, even in states where we already have great results. And we are taking rate as needed and doing other things on a technology standpoint to make sure that we're staying on top of trends.
But again, the last thing I'll say on this, if you look at the results and we spend a lot of time looking at the results here in the last month to make -- the results for August were just such an outlier from any other month we've ever seen.
And certainly, if that had continued into September and October, we would be concerned and we would tell you there's a trend that doesn't look good. It did not happen at all if you look at the charts that we looked at. It dropped right back down to what we would expect. And October was quite a good month from a non-cat perspective, September was also very good. So, we don't see anything like what happened in August repeating itself since then.
Thank you for your answers.
Thank you. Our next question is coming from Elyse Greenspan from Wells Fargo. Your line is now live.
Hi, thanks. My first question, I guess, picking up from where we kind of left off in terms of the underlying loss ratio. So, can you guys -- so I guess, the point is August was that much worse because the underlying loss ratio is above where you've been in many, many years.
And so can you give us a sense, how does October look when you say that everything kind of reverted more back to normal? Just give us a sense, would you expect, all things considered, that the Q4 would go back down kind of in line with the first and the second quarter?
Hi, Elyse, this is Brad. Yes, that's absolutely right. Everyone needs to recognize the seasonality in our business. And third quarter is always our toughest from a loss perspective. Q4 is generally one of our better quarters, so I absolutely would expect a gross underlying loss ratio comparable to what we saw in the first and second quarter.
And I agree with that, Elyse. And it really wasn't just seasonality. As I mentioned, seasonality is one thing. But the 11 liability -- large liability losses, when we typically have zero or one in a quarter drove several million dollars of cost that we just haven't seen before, and it's not being repeated.
So, there were some very strange sort of fluky things in one month, which is why -- where you can't get overly excited or concerned about one month or one quarter. You've got to look at the longer term trends, and there's -- I'm confident that our book is better than it's ever been. And our underwriting is better than it's been, and claims is better than it's been. And there's no reason to think we're not going to go back to where we were, if not better.
So, those 11 claims that you referenced, those all were in August and they were all spread across different states or were they confined more to Florida?
All 11 weren't in August. I think there were seven in August but they were spread out all over. We looked at all the big -- crazy numbers from August and some in July to try to see if there was a common theme, and there really wasn't.
Yes, I would just add that we saw the highest frequency of large losses; we characterize the large losses as anything over $100,000 in our history as well. So, it goes to John's point, where it's just highly unusual, clearly not something we expect to persist. And that drove a severity change, if you compare our severity this quarter to the same period a year ago. It was up 17%, completely abnormal.
And so when we're thinking about what falls within and outside of your cat definition, is your cat definition tied to losses that you have from PCS events?
Generally, yes. But sometimes, there are non-PCS events that do trigger our threshold being greater than $1 million.
Okay. So, the $1 million--
Greater than $1 million and multiple losses. Like one house worth $1 million or $1.5 million that burned down would not be categorized as a cat. It has to be multiple losses and more than $1 million, and almost all of them are PCS events in our catalog.
Okay, great. And then your G&A expenses were a bit higher than where, I guess, I had been modeling. And I think Brad, you had pointed to kind of the expense most -- for the most part kind of being in line with expected, but G&A did pick up sequentially, anything there? Or is it just maybe the Q2 was just -- was rolled for some reason?
In Q3, so for example, we received assessments from quasi-governmental organizations like the North Carolina JUA, FAIR Plan, that was $1 million in the quarter. We started a new contingent commission arrangements with some of our top producers. So, we put up a very conservative initial accrual for that.
We've had other agent incentives, continued hit -- growth investments in technology, but nothing out of the ordinary. No, I mean, the only -- there's probably a couple of million dollars in nonrecurring in there from the assessments and contingency. But overall, the run rate and the guidance on a gross expense ratio between 25 and 27 is really the range.
Okay, great. And then last question, I'm sorry, just go back to the underlying margin. Since it sounds like it's all kind of non-cat weather in the quarter, there was no any kind of true-up for where you were setting your picks from earlier in the year in the third quarter, correct?
All right. Thank you very much. Appreciate the color.
Thank you, Elyse.
Thank you. [Operator Instructions]
Our next question today is coming from Christopher Campbell from KBW. Your line is now live.
Yes, hi, good evening gentlemen.
Good evening Chris.
I guess I'm going to pile on the core loss ratio again. But I think John mentioned about $10 million of abnormally large losses. And if we back out that number from the one --- or from this quarter's number, we still have like a 46% core loss ratio, which is still up about 600 bps year-over-year. So, I guess, what's driving that 600 bps of, I guess, I'd call it more attritional deterioration?
As we said earlier, it was -- most of it is driven by the abnormally high level of large losses. But I pointed out in August; we received 20% more non-cat claims than we did a year ago. The average at every other month this year has been about 8% more claims. So, there was some non-cat weather and there was just -- we just had a month and a half, where we had a lot of claims reported in it.
As I said, that dropped right back down in September and into October. There's nothing we can point to, to say this is exactly the cause. We had a lot of claims for a lot of different reasons. And -- but it seemed to be limited in its duration and it's not happened in the last couple of months.
Okay, got it. And then can we get an update on your Irma gross loss numbers?
We don't have a change in our Irma gross loss number from where we were.
Okay, got it.
At this point, we're continuing to see modest amounts of new claims, nothing like we've seen other people report and some limited reopens. But we're starting to see some of that die off and we've been doing a very good job on those claims. But we're not revisiting the number that we have out there right now. We'll just have to see how things go in November and December.
Okay, that makes sense. And just kind of just a question on the tower. Like how much limit is remaining? And I know you guys have a collateralized piece of that. So, would all the collateralized reinsurance that you would have had in place when Irma hit, would all that still be available and just has any of that collateral been released?
For the 2018, 2019 program, no. I mean we basically got almost all of our limit left for the current catastrophe excess of loss programs. Florence hit our underlying layer and our non-Florida excess of loss layer, the $35 million ex $35 million, so that's gone. But it doesn't matter because as that gets eroded, the program cascades down to its retention of $35 million. And depending on what the gross Michael loss is, even if its $200 million, it's really -- it puts just a tiny dent in a $3.1 billion tower.
Got it. And what about the 2017 program, the one for Irma, would any of that collateral have been released?
We have released some collateral on the very upper layers, where we're highly confident that there would be no potential for session. Yes, there has been collateral release, but the gross loss and the reinsurance recoverables we have are fully collateralized today and we have no concerns about credit risk or collateral.
And keep in mind the--
Okay. And like how high would, like, the gross loss have to be like in the tower for that to even be like an issue for you guys? Like, if you had to take adverse development before you got to the top?
It would have to more than double from where it is right now, okay. And number one, that still would not be an issue for us because we have provisions that enable us to fall back collateral. We've actually done that and it works quite easily within a matter of three or four days with our partners. And so it's just not -- it's not a concern and it's not an issue for us.
Okay. So, if it creeped up, if it creeped up more than double, you would have a, like, your partners want you to be able to get the collateral back to fill back in that tower?
Okay, got it. And then just one final one on the favorable development year-over-year, what was kind of driving that? I noticed that was a little bit higher.
Conservative reserving philosophy. That's what's driving it. We want to aim to be slightly redundant. We don't always achieve that goal. But we see favorable trends in frequency. Frequency is relatively flat. I mean does it bounce around? Of course. But frequency has not been our challenge. It's all been on the severity side. And the runoff of the prior accident years is -- the actual development is better than the expected.
Got it. All right. Well, thanks for all the answers.
Thank you, Chris.
Thank you. We've reached the end of our question-and-answer session. I'd like to turn the floor back over to management for any further or closing comments.
As always, we appreciate everyone's interest in UPC and your taking time to join us on the call. We look forward to being with you again in a few months to talk about the results of Q4. So, thanks everybody.
Thank you. That does conclude today's teleconference. You may disconnect your lines at this time and have a wonderful day. We thank you for your participation today.