濱特爾 (RNR) 2018 Q2 法說會逐字稿

Good morning. My name is Angela, and I will be your conference operator. At this time, I would like to welcome everyone to the RenaissanceRe Second Quarter 2018 Financial Results Conference Call. (Operator Instructions) Thank you.

I would now like to turn the call over to Mr. Peter Hill. Please go ahead.

Good morning and thank you for joining our financial results conference call for the second quarter of 2018. Yesterday, after the market closed, we issued our quarterly release. If you didn't get a copy, please call me at (212) 521-4800, and we'll make sure to provide you with one.

There will be an audio replay of the call available from about 1:00 p.m. Eastern Time today through midnight on October 3. The replay can be accessed by dialing (855) 859-2056 or +1 (404) 537-3406. The passcode you will need for both numbers is 18690172.

Today's call is also available through the Investor Information section of www.renre.com and will be archived on RenaissanceRe's website through midnight on October 3.

Before we begin, I'm obliged to caution that today's discussion may contain forward-looking statements, and actual results may differ materially from those discussed. Additional information regarding the factors shaping these outcomes can be found in RenaissanceRe's SEC filings, to which we direct you.

With us to discuss today's results are Kevin O'Donnell, President and Chief Executive Officer; and Bob Qutub, Executive Vice President and Chief Financial Officer.

I'd now like to turn the call over to Kevin. Kevin?

Thanks, Peter. Good morning and thank you for joining today's call. I'll open with an overview of our performance for the quarter and highlight how I think our strategy performed. Bob will then discuss the financial results. And finally, I will give you a little more detail about what happened in each of our segments before taking your questions.

Last night, we released our second quarter earnings. I'm proud to say that we had a very strong quarter, reporting growth in book value of 4.3% and growth in tangible book value per share plus accumulated dividends of 4.9%. We also reported an annualized return on average common equity of 18.6% and an annualized operating return on average common equity of 20.3%.

These returns reflect the strong performance we've delivered across all aspects of our business as well as our continued focus on controlling expenses. This has allowed us to head into the 2018 wind season just as strong financially as we were at this time last year. And with almost $350 million of operating income in the first 2 quarters of 2018, we have more than earned back last year's loss. We also benefited from higher investment income as well as significant prior year favorable development on the 2017 Catastrophe Events.

As Bob will discuss in greater detail, we grew premium in both property and casualty. I believe it is a significant accomplishment to achieve the growth we have so far this year. This is all the more impressive because we have built the best portfolio that we have had in many years in terms of its capital efficiency, its profitability and its level of risk. Even though this portfolio is larger and more profitable, due to the innovative capital structures we are able to bring to risk using our integrated system and gross-to-net strategy, we required no more surplus than we did last year.

I will provide some more color later in the call in what happened at June 1, but I'm confident that we selected the best risks and constructed an industry-leading portfolio that will generate shareholder value over the long term. For some, however, the June 1 renewals probably did not live up to their expectations. This seems to have resulted in a fair amount of pessimism in the reinsurance marketplace. Frankly, I'm confused by the general sense of malaise. This is the best market we have seen in years. As I've said at the beginning of the year, it feels like the wind is at our backs again. This quarter, you saw what a reasonable tailwind can do for our results.

I think part of what is driving all the pessimists is too much focus on price. Long ago, we recognized that changes to our market are more secular than cyclical due to the increasing efficiency of the reinsurance marketplace. Consequently, we built a business model that can compete as long as prices remain above expected loss. Of course, better pricing makes things easier. But to succeed in this market requires not just great underwriting, but increasingly great portfolio construction. It also requires having great partners and the ability to deploy all forms of capital. Oscar Wilde once said, "a cynic knows the price of everything and the value of nothing." Prices is where we start, not where we end.

Because of our gross-to-net strategy and integrated system, we have more tools at our disposal to aid us in building efficient portfolios of risk. As a result, although we seek rate increases, we are not dependent upon them to produce superior returns. We recognize that the marketplace was changing, and we moved ahead of the change. We're innovators of third-party capital management. We methodically drove down our expenses and our cost of capital. We increased operating and underwriting and investment leverage. We diversified and reduced volatility.

By doing all these things, we reconstructed the foundations of our company, allowing us to become uniquely positioned to navigate the evolution of the supply chain over the next 5 years. Today, we are leaner, more focused and have more options at our disposal. I should also point out that our ongoing success and continued growth trajectory is at least partly due to our increasingly differentiated market position as a standalone reinsurer.

During the quarter, we executed on a significant volume of strategic multiline deals for core clients. We were first call on these opportunities for many reasons, and not competing with their clients was one of them. So even with all the pessimism permeating today's market, I am optimistic about our future and believe that we feel -- believe that we have the right strategy to continue to deliver shareholder value.

Before I pass off to Bob, I should mention that we celebrated our 25th anniversary as a company last month. This happens to be my 5th year as CEO, and I'm heading into my 23rd year as an employee. We made this anniversary about our people, which we always credit as being our most important resource. I'm so proud of what they have accomplished, not only this quarter, but over the last 25 years.

Moving on, I'll discuss our business segments, recent renewals and future opportunities in greater depth later on in the call. But first, I'll turn it over to Bob to give an update on our financial performance.

Thanks, Kevin, and good morning, everyone. We had a strong quarter. And today, I'll focus my comments on the key financial drivers of our performance. I will also give you some additional insights into our investment portfolio and capital management.

In summary, there were several drivers of our performance this quarter, including favorable prior year development from the 2017 Catastrophe Events, higher net earned premiums stemming from targeted top line growth over the past year and continued operating efficiency.

Starting with the consolidated results. For the second quarter, we reported net income of $192 million or $4.78 per diluted common share. Our operating income was $210 million or $5.23 per diluted common share, which excludes $18 million of net realized and unrealized losses on investments.

Underwriting income for the quarter was $227 million, and we reported a combined ratio of 47.2%. Our gross premiums written grew 18% in the quarter to $977 million. These results drove an annualized ROE for the quarter of 18.6% and an annualized operating ROE of 20.3%. All in all, a very strong financial performance for the quarter.

Now before moving on to our segments, I wanted to highlight again for you this quarter the changes we made to our expense allocation as well as the continued improvements in our operational efficiency. We made some minor adjustments at the beginning of the year to our allocation methodology, which we discussed with you on the previous call, where we shifted certain public company expenses from operational expenses in the segments to corporate expenses to be more in line with how our peers treat such expenses.

This allocation's realignment aside, we continue to improve our operational efficiency as we increased net premiums earned by 12% in the current quarter compared to the same period last year, while keeping the sum of operational and corporate expenses relatively flat at about $46 million during the same period. But given the opportunities before us, you should expect operational expenses to increase modestly in the coming quarters as we invest in the business. However, operating expense ratio should remain relatively stable to declining over time due to growth in net earned premium.

Now moving on to our segments, and starting with the Property segment, where Property gross premiums written in the second quarter were up $53 million or 10.7% over the comparative quarter. Gross premiums written in our catastrophe class of business increased by 6.4%. But due to the favorable development we reported this quarter on the 2017 Catastrophe Events, we reversed out about $31.2 million of reinstatement premiums we had previously booked in the third quarter of 2017. Absent this adjustment, gross premiums written in our catastrophe class of business grew by 14%.

In our other Property class of business, we grew gross premiums written to $115 million, a 31% increase over the comparative quarter, albeit off a relatively smaller base. In total, our Property segment generated underwriting income of $214 million and a combined ratio of negative 4.7% in the second quarter. This compares to underwriting income of $107 million and a combined ratio of 45% in the comparative quarter last year.

You will note that our acquisition expense ratio was up in our Property segment by about 5 percentage points. Two factors primarily drove this increase: first, as a result of the 2017 Catastrophe Events, we exhausted a large portion of our profit commissions on certain ceded deals, which are netted against acquisition expense; the second factor was a reduction in net premiums earned due to the reversal of the earned reinstatement premiums I discussed previously. Absent these factors, our acquisition expense ratio would have been roughly flat over the comparative quarter.

Now let me spend a few minutes on the favorable development we recorded this quarter in our Property catastrophe class of business. The net positive impact of $77 million is the result of a deep-dive review we performed on the 2017 Catastrophe Events. While we review our event loss reserves each quarter, we will typically also perform a deep-dive around the anniversary date of large catastrophe events. Now that we are approaching the first anniversary of these events and with the benefit of the ceded loss information we received during the June 1 renewals, we can begin to assign increased credibility to a lower-than-expected level of reported losses we have been experiencing.

In a deep-dive, we take both a top-down and a bottoms-up approach. We begin with a ground-up review, individually reviewing our current losses by cede. In addition, we perform a top-down review to confirm that we are comfortable with our levels of ceding and event IBNR compared to various benchmarks. We believe our reserves appropriately reflect our best estimates based on all of the information available to us. And Kevin will talk more about the 2017 events in greater detail later.

Now moving on to our Casualty segment. Our gross premiums written were up close to 30% in the second quarter of 2018 over the comparative quarter. This increase was principally due to selective growth from business opportunities within the general casualty, other specialty and financial lines of business. Much of this growth is a result of our differentiated strategy to provide bespoke customer solutions.

This quarter, we non-renewed a few proportional deals and also wrote several large excess of loss treaties. The distinction between the business mix is important as it impacts the timing of when the premium is written and earned in our financial statements. Generally, the majority of the premium we write in our Casualty segment is proportional business, which is written and earned pro rata over the contract period as it is reported to us. In contrast to proportional deals, with excess-of-loss treaties, the full amount of the premium is immediately reflected as gross written premium and then earned pro rata over the contract period.

The relative shift to excess of loss contracts this quarter from proportional contracts resulted in a large increase in gross premiums written in the current quarter. While the premium related to this excess of loss contract will earn out over several quarters, it will not influence gross premium in the future quarters. Absent the shift to excess of loss contracts this quarter, Casualty gross premiums written grew just under 13%.

Overall, the Casualty segment generated underwriting income of $13 million and a combined ratio of 94% compared to underwriting income of $3 million and a combined ratio of 98.5% over the comparative quarter.

Our casualty book continues to run current accident year loss ratio of around the mid-60s, printing a 66% for the quarter versus a 69% over the comparative quarter. This is both in line with our expectations and consistent with recent performance.

We experienced modest favorable development in the Casualty segment from prior accident years, of which about half was related to 2017 Catastrophe Events. Positively impacting the Casualty segment's combined ratio was a 6.2 percentage point decrease in the underwriting expense ratio. This decrease is due, in part, to improved underwriting leverage as we continue to grow net earned premiums while keeping expenses relatively flat overall.

Now turning to investments. We reported total investment results of $54 million in the second quarter, resulting in an annualized total return of 2%. This is net of $18 million of net realized and unrealized losses, which is largely from the increase in interest rates this past quarter. Included in total investment results was $71 million of net investment income, largely stemming from our fixed maturity and short-term investment portfolios.

We benefited from higher average invested assets and the impact of interest rate increases during the recent periods. For the quarter, we grew our overall investment portfolio by almost $500 million. We also see that we sold virtually all of our $500 million tax-exempt municipal bond position. These assets were supporting our U.S. balance sheet. Given the reduction of the U.S. corporate rate last year, municipal bonds became relatively less attractive versus other fixed-income securities.

The yield to maturity on our fixed maturity and short-term investment portfolios was 3% at June 30, which is roughly flat from the previous quarter. The duration of our fixed maturity and short-term investment portfolios is slightly shorter than last year at 2.2 years. Our duration shortened due to the increasing amount of business rewrite on a collateralized basis through consolidated in the -- such as Upsilon.

There's also one more aspect to our investments I would like to provide further insight into. We have about $760 million of assets classified as non-investment-grade corporate. These investments are split about 50-50 between traditional, fixed-coupon, high-yield securities and senior secured bank loans. The senior secured bank loans are senior in the capital structure and floating rate. They are managed for us by external investment advisers with over 95% of the exposure to first lien. For these reasons, we are comfortable with our exposure and believe any incremental risk is commensurate with the increased returns.

Now I want to spend a few minutes talking about our third-party capital management. As we continue to grow this business, joint ventures such as Medici, DaVinci, Upsilon and Fibonacci are increasingly important in the context of our financial results. So I'd like to walk you through how to think about these entities' impact on our financials and, more specifically, the investment side of our balance sheet.

While we only hold minority stakes in Medici and DaVinci, we consolidate both of these entities. Consequently, 100% of their investments are represented in our investment portfolio, but their assets only impact our bottom line in proportion to our ownership percentage.

So for example, Medici holds $501 million of catastrophe bonds classified under other investments. But as we only own 17% of Medici, our bottom line is impacted by about $84 million of this exposure. On the other hand, DaVinci's portfolio is slightly more conservative than some of our other entities, and only 22% of its $1.9 billion of assets impact us. It has an investment composition of about 17% AA or higher and 9% below investment grade, with de minimis exposure to non-agency mortgage-backed securities. Its below investment-grade exposure is to senior secured bank loans, which I already discussed.

We also consolidate Upsilon, and I already pointed out its impact on a reported duration. We currently own about 14% -- 14.6% of Upsilon, and its assets tend to be cash or very short duration, high-quality treasuries or other short-duration AAA sovereign or supranational securities. We do not consolidate Top Layer Re or Langhorne, so their investments are not included in our portfolio. However, our investments in those joint ventures are recorded on our balance sheet as investments in other ventures.

Now moving on to our capital management activities. First, I want to point on -- point out that our approach to capital deployment has not changed. We have not repurchased any common shares thus far in 2018 because we have seen many attractive opportunities. The actions we took at this quarter are reflective of this view.

First, we raised over $350 million of third-party capital in Upsilon, Fibonacci and Medici. As of June 1, Upsilon has grown to over $1 billion in capital, and Fibonacci Re is now about $200 million. In both cases, we saw opportunities at the midyear renewal to match desirable risk with efficient third-party capital solutions.

Next, on June 18, we raised $250 million of capital at the holding company through the issuance of new Series F preference shares yielding 5.75%. The new Series F shares allow us to lock in low fixed rate in a rising interest rate environment and also provide us with additional capital and liquidity flexibility heading into the 2018 wind season. I should note that we still have outstanding $125 million of our Series C preference shares yielding 6.08% and $275 million of our Series E preference shares yielding 5 3/8%. Both the Series C and Series E preference shares are now callable.

In summary, we continue to believe we have the right strategy and resources to deliver superior returns to our shareholders over the long term. In executing our strategy this quarter, we were able to grow gross premiums written by 18% as we opportunistically deploy capital during the quarter, improve our operating investment leverage and effectively manage our capital to support both our continuing growth and the continuing market opportunities we are anticipating.

And with that, I'll turn the call back over to Kevin for more details on our segments.

Thanks, Bob. Let me start with Property, and then I'll move over to Casualty. So overall, we grew gross premiums written in our Property segment by 10.7% over the comparable quarter last year. Property cat gross premiums written grew 6.4%, with a substantial portion of this growth coming from a mix of better rate and new business.

Our strategy entering midyear renewals was to grow opportunistically with core clients and leverage our gross-to-net strategy. As a result, we were able to grow the Florida portfolio incrementally despite rates that were flat to up a few points. We also grew Upsilon, which is now over $1 billion in capital, and issued a new series of cat bonds through Fibonacci.

Rate change varied by program and by layer. Pricing in our overall Property cat portfolio at midyear, which would reflect both the U.S. and non-U. S. renewals, was up single digits, in some instances, we saw rate decreases. There's still an abundance of capacity in the market, and oversubscription was a common theme. We were able to grow where appropriate and also exercised underwriting discipline and [came] off business when our return hurdles were not met.

Let me touch briefly on the deep-dive of the 2017 Catastrophe Events Bob already discussed with you. We disclosed that we would perform this around midyear and believe that we have sufficiently credible data this quarter to make better informed decisions about the necessary quantum of our reserves.

When we initially set these reserves, we had to make a number of assumptions around the many variables that can potentially influence an event, and our estimates were dependent on the validity of these assumptions. Since making these estimates, however, more information has become available, and many of the variables are developing more favorably than originally assumed.

One example of this is the low emergence of risk losses from Hurricane Harvey. In Sandy, we saw a number of large risk losses due to extensive flooding. It was reasonable to assume that we would experience similar risk losses in Hurricane Harvey due to the heavy flooding. With the passage of time, this has not proven to be the case. The continuing lack of reported losses in our Property segment gave us sufficient comfort to recognize favorable development, just as we did last quarter.

There is always a significant amount of judgment in any reserving decision, which is often done in a relative vacuum of information. Making the reserving process even more difficult were the estimates around the impact of multiple events on our shared retro program. I believe we made the right call on our reserves last year. I am pleased to recognize favorable development events that could have been much worse for the industry than originally projected.

While we do not break out the individual losses from the 2017 cat events, our reserves for each hurricane were down on a net basis, and the wildfires were essentially flat. As I've said in the past, each of these events is unique, and I would like to provide you some additional insight into how each is developing.

Recently, you've seen our cedents' losses from Hurricane Irma move up substantially. Because of the [time] litigious environment and the omnipresence of public adjusters, claims settlement in Florida has always been difficult. Due to our decades of experience and deep expertise in the Florida marketplace, however, we were able to anticipate that material adverse development would be likely and we reserved accordingly. As you may recall, last quarter, I said to you, we were not surprised by the large number of reopened claims insurers are experiencing with relatively high levels of loss adjustment expense. For this reason, we are comfortable with our current position. In addition, many losses are now at a point for us where they will develop into the cat fund, which should mitigate the impact of further adverse development.

For Hurricane Harvey, a lot of our cedents incurred losses are at or below their retentions. Based on recent patterns, many of them may not breach these retentions on a paid basis. Because of this, we were able to revisit many of our assumptions around Harvey with our new lower loss estimates reflecting better information.

More so than the other 2 storms of 2017, Maria still has a fair amount of uncertainty associated with its development. Insurance companies are overwhelmed by the sheer volume of claims and struggled to access the personnel necessary as many contingency plans revolved around bringing in U.S.-based adjusters after a large event. These adjusters, however, were occupied with hurricanes Harvey and Irma. The filing of claims was further delayed due to the extensive infrastructure damage on the island. As a result, Maria claims are taking longer to close than those in Harvey or Irma. In addition, public adjusters have found their way to Puerto Rico and appear to be targeting retail and condos. These considerations weigh heavily on the commercial market, but have been anticipated by us, which is why we continue to be comfortable with our reserves for Maria.

With regards to the California wildfires, we're hearing reports of increasing demand surge due to lack of licensed contractors. As I've previously discussed, wildfire claims tend to settle much quicker than wind claims, which gives them a shorter tail. And we remain comfortable with our reserves against the information we have.

Moving from Property cat to our other Property portfolio. There was fairly substantial growth in our other Property portfolios this quarter as gross premiums written increased 31%. This growth is the culmination of many years of hard work, building leadership positions in other Property. The many seeds we have planted and diligently cultivated are now bearing the fruit of both new business and increased shares on existing programs. The consistent application of our 3 superiors and integrated system is allowing us to enjoy the same advantages in other Property, as we have for decades in the Property cat market.

The growth in other Property came across various lines of business and platforms. This includes property per risk, proportional business, delegated authorities and our direct and facultative business, both in the U.S. and Europe. I should note that we are keeping our eye on several large losses in the per-risk portfolio. We are comfortable with our reserve position on these losses, but they are young, and the full magnitude of the events may not yet be properly known.

Moving over to Casualty. We grew our Casualty segment gross written premium almost 30% year-on-year in the second quarter. I feel this is an outstanding result and want to give you some context to better understand this growth and why it occurred in the second quarter. As I've said in the past, the Casualty business is often influenced by a small number of large deals. In this quarter, we saw several unique opportunities where we're able to write material participations on large programs, which were open to only a few select markets. We achieved this outcome through the diligent application of our 3 superiors on the Casualty segment this quarter. We worked closely with our customers and were able to bring better understanding of risk to their underwriting.

Due to our gross-to-net strategy and our superior access to efficient capital, we were able to take large participations on the most attractive deals. Our results this quarter confirm our strategy and our leadership role in the casualty market. As an example of this leadership, we had a first-call opportunity to write an event-based excess casualty deal for the dislocated California wildfire liability market. This large transaction accounted for a material portion of the Casualty segment's growth in the quarter. As always, we increased on the best and decreased on the worst. And in several instances, we [were] non-renew deals that did not meet our return hurdles.

In general, the trends we saw at January 1 continued into the second quarter. We experienced improving terms and conditions on the more troubled accounts. And unlike January 1, however, stable accounts tended to renew as expiring as there was less hardening. Underlying rates continued to increase across casualty and liability lines, but loss trends are also up. We think these 2 trends probably cancel each other out and are keeping a close eye on further development in the underlying businesses.

As I've discussed in the past, our ceded strategy is critical to maximizing the efficiency of all our lines of business but I want to focus on our Property ceded program for this call. Most of the ceded protections for our Property segment are aligned with the midyear renewals. This year, we purchased significant additional limit, spending an additional $92 million to purchase retro cover. This may seem like a lot but our goal was to [mean] relatively flat exposure to Southeast hurricane, our peak risk relative to 2017 hurricane season.

Several factors were at play in our increased spend. As you remember, we wrote significantly more Property business at January 1 when rates were up. We also ceded a little less on a quarter share basis. So our increased purchase at midyear was partly us trueing up the portfolio for prior growth. Having the flexibility to know how, what and when to buy retro is an important differentiator of our gross-to-net strategy.

The majority of our retro capacity we purchased was provided by third-party capital funds and is fully collateralized. Overall, the ceded placement was a notable success within our gross-to-net strategy and is an important component in building one of our best net portfolios in recent years. By purchasing retro, and thanks to precise analytics and a deep sense of the market, I am pleased to report that we were able to construct a portfolio with risk characteristics for Atlantic hurricane, similar to the prior year.

In conclusion, we had a strong quarter. We successfully managed expenses, grew broadly across our business, kept our catastrophe exposure relatively flat and managed our capital effectively. For these and many other reasons, I'm confident that we continue to have the right strategy, regardless of market cycles, to deliver shareholder value going forward.

With that, I'll turn it over for questions.

(Operator Instructions) Your first question comes from the line of Kai Pan with Morgan Stanley.

So thank you so much for all the details about the reserve releases. I just want to follow up on that, to see, is that -- you do this study, you said, at anniversary day, which means we would not see sort of big reserve movements related to these hurricanes or 2017 cat events until second quarter of next year?

Kai, you came across a little broken out there. Can you actually just repeat the question?

Sure. And you said you typically do these, like, deep-dives, reserve study at the anniversary of those events. And that's the time you decided, let's say, you've added to these reserves. So -- which that imply would not -- we would not see significant movements on the reserve related to 2017 catastrophes like another year from now, basically, second quarter of 2019.

Yes, I think what we've announced earlier was that we would do a midyear review of these events. And part of the reason we use the term midyear is because we were aggressively collecting as much data as we possibly could, so we weren't sure whether we'd be able to do it in the second quarter or the third quarter. The combination of our work and then the reports we receive for the June and July renewals put us in a position to feel comfortable to do it in the second quarter, not at the anniversary, which would be the third quarter. So I wouldn't read too much into it other than we responded to the information that we had, and we felt comfortable that in the second quarter, we were in a strong position to be able to make the judgments that we made. I wouldn't read forward that we would do the same in the second quarter next year.

Okay, that's fair. But if I'm looking back at your 2004, 2005 hurricane losses, we have -- we didn't see sort of meaningful reserve releases from those events until probably 2007, 2008. I just wonder, is that correct or not? And what has changed here? Is that -- why your release is so soon?

I don't have the information to hand as to what happened with the 2004-2005 events. What I would say here is each event is different. We tried to point out is the California wildfires pay more quickly. But I think the important thing to focus on is that we identified, I think, the important variables that would control the loss or control our assessment of the loss, and we collected more information on that. So thinking of Irma, where there's an increasing industry dialogue around the growth in several elements of the loss. If you look back over the things we talked to you about since the loss, I think we have correctly identified that we expected to see an elevated loss adjustment expense. We expected to see more reopened claims. We talked about the impact of, not only Irma, but all of the events on the aggregate covers. And we all recognize that Florida is a litigious environment. So the fact that we kind of highlighted what we thought the important variables allowed us to focus in narrowly as to what was the important information for us to provide an update on our -- the quantum of our reserves against each of these, and that's what we did and we're able to achieve it in the second quarter. I think it would have been the same process we applied after each of the events, it's just that this event became transparent in the second quarter, not the third.

Okay, great. My next question is on the underlying combined ratio in the Property cat segment. So it looks like it's kind of elevated comparing the 49% versus like low-40s for the prior years. I just wondered, is there any sort of one-off item which is not including the cats?

No, Kai. The cat -- the [same with the] development on the prior year 2017 Catastrophe Events drove it. I did point out some anomalies in the acquisition expense ratio related to the events that actually drove it up. But absent that, they were fairly normal. Really nothing that was other than the cat events was a big driver.

You said, so it's just one-off for this quarter?

Then the large item for the 2017, yes.

Okay. And then last one, if I may. If you're stepping back, Kevin, if you look at first quarter operating ROE, about 13.5%. The second quarter, if you strip out all the reserve releases coming from 2017, that's still 13%. So I just wonder, is that 13% ROE sustainable in your view going forward?

Our strategy is the same. In the way we're looking at the business, we're thinking about targeting the lines of business that best serve our customers, and then thinking about the most efficient capital to match with it. So without giving guidance as to what the return expectations are, I think our strategy is working well in this market, and we hope that it continues to work well as we move into the third and fourth quarters.

And your next question is from the line of Amit Kumar with Buckingham Research.

A few questions. The first question, maybe going back to Kai's question on ROE. If I flip the question, do you think that the better way to look at RenaissanceRe is looking at the value creation over time? Do you think is that better? Or should we continue to focus on near-term ROE moves?

I'll pick it up. I mean, I think, Amit, our primary metric that we look at is value creation through our book -- tangible book value per share plus accumulated dividends, and that's the one we really look at and what we focus on. ROE and operating ROE is a relative measure off of that. Kevin, I don't know if you want to add anything else to that or...

No, that's exactly right. I think if we do our job and grow tangible book value plus accumulated dividends, I think that's our best path to creating long-term shareholder value.

The second question, and maybe I might hope for a bit more color on the reserve releases. If I understand correctly, the majority of the release came from Harvey, but Irma and Maria were sort of unchanged. Maybe just go a bit deeper and also talk about the LAE issue, which is probably running materially higher versus the past storms?

Okay. So let me start with the loss adjustment expense comment, which is within the Florida market, the way the market is structured, we've always anticipated higher loss adjustment expenses for Florida-related losses. And I think we've talked about that consistently on each of the calls since the Q3 event reserves were first put up. So I don't think that's something that we are particularly surprised at. But again, we've got 25 years of deep experience within Florida. With regard to the events, what I said in my comments is that, on a net basis, each of the events are down. So what I tried to do is to give a little color from a market perspective as to what's driving the events. But from our perspective, when we look across the board, each of the hurricanes are down on a net basis. And what I said is the wildfires are at about the same as what they were when we put them up in the fourth quarter.

The -- I guess, what I was trying to ask is, is Harvey like 80%, 75%? Like, is there some sort of proportion which would help us sort of do some sort of a comparative analysis with other companies?

I think we look at it as a portfolio and we manage it as a portfolio, adding to the complicating factor as to how the retro, which supports all of the events with a single program responds. So I don't think it would provide the insight that you're hoping for us to dive more deeply because the net changes to us are unique to us. And a lot of it depends on the assumptions you made with the original loss estimates that you used for reserving.

Fair enough. Last question, I'll re-queue. On the investment income side, you alluded to the higher invested assets with the new business, et cetera. The investment income this quarter is materially higher than the last, I guess, couple of quarters' run rate. Should we use this sort of as a guide going forward in terms of our modeling purposes?

I think there's 3 things driving the growth, and that is, one, is obviously we've seen interest rates increase over the last year, and that's driving probably the majority of the increase. You can see that in the fixed maturity where that's up several million dollars. Now there's some balance growth as well, but principally, interest rates are the biggest catalyst. Short term, which is up significantly, comes a lot of it on the Upsilon side that's helped part of the growth when you think about it in volumes, and then you can look at the -- the interest rates have been the biggest factor. As we look forward, we expect to see benefits from that as the rates go up. New money yield is 3%, probably a little bit higher than that with the rates going up, so that's how we focus on it.

And your next question is from the line of Yaron Kinar with Goldman Sachs.

I guess my first question, may be a little premature, but I'm going to take a stab at it anyway. Just given where pricing is today and with the benign start to the wind season this year, if the wind season proves to be benign this year, at the end of '18, (inaudible) of the structure in the portfolio and to have 1/1 renewals, are you going to be shifting into quota share from [extra swaps]? Are you going to be using more retro? Maybe any thoughts on that would be appreciated.

I think when we think about any renewal or any time we're going to engage with our customers is we try to think the best way which we can serve them, what's the right capital to bring to them and what's the right level of risk for us to accept from them. So I think when I look at what's going to happen at 1/1, I don't think there's a lot of information in the system right now, and we are going through kind of the peak Atlantic hurricane season. So I say that I don't have any meaningful comments about what we expect for January 1 at this point. But I can say that our strategy to go in thinking about serving our clients and bringing the right capital to their needs will be unchanged.

Okay, I appreciate that. And then my second question, just in terms of your partners with the perspective of a year after the storms, have the conversations changed? Are the issues that they're focused on changing at all? Or is it pretty much consistent from -- relative to where it was a year ago?

By partners, I'll divide my comments between our customers and then those from whom we share risk with. Our customers' conversations are, I think, increasingly about how we can provide more protections more broadly to them. And I think that's been benefiting us not just in the Property cat portfolio, but other Property across Casualty and Specialty as well. From our capital providers, I think they are appreciative of the transparency that we bring to them and understanding the risk from which they accept to us. So those conversations have been very robust. So I feel like we're in a great spot with where the market is and within our access both to desirable risk and efficient capital.

And your next question is from the line of Elyse Greenspan with Wells Fargo.

My first question, you gave some color on your gross-to-net strategy within your Property book this year. I guess, if we have a repeat of last year's events, not specifically the same events, but meaning more of a frequency of events than a severity event, do you think that the level of loss that went to the reinsurance market would be materially different than the share of the loss that they saw from the few -- the 3 hurricanes that we saw last year? I guess I'm trying to see, like, were there more aggregate covers purchased? Or was the way reinsurance was purchased this year different than how it had been purchased last year?

I think from a market perspective, demand has been pretty stable. And the -- those on the aggregate was about the same level of risk being transferred roughly. And the types of product that are being sold has not materially shifted from last year to this year. So I think from those 2 perspectives, we should see a reasonably consistent risk transfer from an insurance to reinsurance. I'd also say that retentions have been reasonably stable. So an aggregation of smallish-type events, even if it's a heavy aggregate year, I'd say it'd be comparable to what we saw last year. The one thing I will mention for us is, as I've commented in my prepared remarks, is if there are more wildfires, we saw an opportunity in the dislocated liability market or public utilities in California, so we could have more exposure to something like that, but I don't think that's an industry trend.

Okay, that's great. And then in terms of as we think about the takedown for the 2017 losses that you guys had this quarter, could you provide some color in terms of the IBNR that you guys have upped for the events, and how that will compare to past events, I guess, just as we think about the IBNR, I guess, is still there following the takedown that you saw this quarter?

Yes, I think it's different by event. As I mentioned, the California wildfires paid more quickly than something like an earthquake. And then the hurricanes are a little bit of a mixed bag with Maria being the slowest of the hurricanes. I think that's an indication that as things are slower, one should expect there to be more reserves associated with the claim. But I don't think, at this point, we're going to provide a breakdown of paid to case to IBNR. But we recognize that each of these have a unique development pattern, and I think we were thoughtful on thinking about those and how to set the original reserves and also how to set the current reserves.

Okay, great. And then in terms of the expense ratio this quarter, I know you guys pointed out a few factors in the hurricanes that impacted the number. If I kind of adjust for the reinstatements and the profit commission impact, again, in expense ratio, and this is for the entire company, it's about 29.6%, then it comes out a little bit lower than where you have been trending. Bob, I know you mentioned, I think, expenses might go up, but earned premium offsets that. So is that about the right expense ratio to assume going forward? Or was there something kind of one-off when we back out the impact of the 2017 loss adjustments?

No, I think we tried to highlight in my prepared comments the factors that were driving the cost that were outside of what you'd see in normal growth. They will come in and probably correspond to premium growth, like acquisition expense cost. We will invest in the business. We'll do it modestly as we see the need to expand the platform for growth. But again, at the same time, maintaining that discipline and try to point you that the ratio should remain at or around, if not below, where they're currently at right now. And it did point the levers that we built into the system across all of our expense base as well.

Okay, great. And then one last question. You guys mentioned that the surplus required for your book this year was about the same as last year. If you're carrying a bit more capital because you had the preferred that you issued this quarter, so I guess, if we were thinking about getting to a point of capital return depending upon opportunities, would your process be to wait, I guess, until later on in the year after wind season?

Listen, my comment around -- to point at the same level of surplus with the growth that we'd achieved was to give a little bit of color on the efficiency we're building into our risk portfolios. As far as our buyback positioning, nothing has changed. We continue to look to deploying to the business first. And our preferred method of capital return remains share buybacks, which have been accretive over time.

And your next question is from Ryan Tunis with Autonomous Research.

I just had a few here. I guess, my first one, following up on Elyse's question, you guys did a really good job giving us the kind of the bottoms-up view on why to get comfortable with a big reserve release. But just looking for something more from a top-down perspective, if you'll give us, I guess, paid to incurred, that would be helpful. But if not that, is there anything you can give us that we can see in publicly available data that you think kind of adequately says that, hey, like this is one reason why you should be comfortable with where we have net reserves set at this point?

Ryan, this is Bob. Thanks for the question. I think there's a couple of things. As Kevin pointed out, there is the complexity of the business model makes it difficult just the way we manage the risk. Yes, there's the individual storms, you layer that with the retrocessional across those -- across the portfolio, and then you have the aggregates. In the supplemental, we do lay out on one of the pages, on page, I think, it's 13, we do kind of show the changes in the reserves that breaks it down with a little bit more detail. You can see how it spikes up. That's probably where you'll get the most information on the detail for it. But again, we look at it on an aggregate basis and a net negative basis, just given the level of complexity.

Okay. And then just on actual versus expected, thinking about Irma, it sounds like you guys got it right in being pretty conservative around the LAE, a lot of the litigation stuff. So it's developed favorably relative to your estimates. I think the PCS estimate's at, like, $20 billion now as of June. I mean, where you guys have it versus where the PCS is coming in, do you guys -- is your assumption that there is still more creep and you're leasing off of that? Or do you think that kind of what you're seeing now paints a pretty good picture?

So I think when we first set up a reserve, we do look both top-down and bottom-up, top-down being market share-event driven as a way in which we can begin to zero in on the appropriate level of loss when there's -- when uncertainty is significantly higher. As the loss matures, we increasingly rely on more observed information on specific programs, so we're more from a bottoms-up approach, so we do less reconciliation as to where the industry loss is. Specific to Irma, I think the things that are being discussed as adverse development are things that we identified and thought would continue -- we thought would deteriorate. It would be difficult for us to comment whether the market believes there'll be more or less deterioration in it. We think we have a robust process to assess it, and we feel comfortable with the level of reserves we're carrying.

Okay. And then I just have a couple, thinking about, I guess, retro in Upsilon. If I have this right, so there's been a premium capital raise in Upsilon. And I mean, correct me if I'm wrong here, that's a lot of business that you've written because you raised capital in Upsilon than you would have otherwise written, is that right, on a gross basis?

So really, what Upsilon is, is it's a portfolio which, on a rated balance sheet, is heavily capital consumptive. So we like the business that is in Upsilon. We simply put it into that vehicle because it fits better on a nonrated collateralized structure than it does on the rated balance sheets that we write our -- the rest of our book on. So -- and as Bob had mentioned, we do own a portion of the capital that's deployed within Upsilon. So it's a growth portfolio we like. We just recognized that we can bring more efficient capital by moving it off the rated balance sheet and delivering that product to our customers.

Got it. I guess, what I was trying to maybe get at is, if you think about -- I mean, I guess, it's tricky if all you're doing is moving it from your balance sheet over to Upsilon. But just trying to think about it, if you have like a no loss year, your decision to raise all this capital in Upsilon, how can you quantify what the earnings pickup would look like to RenRe's shareholders from having done all that? I mean, if there's any way to really think about it like that?

Yes, I think that's a good question. The way I think about it is that business is best suited to that source of capital. So thinking of it, what the return could be if we retain that risk is a bit flawed because it doesn't belong on our rated balance sheet because that's not the best way for us to serve our customer. So I look at it as the opportunity over the long term to earn appropriate margins for our capital -- for our partner capital and for the fees which we charge to administrate the vehicle is through that, and it serves our client best. So I think over 10 years, regardless of whether there's losses or not, it's the best structure that we can and it's the best way in which we can participate and support our customers with that capital.

Your next question is from the line of Josh Shanker with Deutsche Bank.

You guys have been growing fantastically in non-cat lines. And I'm trying to understand the picture of the pie getting bigger or you getting better, and that might be a combination of both. You discussed the -- the California wildfire liability transaction, I think, is an example of the pie getting bigger. My question is, when you're winning business from business held by competitors, what types of competitors or what type of capital structures are losing business in the market to RenRe's offering?

So what we've talked about before is that a lot of our Casualty and Specialty clients are heavily overlapped with our Property cat clients. So I think we've done a very good job getting on the most desirable accounts that are difficult for others to access because we've had long relationships in other lines with them. So I wouldn't point to their being, we're targeting a specific competitor. What we're doing is targeting what we think are the best accounts and then working hard by bringing our expertise, not only in Property cat, but increasingly in other lines of business, to the [sessions] that they're making outside of Property cat. So it's more of about an [act], it's access for us more than competing.

Is that a conversation that begins with the broker or begins with the client?

It's increasingly -- we are a broker market. But increasingly, as we write more diverse lines for large counter-parties, many of those lines are written through different brokers. So we find it increasingly important for us to directly engage with the ultimate buyer so that we can talk broadly across the portfolio of coverages that we're providing. I think we do a good job balancing the important role that the broker plays and bringing our expertise directly to the client so we can have the most robust conversations.

So ultimately, this sounds a bit like a cross-sale to me. I'm wondering if you have an idea of a possible market for yourself and how well penetrated you are in those goals that you have and whether we can look at 2019 and say, well, gee, we're just at the beginning, we should have another great year of getting on those accounts that we've been targeting. I mean, do you have sort of a multiyear plan about where you want to be?

We have lots of different plans. Included in that is we have development plans for each of our core clients. Those include the types of deals in which they're ceding the lines in which we have on their most desirable programs now to target for growth. So it's something that -- it's pretty tactical in how we attack each of these opportunities, but it's something that is highly coordinated. I mean, we empower the underwriter to make this decision, but we make sure that every underwriter is fully informed about the quantum of the overall relationship.

And your next question is from Meyer Shields with KBW.

Just a couple of, I think, quick questions. One, Kevin, you mentioned that the standalone reinsurance business model is attractive to clients. Is that changing? In other words, is it becoming more attractive than it was a year or 2 ago?

We've always found it to be attractive, so I think there's fewer of us. So the scarcity value of having the ability to have a conversation with somebody who doesn't compete with them, I think, might be increasingly important, but it's something that we always thought was important.

Okay, no, that helps. Also, within Casualty and Specialty, so you talked about rates and loss trends basically offsetting each other from a, well, I'll call it a margin perspective. When you see loss trends accelerate, does that itself catalyze more demand for reinsurance solutions?

There's a lag between what you're observing and then what's happening in the current market. I had said that it's not always as cleanly connected as that.

Okay. Then the final question. When you look at, I guess -- so you correctly anticipated, for example, the LAE inflation or aggregate covers. On the LAE side, do current 2017 catastrophe reserves include any potential acceleration of 2018 as a difficult year?

Let me rephrase your question and see if this is what you're asking. Are we anticipating that if there's losses in 2018, loss adjustment expenses would be even higher because some of the adjusters are deployed on 2017 events?

Well, I meant the other way around. In other words, since these claims take a little bit longer to settle, if there are events this year, then loss adjustment expenses related to claims incurred last year would go up even more?

Yes, okay. So the scarcity of adjusters could impact either this year's or last year's loss adjustment expense. I think -- so from our perspective, I think what we have within our, particularly for Property cat, we have a demand surge, which is a component of how we think about pricing. So I think we do consider that there's inflation associated with events. I think what you're pointing to is, what about this specific element? And it's something that, I think, we would re-reset reserves for events that could occur in third or fourth quarter for 2018. We would certainly consider the availability of adjusters to make sure we understand the impact on loss adjustment expense.

And we have no further questions.

Thank you, everybody, for participating on today's call, and we look forward to speaking with you next quarter. Thank you.

This does conclude today's conference. You may now disconnect.