濱特爾 (RNR) 2011 Q1 法說會逐字稿

Good morning, my name is Nicole, and I will be your conference operator today. At this time, I would like to welcome everyone to the RenaissanceRe first-quarter 2011 financial results conference call. All lines have been placed on mute to prevent any background noise. After the speakers remarks there will be a question-and-answer session. (Operator Instructions) I would now like to turn the conference over to Mr Hill. Sir, you may begin your conference.

Good morning, and thank you for joining our first-quarter 2011 financial results conference call. Yesterday, after the market closed, we issued our quarterly release. If you didn't receive a copy, please call me, Peter Hill, at 212-521-4800 and we will be sure to provide you with one. There will be an audio replay of the call available from approximately noon Eastern Time today, through midnight on May 18. The replay can be accessed by dialing 800-642-1687, or 706-645-9291. The pass code you will need for both numbers is 59290688. 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 July 6, 2011.

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 me today to discuss the results for this quarter are Neill Currie, Chief Executive Officer, Jeff Kelly, Executive Vice President and Chief Financial Officer, and Kevin O'Donnell, Executive Vice President and Global Chief Underwriting Officer. I would like to turn the call over to Neill. Neill?

Thank you, Peter. Good morning everyone, and thank you for joining us today. During the first-quarter of 2011, RenaissanceRe experienced losses from the flooding in Australia and the earthquakes in New Zealand and Japan. These events caused our tangible book value per share, plus accumulated dividends, to decline by 7.1%. Quarters like this highlight the low frequency, high severity characteristics of our business. As a leading reinsurer of major catastrophic risks around the world, we expect to have losses of this magnitude in some quarters.

Our aspiration has always been to produce attractive returns on equity over the long-term, in exchange for accepting short-term volatility. We have achieved this since the beginning of our Company. As we have done throughout the history of our Company, we will build on our existing strong relationships with our clients by paying claims with industry leading speed and responding to their needs quickly. Of course, it is still relatively early in the information gathering process following the earthquakes. And the extraordinary complexity of these events will take time to assess fully.

But, as our track record has shown, our risk assessment capabilities enable us to write business confidently rather than back away from business in regions that have been impacted by catastrophe losses. In the wake of these events, as for past events, we have been assimilating information that our in house scientists are extracting from the field to fine tune our understanding of natural peril exposures.

These same risk assessment capabilities enable us to be uniquely situated to make sense of the repercussions of the new vendor model releases. The combination of our strength in this area and a world class underwriting team gives us a unique perspective. For example, there are some elements of the new model releases that have been incorporated in our view of risk for many years. By the same token, our granular research shows variation in potential outcomes for some regions that external models do not always capture.

Our clients and partners have come to rely on us over time for our ability to bring a combination of best science, underwriting skills and sound judgment to bear when underwriting their account. Each client feels like they are unique. They are unique and we treat them that way. Once again, this goes a long way to further cementing the long-term relationships we have forged.

In a testament to our underwriting discipline and the way we manage our capital from a quarter like this with a strong balance sheet and substantial liquid resources, we are well positioned, not only to meet our obligations to our clients and bring additional capacity to them when they need it, but also to take up new underwriting opportunities as they arise.

We do anticipate a gradual increase in demand going forward. It is our view that this will play out incrementally over time, perhaps over several renewal periods, as a result of a number of factors including the extension of Japanese renewals at April 1, the occurrence of major catastrophic events after the January 1 renewals, and the time that it will take for the model change implications to filter through the insurance and reinsurance markets.

At this time, we feel the combination of our uncapital position, long standing proportional retro session partners and our joint venture partnerships with Top Layer Re and DaVinci Re will enable us to match capital opportunities when they come. If demand is greater than we anticipate in peak areas, we may set up short-term joint ventures to meet the demand as we have in the past with side cars like Tim Re and Starbound Re. However, we will do that only if we feel confident the capital can be deployed to achieve attractive risk adjusted returns.

So, we have an interesting time ahead of us as clients absorb to effects of catastrophic losses and the implications of the new model changes. We stand ready to support our clients by providing thoughtful solutions and substantial capacity. With that I would like to turn the call over to Kevin.

Thanks, Neill, and good morning. I would like the to pick up on Neill's main themes in my comments this morning and give more color on our Japan losses, on our view of risk and the impact of the new model releases.

In general, quake losses are difficult to estimate and the Japan quake is no exception. On the personal property side of the loss, the unique nature of the coverages offered and the role of the government reinsurance scheme adds complexity. On the commercial property side, there is a big unknown with regard to non-visible structural damage, as well as with business interruption and contingent business interruption.

As with any loss, we try to vector in on our estimate using as many tools as possible. Typically, one of the most useful tools is direct feedback from our customers. But, at this early stage of the loss, we have few loss notifications to base our estimate on. We have created a groundup loss estimate by looking at satellite photos and supplementing our information with input at the cool level from our weather predict seismic team.

Additionally, our loss is made up of half retro sessional contracts and half primary reinsurance contracts, including both international and local Japanese covers. We are not a broad player in the market, so we are focused on a few large contracts. This makes our own loss estimates subject to change if these contracts develop.

It's important to note that our franchise is built on risk taking, underpinned by rational analysis of exposure. We are not backing away from Japanese risk post event and are actively working with our clients to assess their exposure and provide coverage where it ads value to both client and reinsurer. Throughout our history, situations like this have pushed a flight to quality, which enhances our ability to source and construct a portfolio that we find attractive.

Next, I would like to spend a few minutes talking about our view of risk. There has been a lot of talk in the market about the recent changes to the AIR and RMS North Atlantic hurricane models. At RenRe, we are in a continuous process of trying to bring the best data and science to our representation of natural hazard risk, both at the industry level and then a client by client basis. As we have done continuously throughout our history, we have updated our view of US risk in REMS, our proprietary risk modeling system, and have just completed our most recent update. These changes draw from what we learned during the 2008 events, our own research, and from the analysis of the enhancements of the new hurricane releases from both AIR and RMS.

Simultaneously, we are in the process of updating our view of international risk in REMS, incorporating what we learned from the recent earthquakes in Chile, New Zealand and Japan. The changes I am discussing have been incorporated at a high level, and over the remainder of the year, we will continue to refine our models at a more granular level. This is the same process we followed in implementing the much simpler changes that we incorporated in late 2005. In short, we are prepared for the June renewals and will look at business with an updated view of risk with which we are comfortable.

The market conversation about the models revolves around one or two changes that have been the largest focus. We do not think of the models as a series of independent variables. It is a comprehensive system with one assumption very likely influencing another. We feel that we need to understand all the variables in order to assess the credibility of the models and to better evaluate our own view of risk. This deep understanding differentiates us in that we can identify the key elements affecting loss curves, both for our clients and for ourselves.

Some high level comments I would make about the changes to our model are, our view of hurricane risk in the US has increased. The increase in our REMS loss curves are different regionally from the vendors and the magnitude of our changes vary substantially from what is observed in the new models. We believe that our ability to understand the new vendor models and how they impact our clients' portfolios, as well as the speed with which we have been able to incorporate adjustments into our REMS model, differentiates us in the market.

Our customers understand that our pricing is exposure based and that we will continue to seek adequate returns for our capital. We have rerun all of our accounts through our revised model and are discussing our findings with our customers. There have been big differences from one customer to another in the amount of change to their loss curves. As in the past, we will work with our customers to help them understand how and why our view has changed and how that affects our pricing of their risk.

I would now like to discuss the current market. I think it's helpful to divide my comments between international and US markets. On the international side, we have seen some pretty large increases in the areas with losses. Overall, the markets saw sizable increases in Chile, we expect to see large increases in New Zealand. We have provided temporary extensions for Japanese covers and anticipate rate increases once these covers renew. It is our view that many regions beyond those with losses need rate increases to reach our return hurdles, but, so far, the rate increases have only been in the areas that have experienced loss.

Unlike the international markets where we have physical events to learn from and to calibrate our models against, in the US, most of our clients and brokers are focused on the changing view of risk as represented by the major vendor models. Because we have our own view of risk, our job is to look for enhancements in the new models. Without this independent view we would be similar to others in the market who have to make a decision to either accept or reject the new model. In some ways, this is a much harder decision because their own view of risk changes dramatically from one day to the next, or from one version to another. With all model changes, there is a lot of focus on what is going to happen in the short-term. I'm not sure that what we see over the next few months will be the whole story on rate changes and that it may be a much longer period of time before rates catch up to the new view of risk.

Markets tend to move either by reinsurers leading the increase, or more organically with primary insurers raising rates. In 2005, the reinsurers led the market by more quickly adapting to the new understanding of risk. While it's our understanding that primary insurance property rates are starting to go up, we don't expect to see a Katrina-like dislocation, but rather a more gradual adjustment with rates moving over a period of time.

In closing, I will touch briefly on our other businesses. Our specialty business is performing well and we continue to see low loss emergence. There are already a few bright spots in this market and we are hopeful that we are moving to a better market generally. Our Lloyd's business is experiencing good growth, which is our primary focus at this point in its development. We did have cat losses in the quarter which will hurt our results for the year, but outside of our cat, our results are good and we remain optimistic and excited about this platform. Thanks and I will turn the call over to Jeff.

Thanks, Kevin, and good morning everyone. On today's call, I would like to go over the first-quarter results and update you on our 2011 top line forecast. The first-quarter was a challenging one for RenaissanceRe, as it was for the rest of the industry. The high frequency of large losses, such as the Australian floods, February New Zealand earthquake and the Japanese earthquake during the quarter resulted in us reporting a meaningful operating loss.

The total net negative impact of the three events on the financial results of RenaissanceRe was $427 million. The net negative impact is the net loss amount after accounting for reinstatement premiums assumed and ceded, loss profit commissions, equity and losses of Top Layer Re and non-controlling interest in joint ventures. It also includes recoveries on ceded reinsurance contracts accounted for at fair value, that impacts the other income line item. We have provided a detailed table in the press release relating to the calculation of net negative impact. As Kevin mentioned in his remarks, there is still considerable uncertainty around the loss estimates given the scale, complexity and the recent nature of these events.

Favorable reserve development in our specialty and cat units was a slight offset to the catastrophe losses during the quarter. Investment income remained under pressure as a result of the continued low interest rate environment.

We reported a net loss of $248 million, or $4.69 per diluted share, and an operating loss of $243 million, or $4.59 per diluted share for the first-quarter. Net realized and unrealized losses totaled $5 million. Our annualized operating return on equity was negative 30.7% for the first-quarter, and our tangible book value per share, including change in accumulated dividends, declined 7.1%.

Let me shift to walking you through the segment operating results. In the fourth-quarter, we changed our segment reporting to reflect the sale of our US based insurance business, which we began reporting as discontinued operations. During the first-quarter, we completed that sale transaction. I will go over the results of our reinsurance segment, which includes cat and specialty, followed by our Lloyd's segment and retained insurance.

In our reinsurance segment, managed cat gross premiums written in the first-quarter totaled $529 million, compared with $456 million in the year ago period. The year-over-year increase was driven primarily by higher reinstatement premiums related to the large loss activity during the quarter. Adjusting for $113 million of reinstatement premiums in the first-quarter of 2011, and $27 million of reinstatement premiums in the year ago period, managed cat premiums declined 3%. The top line decline during the quarter was primarily a result of softening market conditions at the January 1 renewals, partially offset by increased participation on certain programs and improved terms on a few contracts that renewed following the recent catastrophic events. As a reminder, managed cat includes the business written on RenaissanceRe Limited's balance sheet, as well as cat premium written by DaVinci, Top Layer Re and our Lloyd's unit.

The first-quarter combined ratio for the cat unit came in at 248.8%. This included net underwriting impacts of $34 million for the Australian floods, $179 million for the February New Zealand earthquake and $294 million for the Japanese earthquake. The cat combined ratio benefited from $20 million of favorable reserve development during the quarter.

Specialty gross premiums written totaled $75 million in the first-quarter, which was approximately flat with the prior year period. The combined ratio for the first-quarter came in at 65.7%. The specialty combined ratio included $39 million of net impact from the large catastrophe losses during the quarter. This was largely offset by $52 million of favorable reserve development, approximately half of which related to actuarial assumption changes.

In our Lloyd segment, we generated $37 million of premiums in the first-quarter, compared with $14 million in the year ago period. Specialty premiums accounted for most of this amount, although premium volume was higher across each of the business lines. The Lloyd's unit came in at a combined ratio of 267.7%. Claims related to the February New Zealand earthquake and the Japanese earthquake totaled $21 million. The expense ratio for the segment was 73%, although we expect it to decline meaningfully over time from this level, as we continue to expect business volume written on this platform to expand.

Moving on to our insurance segment, as we have mentioned on our last conference call, the Company is currently not actively writing business in the insurance segment, but may choose to do so from time to time in the future. Gross premiums written totaled $300,000 during the quarter. The segment incurred a $3 million underwriting loss. On a go forward basis, we would not expect much contribution from this segment.

Moving away from our underwriting results, the other income totaled $50 million in the first-quarter. The major driver of other income was $44 million related to recoveries on ceded reinsurance contracts that are accounted for at fair value. These recoveries are not included in the calculation of the combined ratio, but are included in the calculation of net negative impact from the catastrophe losses. Other contributors during the quarter were $3 million of pre tax income from real and a $3 million mark to market -- not a mark to market gain, a gain on the sale of our platinum warrants in January of this year, as we discussed on our February conference call. Equity and losses of other ventures included a loss of $23 million for Top Layer Re as a result of exposure to the February New Zealand earthquake.

Turning to investments, we reported net investment income of $60 million, with our other investments portfolio contributing $34 million of this amount. Recurring investment income from fixed maturity investments remained under pressure due to low yields on our bond portfolio. Private equity had another strong quarter, the total return on the overall portfolio was 0.9% for the quarter. Net realized and unrealized losses totaled $5 million in the quarter. Our investment portfolio remains conservatively positioned with a high degree of liquidity and modest credit exposure.

During the first-quarter, we reduced our allocation to US treasuries and increased our allocation to short-term investments. The duration of the investment portfolio decreased to 2.5 years. The yield to maturity on the fixed income and short-term investments remain flat at 2.1%. The credit quality of our fixed income portfolio remains high with 58% of our fixed maturity securities rated AAA.

Our capital position remains strong despite the high loss activity and we have ample capital and liquidity at the holding company to meet market opportunities. In the event of additional market dislocation, we have options to raise capital through numerous sources and a number of entities such as joint ventures, and potential side cars through which we could deploy it.

During the first-quarter, we repurchased 2.7 million shares for a total of $175 million. The majority of share buy backs were conducted in January and February. Following the Japanese earthquake our 10b5-1 plan automatically terminated in accordance with our terms due to an automatic trigger hit by the estimated size of the industry loss. While we have ample capital, we do not anticipate further share or purchases in the near term as we evaluate opportunities to deploy capital on our underwriting businesses.

Finally, let me give you an update to our top line forecast for 2011. We are raising our forecast for managed cat premiums to an expectation that we will see modest growth compared to our prior guidance of down approximately 10%. This is adjusting for reinstatement premiums in both years. As a means of providing context for that change in outlook, approximately half of our current book has already renewed for 2011. While we would like to be more specific, it is still very early in the June renewal season and we don't have a clear enough view to land on a specific number at this time. We are maintaining our top line forecast for specialty of up 10%, and our guidance for Lloyd's up over 50%, albeit off of a small base. With that I will turn the call over to Neill.

Thank you, Jeff. Nicole, let's open the call up for questions, please.

(Operator Instructions) Your first question comes from the line of Jay Gelb with Barclays Capital.

Thank you and good morning. In terms of the impact in Asia Pacific over the past quarter, currently Japan earthquake is a global peak catastrophe zone and even RenaissanceRe has had larger declines in book value from other catastrophe events in the past. So, my question is, given the substantial amount of catastrophe activity in Asia Pacific, to what extent do you think this is a game changer for the US market including the mid year renewals? Thanks.

Jay, it's Neill. I'll start off and then turn it over the Kevin. I think it's all part of the holistic number of things that come in to play. I think we can never forget human behavior is part of the inputs here. I don't think other than learning from the Japanese event and the learnings from that it could be utilized in underwriting US earthquake risk. I don't think it's going to have that much effect on the US. It takes a couple of bucks out of the system, but I think it gives us more insight into earthquake exposure in that particular territory. So, I don't think that has that much effect on US pricing. Other things have more effect there.

Neill, when you are talking about that, you are talking about wind as well in terms of impact on pricing.

Yes. So, what I am saying is there are certain learnings we will take away from the earthquake standpoint. A far amount of learnings there in Japan teaches us more about the risk in Japan, not necessarily the risk in California. We think in the US most of the effects on increased demand will be driven by model change and new insights there.

On the model change, can you -- I know it is early days, but can you give us a sense in terms of what influenced the change in the cat models would have on RenRe's model PML's going forward?

Sure. I'll turn that one over to Kevin.

Sure. We have done a ton of work trying to evaluate the new models with teams of scientists, engineers working collaboratively with our underwriters. With that, our model reflects our best view of risk at any given time. With the new models that have been released by AIR and RMS, we have incorporated already some of the features that we think are best. Some of the features that are enhancements to the new versions are things that we already had in our model and then some of the new features of things we don't want to add to our model.

We have taken a view and are reflecting that against our current portfolio, it has increased our hurricane risk. I think we have a lot of different levers to help us manage that. We can think about the spread over book and we can think about changing the ceded nature of managing the portfolio. We can certainly change where we are attaching on an account by account basis to reflect how it is individually being reflected on the clients' risk exposure. We can also change price. We have got a lot of different levers to pull here. It's not going to be one thing moving it. It's how we want to shape the portfolio based on our updated view of risk.

It's an input, not necessarily a driver?

It's one of the key -- it's any time there is a model release, based on a lot of good modeling firms spent a lot of money trying to bring the best science to their model. Our job is to extract out the best features of that and continue to incorporate those into our model.

Right. Then just a quick one for Jeff, on the share buy backs, should we make your comments saying that no buy backs in the near term, does Ren expect to wait to get through hurricane season before reevaluating buy backs of more like wait until the fourth quarter until we see anything again on share buy back activity?

Jay, I think without a detailed forecast on that, typically we do -- we are not larger purchasers of our shares during the third quarter. I do think it's, as I said, it's probably not likely that we will incorporate very much share repurchase in the coming months. I think it is fair to say that at least our view at the moment is that we would not have large scale repurchases probably until later in the year.

Jay, we want to see what opportunities come our way as well.

Makes sense, thanks again.

Your next question is from the line of (Inaudible) Walsh with Citi.

Buddy. First question, apologies if I missed it, but why were the -- if you could just talk through the reinstatement premiums, why they were so high again? And then I've got a follow up, thanks.

It's a feature of the compositions of book that was hit. One of the things I commented on is about half the loss is coming from our retro book and the retro rates on line are substantially higher than what you have in the primary.

Then the follow up, just on our RMS 11, maybe if you can talk about when you look at the model, the exposure changes you're seeing for your client as you run it through the model? And specifically, I'm asking average in specifically in the mid Atlantic, Texas and Florida? Thanks.

I think, going back to what I said, we are in a good position. We can talk about the differences between version 10 and version 11, or 11 and 12 with our clients and help them understand the changes and how they are seeing it. It's independent of how we look at it. Our view of risk has always been separate from the vendor models.

The fact that we are seeing big increases, we can explain why we are seeing those increases in detail, our model reflects very different losses regionally than what you can see in the vendor models. So, we don't need to engage in saying we think this is better or that is better. It is much more factually about here is what is changed and why, rather than here is what we believe and why. It's a better dynamic for us to have the conversation with our clients.

Keith, one thing we probably don't bring out quite as much as we could is the way we are staffed to help analyze these models. We've got a team of very experienced scientists, of modelers, and exposure analysts. So, it's not just Kevin and I sitting in the corner having a beer and saying what do you think. We've got folks that really poke and prod and we have our own view.

Your next --

Operator, are there questions. Or, Keith, does that answer your question?

Yes, your next question is from the line of Greg Locraft with Morgan Stanley.

Thanks. Good morning. I wanted to just follow up on the guidance on the managed premiums. I'm trying to think about how the world has changed in the last -- through the quarter. Just to sort to reiterate, what you said. Half the book renewed at a down 10. Now the guidance is for the total book to be up modestly. Does that mean that, if half the book renewed at down 10, you're expecting the rest of the book to renew at up 10 or greater for the rest of the year? That's just the math, right?

I think that's roughly the math. I don't think, Greg, we said that the -- all of the renewals took place at down 10. I think we probably saw some improvement in pricing on some of those deals, and as I said, we did a couple of other deals. So, that resulted in I think managed premiums being down only 3% year to date. Net of reinstatement premiums. It's still to get through the rest of the year, just to get to flat, to provide that sort of point estimate would require about 7% increase in premiums.

Greg, it's Neill. Couple of points that you need to remember. We are not making a statement saying we expect rate decreases or rate increases of a certain amount. That down 10, what our premium volume was, that could have been solely as a result of not renewing one account. There is lots of moving parts here.

I would also say that we try to help you guys out. You have hard jobs trying to predict the future. It's very difficult to predict the future. Please don't be too scientific when we give you these estimates. They could be significantly higher or they could be lower.

I appreciate that. I'm trying to assess the rapidly changing supply-demand dynamic in the marketplace. So, you've mentioned the gradual increase in demand. Sounds like there is roughly 1,000 basis points swing then in terms of what the market was seeing six weeks ago and what it's seeing today. Is that a fair statement?

I think, the guidance is directionally we think we are going from a shrinking book to a growing book. There is a bunch of things that are going to change, even looking back as to what happened in the Q1. We got off a fair amount of business, wrote some new deals that came to the market. A lot of the growth to the extent there is more demand put into the market. I think we will be able to play very effectively into that demand.

There is a degree of uncertainty with how the new models are going to be adopted. So, we are taking an estimate as to what the future looks like based on some adaptation of the models. We are also thinking that the models could be adopted more slowly and that whatever demand appreciates -- increases are associated with the models, could come out over a longer period as the models are more slowly adopted.

That kind of brings me to my next question, which is the longer period, is this -- it sounds like from your commentary you are thinking this is through all of '12 as well? This continues to play through the market, whatever we are seeing now is a multi year phenomenon or a one year phenomenon in your opinion?

We don't know. If we had to guess, we think it will be a multi year phenomenon. We don't see it being something like after Katrina where there was one big push and that was that. We think there will be different parts that will cycle in at different time periods over the next few years.

Just adding to that, we have seen a lot of things in the new model that we like. We think over time there could be greater adoption of the new model. With that, there could be more primary rate increase and associated reinsurance rate increase. It's not like there is going to be a sudden shift where after 2005 there is a change in frequency, that kind of was adopted very quickly. This could be adopted over time and associated with that could be changes in behavior as to people how people buy or sell reinsurance.

Okay. Okay. Then the last is just on the Japan loss, it looks like most of it is IBNR at this point. Can you talk a bit about how you put it up as IBNR and maybe compare it to other events where perhaps not as much IBNR was put up at first, and how it plays through the next several quarters in to your income statement?

I think, frankly when I think about it, I try to get the estimate of loss by account as accurate as possible. When we have information from a client we move it more typically is when we move it from IBNR to ACR, we'll get their estimate and bring in ACR and reserve something for IBNR. It's very early on. I wouldn't pay too much attention, frankly, as to whether it's in ACR or IBNR. I think as we get more concrete information from our clients, it will migrate in to an ACR.

This was built ground up client by client with some additional IBNR on top.

If you look -- can you maybe go back to Katrina or another big loss. Did you see the same amount of IBNR day one as you are seeing in this? How is this different or the same?

Off the top of my head I don't remember exactly the booking process on those. I can tell you the underwriting process to estimate the loss is exactly the same, where we vector in on it from multiple different levels. If you are trying to drill out, do we have more or less confidence because it's in ACR or IBNR, I wouldn't think of it that way. I would think of it from underwriting perspective, it's our best estimate and it's based on the tried and true process we followed on other events.

Okay, that's great. I will get back in the queue, thanks.

Your next question comes from the line of Vinay Misquith with Credit Suisse.

On the Japanese quake loss, curious as to whether you are assuming a total loss on the large mutual contracts? And specifically was a little surprise to see that on Top Layer Re there was no loss to RenRe from that transaction.

Yes. I know you are not explicitly asking about a specific account, but it's our policy not to comment on specifics with regard to our customers. Top Layer Re writes more than one account in Japan. It's our job to assess whether we think they have a loss or not.

We took each of the Top Layer Re accounts, developed a ground up loss assessment, vectored in on it with industry loss information, the weather predicts seismic evaluations and everything else to make the determination we don't think they have a loss. It's not looking simply at one account. It's looking at multiple accounts that Top Layer Re is on.

Fair enough. The second question is on retro. You seem to have bought more retro sessional reinsurance in Jan 1, maybe about $40 million more this year first quarter versus last year. Can you help us understand your retro purchases versus sales for the second quarter on for this year and that's going to effect your PML for the rest of the year?

We uniquely with how we think about retro, our own retro purchases is we think of it simply as a portfolio tool. It's generally not used to control PMLs, it's used to shape the portfolio and to increase returns. We saw a few good opportunities in the first quarter. I think there is still a few more opportunities that we are looking at as far as buying additional retro session. Being that we don't have a program that we go out with, it really is much more of a trading mentality and a portfolio shaping mentality. It's not something I can concretely focus as we are going to do x or y, we are going to continue to optimize the portfolio, writing more inwards and potentially buying more outwards.

And the last question is impact of new model on your capital. I understand that you already answered the question, but just a bit hazy, has the new model reduced your payment of producer capacity by 10% or 15%? Just want to get a sense for what impact your model has on your capacity. Thanks.

Sure. It has some effect, not as much effect as one might think. We look at the entire portfolio. Part of it is the way we already had a fair amount of this information embedded in our own model and the way we looked at risk on and account by account basis. Our utilization of capital to support our business is up a little bit.

Thank you very much.

Your next question comes from the line of Cliff Gallant with KBW.

Good morning. I just wanted to ask about the Japan market itself. We expect after an event of this magnitude that perhaps structurally that market might change and so there might be more opportunity in the future. If so, is Japan a geography or is it data friendly, such that it will be good for -- a good match for your models and could you see yourself being more involved in that market?

Yes. To all of those. There is good data, it's a market we like, it's a market we know well and it's a market we have good relationships. The question really is price is what is limited our entrance in to the market. That I think we are going to see some adjustment. The other thing is there are large proportional placements out of Japan, and we are not a large proportional writer. To the extent there is contraction on the proportional risk or reinsurance that's available to them, there could be more excess of loss purchased, which will play very nicely into our hand.

Okay. Thank you.

Sure.

Your next question comes from the line of Doug Mewhirter with RBC Capital Markets.

Good morning. Most of my questions have been answered, I just have one question about the RMS models versus your models. Could you, I guess help me understand the kind of conversations you have with your clients that are affected by the models say in hurricanes like the Gulf of Florida?

As I understand, a lot of the rate agencies put a lot of stock in the vendor models to varying degrees, which effects your customers demand. So, maybe they get this updated model and maybe AM Best says, well, or they think AM Best is going to tell them you need to buy $50 million worth of reinsurance this year versus $40 million last year, they approach you and you say, based on our models you need to buy $60 million or maybe its less, maybe you only need to buy $30 million. Who wins that battle? Does the customer fall back on the vendor model to I guess satisfy the rating agency or do they steer it towards your recommendations because they may think you have a better view on their risk?

It's a great question. I think the conversations starting with the customer, really is, it's two separate conversations almost. One is if they are an AIR user or an RMS, they may ask us to help them understand why their risk looks different between the old and new versions. We will happily engage in that conversation based on how they are thinking about their risk.

The second conversation is really how we are thinking about their risk based on our model. That may inform how they are thinking between the two versions that they are looking at. It's a very different conversation than the comparison between one version to the other within the vendor models.

The second element of it is, what happens to them after they leave us and talk to the rating agencies. That's something we are really not -- we don't participate in that dialogue with our customers. Whatever the rating agencies determine, they are going to need to make a decision as to how they want information presented. Historically, they tended to move to the new model. Looking at past behavior. We don't have a clear definition from the rating agencies as to what they are going to do at this time.

Okay. I guess to follow up that, maybe put a finer point on it, you would think that the vendor models -- the effect, the overall effect on the market of the vendor models may influence the demand for you're reinsurance disproportionately towards than how you've changed your own models? If I worded that correctly.

Sounds like there is two questions there. One is has our model, our new to old model for lack of a better way to put it, changed more or less than the vendor model changes. Looking at it in an industry basis, it's very different than looking at on a customer basis. In most instances, our assessment of the customer risk has changed less than version one to version two of the models. I think we are better positioned if the market, if our competitors or customers are adopting from one version to the other.

The other question really is, kind of how did the rating agencies look at it. That's one I'm less qualified to talk to about what the rating agencies are willing to do next. I think they -- there is a good chance that they'll want to understand the different versions and why someone chose, if they did chose not to adopt. The final point I would make is not every risk is higher in the new versions of the models. We've seen certain customers with a lower risk curve under the new version than the old version. It's a little bit of a mix.

Thanks. That was a complicated question, I appreciate the answer. That's all my questions.

Your next question comes from the line of Brian Meredith with UBS.

Good morning. I was hoping you would comment a little bit more about opportunities in the retro sessional marketplace. What do things look like right there for you? Do you have capacity to write more business there? And then on the other side, as far as your ability to buy retro out there, is it cost effective right now as you are looking at some deals here going forward? Is that an impediment potentially of doing any business going forward?

We are one of the large recognized leaders and large writers of retro, so we are well positioned to seeing what is going on in the market. There hasn't been that much of an increase in demand, there has been some. For the number of covers that we think are impaired on a first event, we haven't seen as many back ups as we have impaired from the first event. There seems to be more of retention with some of our customers, at least for subsequent event covers.

The outwards reinsurance, I think, is -- that's always a bit of a moving target. We don't look at it as a -- at this point we buy at this point we don't. We tend to look at it more fluidly against the changing nature of our portfolio. As you our portfolio becomes higher returning, we may have tolerance to buy things we didn't have tolerance to buy before. I think we still have opportunities to buy more, but again, the market seems to be a little bit quieter than it typically would be after large event.

Lastly, can you chat a little bit about opportunities at the Lloyd's marketplace, as a result of what has been going on? And there is a lot of chatter out there about price increases and are you kind of in a good position right now to take advantage of any opportunities?

We are in a better position. I wouldn't classify as in a good position. We are in a good positions in the business that we are in. For cat we've got some specialty underwriters there now, but again we are still trying to grow the platform, we are trying to hire more underwriters. It will take us a little bit of time to leg in to any changes into the market. Because we are not fully staffed at this point.

Thank you.

Your next question comes from the line of Ron Bobman with Capital Returns.

Thanks a lot. Appreciate the time. Could you provide some sort of metric with respect to sort of potential Japan net loss addition? I know you reversed to specifying clients of course, but could you give us some sort of metric as far as how far away from limits you are on a net basis or some sort of element there?

Then I had a question about the Japan data that you are provided. I think you responded to somebody's question about the quality of the data and you responded with price is a big driver going forward. Do you get location specific data, I presume like you do in the US and in Florida as far as homes insured and the addresses of the underlying books or do you not have that for Japan? Thanks a lot and best of luck.

The first question is about kind of development on the loss. I think as -- it's irrational to blow everything that potentially has exposure often times from an event because things can be so remote that it doesn't make sense. We take the same assumptions on our retro -- when we assess the net loss, we are looking at it on the same way with how our ceded responds and how our inwards loss is developed. I think there is always a chance for upward or downward movement on a loss, but I don't feel that we are particularly exposed with the way we assessed this loss. I think we've done a good job. Yes, there is move, there is room for it to develop, there is room for it to decrease, but right now we think we have it at the best estimate that we possibly can.

The other thing on the Japan data, I think we get different levels of data, actually, all across the world depending on the client. One of the things I mentioned is we write about half our losses coming from our retro book, which is not the same degree of granular data that you would normally receive on a primary basis, but we look at it and we have done this for a long time, we've built models that allow us to break down into the markets, I think, at a reasonable level. Looking at all of that, I think we do get pretty good data, for Japan, certainly comparing it to other regions around the world and outside the US.

One of the complications I think in Japan is, there is a lot of retentions that come in before the data is released to reinsurers between the government reinsurance teams, between their proportionals and ultimately what we are presented is their net retained exposure. I think that's a degree of complication in there, but it's one in which we have gotten comfortable with over the years working through.

Thanks.

Your next question comes from the line of Ian Gutterman with Adage Capital.

Just to follow up a little bit on the development risk and just international retro. From the supplement, from the balance sheet, it looks like you have a little bit of over $200 million of recoverables from this event. Can you give any break down on how much of that was from New Zealand versus Japan?

Definitely more from Japan. We have, as Neill mentioned, there is two -- there is three types of cover that we buy. We buy, what Neill mentioned is our proportional cover, that would respond, pretty uniformly between the two events. We have UNL traditional retro session cover protecting our loss. That is more -- we have more of that available to the Japanese loss than we do to the New Zealand loss. The third thing we buy is more of index based products, including ILW's and we have more ILW recovery in Japan than we do in New Zealand.

I guess that's helpful. I'm trying to think through, given we have had two major international events, I assume most of that retro was probably global ex-US, how much limit do you have for the rest of the year if we do have another major international event, or did you need to buy back up cover post Japan to protect yourself for the rest of the year?

Some of, frankly, is single shot. We do have -- some of the limit is exhausted. It is, it's not something we need to go out and replace. Again, thinking about it, putting into context of how we think about, these are portfolio shaping. The shape of our portfolio is a little different. We have multiple different levers to use. Very, very comfortable with the way the portfolio is situated. Post loss of some of that retro sessional cover.

So, there aren't any individual countries where you might be a little bit uncomfortable and you may have to cut back your gross because you don't have that retro or maybe they are not peak zones, but okay. Just is there enough limit left in those retro covers that you did hit to handle further development or is any adverse development we might see from Japan going to be gross equals net?

We have additional retro that has not triggered. It won't flow through dollar for dollar necessarily. Some of the retro is more explicit than the size of the loss. Than the type of covers that have been triggered. It's a little bit of a mixed bag. We definitely have more retro sessional available. It depends on how the loss flows through whether it will be triggered or not.

And do you have any thoughts if we do see development for the industry? What do you think the risks are from where it's coming? The ones that I through out there is, is this a debate over CBI, and then maybe, is that a bigger issue or is it related to typical quake, it takes a long time, especially given the slow reporting of this and to find out if the damage takes a long time to report structural damage that we don't find for a while?

I think it's, frankly, all of the above. It's -- quake losses are complicated. I think the CBI component of this loss, particularly for the large players outside the US, is a risk. I think there is a couple of places where it can come from, but there is nothing that is glaring unique about where this development might come from.

We do, for the record, if you would like to it, the breakdown on the recoverables.

Just in terms of the recoverables, about three quarters of the recoverables are due to the Japanese quake, and that breaks down both in terms of reinsurance claims, as well as the recoveries that are included in the ceded reinsurance contracts covered or accounted for at fair value. The majority of the rest of it is the New Zealand quake.

Got it. And if I could just ask one more quick one on a different topic. Your paid losses were very low in the quarter, they were pretty low all of last year, which I've been waiting for the Chile and New Zealand one to flow through your pays and it seems very little of that has hit and it seems almost none of the Q1 paids, Q1 events came through paids this quarter. Can you give us some sense for 2010 plus Q1 events, when we will start to see the paids come through?

I honestly can't. It's really a matter of how -- we are paying as paid. We've got a turn around time once we get the notification is to try to get the payment within 48 hours. So, it's reflected very quickly. There is no lag from this side. The one thing I would say is, as there is more of a retro component to any of these losses, it tends to take longer for payments to work through the system. Other than that, it's very difficult for us to know how the dollars are being paid on the ground and how it's going to flow through to us.

Are the events from last year paying out very slowly?

It seems looking at some of the other stuff, the information losses pay, certainly paying slower than hurricane. Not a ton equates to comparative, but it does seem a little bit on the slow side.

Great. I think that's all I have for now, thanks.

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

Earlier in your prepared comments, you spoke about the model changes that some of your clients were affected more than others, is there any broad strokes that you could describe the kinds of clients who would be effected as opposed to others?

I think that's the kind of comment we are trying to avoid because it's not -- we think about it on a customer by customer basis. There are things within the vendor models that when you look at it you can determine what it is, whether the filling rates are causing more inland losses or whether they have reduced the losses on the coast. But really it's not something I would look at on a macro basis, it's really the construction of an individual client's book.

Maybe asking another way, compared to the advice that you give clients, to what extent do you think that your client are sophisticated in their understanding of catastrophe risk?

I think, what we try to do is help them -- this is what we live and breathe. They are in often times in the insurance business managing lots of different types of losses, not just cat. We think they are sophisticated in how they are thinking about it, but they have so many other things to think about and we are specialists, that it brings a unique perspective to them.

The other thing I think is kind of dividing the conversation, as I mentioned earlier, is we help them understand version to version, but we also give them an independent view of their risk and hopefully that is helpful in letting them calibrate as to how they are thinking about vendor models across the vendor models and versions within the vendor models.

Thank you. My other questions have been answered.

Thanks.

You have a follow-up question from the line of Greg Locraft with Morgan Stanley.

Thanks for taking the follow up. Wanted to nail down the guidance on the top line. I apologize, as I'm doing the math, I just want to be clear, you're managed cat premium guidance is for, is to be up. Now modestly, as opposed to being down 10. Half of the book is renewed and it's renewed at down 3%, so mathematically, that means that the next half, just to hit your guidance of up modest, and the up modest I'm going to define as up 5%, means that the remaining book is going to be up 13%. Is that what you are trying to say? Is my math wrong or where is the logic wrong?

There is one thing, we are going to lose some business and we are going to write some new business. The change of the construction of the portfolio is going to look a little bit different, I think is one thing. The other thing, is we don't have -- we may change where we are attaching depending on what we see in the market as we may move up on programs or down on programs. I think it's harder to put it as your book is static, if your book remains static, here is the price increase. We look at it as, our book looks like this, with the way the world is going to change and how we want to shape our portfolio, we think it's going to like that. Your math, obviously, is right if you are just taking a percentage increase on the static book, but our book won't be static and will not look exactly like it does now by year end.

Okay. No I totally appreciate that and you've even said you have gone from an expectation of a down 10 to flat and you have already baked in a down 3 for half the book. We are all trying to wrestle with where is the market at the margin. You are obviously confirming that demand is picking up and there is a supply demand dislocation. It's very hard to model this when you sit in our shoes and we have to go off of what you guys are saying.

You are now getting a double -- the rest of the book should renew up double digits. Premiums should be up double digits from here is how we should be thinking about that, and the other big thing I'm taking away is, that's got a multi year projectory to it based upon what you have laid out on this call.

Greg, just backing up at the risk of getting too in the weeds on this, I think the -- I was with you until, when you got to the point about the first half being down 3%. I think the way you outlined the math is probably the right way to think about it. I think the difference is what anyone's, we don't want to pin down what modest growth means in terms of a specific point estimate. That's why when I gave the math at up 7%, to be flat, anyone can pick their own estimate of what modest growth can be, we just didn't want to give a specific point estimate, but I think generally speaking you've got that math right.

I think the other point Kevin was trying to make is that we are not forecasting double -- necessarily a double digit increase in prices, we are talking about premiums which reflects changes both in the level of demand and price.

Absolutely. Yes. I totally understand that. Great, thanks for clarifying that. That's actually very helpful for us modeling going forward, thank you.

Thank you. Operator, I believe this maybe our longest earnings call I've ever on been, so I think at this we probably ought to close it down. We've got some good questions today and we appreciate everybody calling in and look forward to speaking with you next quarter.

Thank you for participating in today's conference call. You may now disconnect