Arch Capital Group Ltd (ACGL) 2018 Q1 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the First Quarter 2018 Arch Capital Group Earnings Conference Call. (Operator Instructions) As a reminder, this conference is being recorded.

Before the company gets started with its update, management wants to first remind everyone that certain statements in today's press release and discussed on this call may constitute forward-looking statements under the federal securities laws. These statements are based upon management's current assessment and assumptions and are subject to a number of risks and uncertainties. Consequently, actual results may differ materially from those expressed or implied. For more information on the risks and other factors that may affect future performance, investors should review periodic reports that are filed by the company with the SEC from time to time.

Additionally, certain statements contained in the call that are not based on the historical facts are forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. The company intends the forward-looking statements in the call to be subject to the safe harbor created thereby.

Management also will make reference to some non-GAAP measures of financial performance. The reconciliation to GAAP and definition of operating income can be found in the company's current report on Form 8-K furnished to the SEC yesterday, which contains the company's earnings press release and is available on the company's website.

I would now like to introduce your host for today's conference, Mr. Marc Grandisson and Mr. Mark Lyons. Sirs, you may begin.

Thank you, Crystal, and good morning to you all. Overall, our first quarter results were excellent and demonstrate the value of our diversified specialty insurance platforms. Before commenting on market conditions, I would like to review the core tenets that successfully guide us at Arch.

Our primary goal is to produce superior risk-adjusted returns in order to drive long-term growth and book value per share while providing customers with quality insurance products. To support this goal, we hold dear a few core principles such as cycle and capital management as well as being intellectually honest about the probability of achieving the risk-adjusted returns offered by the marketplace. Our shareholders, policy holders and employees all gain from this approach.

Currently, market conditions are stable to slightly improving in the P&C arena. Operating margins expanded slightly in insurance in the first quarter, while the interest rate environment has lifted expected returns. Despite growth in some niche areas, we remain cautious and do not see a broad-based market turn in the near term given abundant capital across the market.

It's important to keep in mind that loss trend is picking up and is, at best, a guess. We will not know for another 5 years what the recent changes in trend will mean for our P&C businesses. Even though there may appear to be an increase in ROEs, the uncertainty of the impact of inflation as well as some of the negative effects of changes in terms of conditions of late tempers our enthusiasm.

In our insurance group, underlying the growth in our gross written premium, we continue to deemphasize some lines such as casualty, excess D&O and some London market business, all owing to an overly competitive marketplace. Our insurance growth is coming from travel and small to medium enterprise professional lines. In addition, premiums increased in loss-impacted property lines, where rates and returns are improving.

In reinsurance, our growth in European auto quota share and in excess of loss as well as property is balanced out by decreases in casualty and D&O. As you can see, our insurance and reinsurance operations are in sync as to where capital needs not to be deployed.

Lastly, keep in mind that the reported growth in premium was magnified by foreign exchange impacts this quarter, to the tune of 1/3 of growth in both insurance and reinsurance.

Turning briefly to capital management. We entered into a loss portfolio transfer on certain discontinued liability lines and program businesses, predominantly from years prior to 2012. We are no stranger to the runoff market, and we value what this product can offer. The transaction also included reserve development protection above the carried reserves. We did that transaction with 2 main objectives in mind: first, it will reduce the volatility of future reserve development, narrowing the ultimate payments around the level currently expected; and second, it will enable Nicolas Papadopoulo, who's our new insurance group CEO, and his team to focus on ongoing projects without being distracted from running off a business no longer core to Arch. Mark will further address this and additional capital management actions in his own comments.

Turning now to our other specialty segment, mortgage insurance. It was an active quarter in the press to say the least. We had a great quarter, and we are even more convinced that our risk-based pricing framework, RateStar, is the best way to approach this marketplace. Our new insurance written for the first quarter was $11.4 billion, of which 82% was through our RateStar platform. The pricing outlook was looking fairly stable until recently, when competitors announced the cut to their rate cards. You will remember our comment on last quarter's call that we expected this reaction. However, like you, we did not think it would occur this quickly, especially in light of the uncertainty surrounding PMIERs 2.0. We are, as is everyone, currently evaluating the competition's rate card, and we will decide whether to take action soon. Bear in mind that our production from rate card is less than 20% of our NIW, and we believe that RateStar will still attract the better risks even after the rate cuts announced by some in the industry are put into effect.

As I mentioned a minute ago, RateStar has proven to be a great way for Arch to enhance risk selection. One example of this is that RateStar steered Arch away from originations in high-LTV, high-DTI products over the last few quarters. Along with single premium products, we purposely remain underweight in these higher-risk areas. Our expected returns on all U.S. MI business are still in excess of 15%.

Of note, we closed another Bellemeade transaction for the second half of 2017 production at tighter than the one we did in last year's third quarter. Bellemeade structures provide capital market protection for Arch for deterioration in the mortgage market. Think of it as an aggregate excess of loss covering -- cover attaching excess of a 23% loss ratio.

Turning now to IMAGIN, the Freddie Mac product announced last month. We believe that this product was an evolution of GSE credit risk transfer and not a revolution. This pilot is still very much in its infancy, and we believe it has the potential to do 2 positive things for us: first, it establishes Arch as a go-to innovator in mortgage insurance; and second, it leverages our underwriting expertise to managing insurance platforms and third-party capital. IMAGIN targets the discounted single LPMI product, and it is capped at $2.5 billion of NIW, which is projected to be less than 1% of the expected MI industry production in 2018. In addition, this new structure fits our core principle of cycle management and allows us to be a low-cost provider in a highly commoditized business environment. Once you factor in the fees and the expense savings, the expected returns are appropriate benefit for the risk that we're assuming.

Last but not least, the new CRT advisory relationship that we have agreed to with Munich Re is yet another example of our ability to leverage our experience and expertise in executing various types of MI risk transfer. On the investment side, we continue to position our portfolio to be flexible and poised to recover quickly from an increase in rates and yet remains liquid enough to allow additional longer-term alternative investments. Our property cat exposures are substantially the same as last quarter, with our 1-in-250 year peak zone, the Northeast PML, the largest, at 6.2% of tangible common equity. Our RDS for mortgage insurance, driven largely by the U.S. primary exposure, is stable at 16.4% of tangible common equity as a result of the growth in insurance in force and the increase in persistency in U.S. primary MI, largely offset by the new Bellemeade transaction. We are continuing to refine the RDS for our non-U. S. business, and we will report any changes to the current view as they evolve.

In closing, book value per share grew to $61.24 at March 31. Strong operating results were partially offset by the effects of volatility in the financial market. In summary, a good quarter with some very early positive signs in our P&C operations and a continuing well-performing MI on the back of conservative, proactive capital and investment management.

And now I will turn it to Mark.

Great. Thank you, Marc, and good morning to all. I will make some summary comments for the first quarter of 2018 on a core basis. And as I say every quarter, the term core corresponds to Arch financial results excluding Watford Re, whereas the term consolidated includes Watford Re.

So big picture perspective, after-tax operating earnings for the quarter were $235-plus million, which translates to an annualized 11.3% operating return on average common equity and $1.69 per share. Book value per share was, as Marc just said, was $61.24 at the end of the quarter, which represents a 0.5% increase from last quarter and 6.2% increase from 1 year ago despite a negative total investment return for the quarter. The diversification of our operating platform and within our investment portfolio proved invaluable towards increasing book value per share in a very challenging economic and insurance environment.

Moving on to operations. Core losses recorded in the first quarter from 2018 catastrophic events, net of reinsurance recoverables and reinstatement premiums, were $2 million or 0.2 loss ratio points compared to 1.2 percentage points in the first quarter of 2017 on the same basis, approximately evenly split between our insurance and reinsurance segments. As for prior period, pure net loss reserve development, approximately $52 million, a favorable development of 4.7 loss ratio points, was reported in the first quarter compared to 8.3 loss ratio points in the corresponding quarter of 2017. This was led by the reinsurance segment with approximately $37 million favorable, the mortgage segment at approximately $13 million favorable and the insurance segment contributing $2 million favorable. The reduction in net favorable pure loss development relative to a year ago was driven by a lower level of reinsurance casualty releases and a lesser amount of U.S. mortgage second lien subrogation recoveries and fewer accident years contributing to U.S. mortgages' first lien releases. Net favorable development associated with prior year catastrophic events totaled approximately $12 million this quarter, predominantly driven by releases on Hurricane Harvey.

Before I comment on our individual segment results, I'd like to update you on capital management actions we've taken through the first quarter of 2018. As you recall, in the fourth quarter of 2017, we executed roughly $1.4 billion of internal loss portfolio transactions between our U.S. property and casualty insurance subsidiaries and our Bermuda operating company. Additionally, effective January 1, 2018, we canceled all internal property and casualty insurance and reinsurance in force quota share treaty on a cut-off basis, the net effect of which was to improve the risk-based capital ratios of our relevant U.S. subsidiaries. In future quarters, we will provide updates on any further actions taken, and I will comment on share repurchases later in these comments.

On a related topic, as Marc just referenced, we've announced in latter part of April that Arch Re Ltd. entered into a transaction with Catalina General Insurance Ltd. At inception, approximately $400 million of subject reserves were transferred, accompanied by an approximate $200 million adverse development cover. Catalina will assume all claims-handling responsibilities, and the transaction is heavily collateralized to secure Catalina's obligations, with a meaningful margin above 100% of all transferred reserves throughout the life of the contract. It should be noted that although this was a transaction between our Bermuda operating company and Catalina, the underlying exposures emanated from the U.S. insurance group.

Moving now to -- more so into operations. The calendar quarter combined ratio on a core basis was 78.8%, identical with the first quarter of 2017 and lower compared to the 82.5% serially for the fourth quarter of 2017. The core accident quarter combined ratio excluding cats improved to 83.2% compared to 86.1% for 2017's first quarter.

The reinsurance segment accident quarter combined ratio excluding cats of 93.4% showed 420 basis points of improvement compared to first quarter of 2017's 97.6% combined ratio. This was driven by expense ratio reductions, with a corresponding accident quarter loss ratio quarter-over-quarter. The reinsurance segment expense ratio benefited from reductions of operating expenses in the dollar sense combined with larger net earned premium base. In addition, a reduction in federal excise taxes of $2.5 million or 90 basis points due to a reduction from the cancellation of certain intercompany property and casualty quota share agreements that I've referenced earlier. This benefit will continue to accrue for the remainder of 2018.

The insurance segment's accident quarter combined ratio excluding cats was 98.7%, up slightly from the 97.8% in the first quarter of 2017 due to higher acquisition expenses resulting from mix of business changes, with also a corresponding flat accident quarter loss ratio. However, on a sequential basis, this quarter's accident quarter combined ratio improved 100 basis points over the fourth quarter of 2017 largely due to a lower level of reported large attritional losses relative to recent quarters.

Moving to the mortgage segment. There, accident quarter combined ratio improved to 43.4% from 50.4% in the first quarter of last year as net earned premiums remained relatively flat as a percentage of total, being approximately 25% to 26% in both quarters. The accident quarter loss ratio of 20.1% in the first quarter of 2018 compares favorably against both the 21.5% ratio in the same quarter of 2017 and the 25% ratio in the fourth quarter of 2017. The expense ratio also improved from the 28.9% in the first quarter of 2017 to 23.3% this quarter, reflecting the benefit of a full year of integration efforts following the acquisition of United Guaranty Corp. However, on a sequential basis, the expense ratio increased 120 basis points from 22.1%. As we have previously discussed, this is driven by an increase in the amortization of deferred acquisition cost. Remember that at the closing of UGC transaction at prior year-end, all deferred acquisition expenses were written off to 0, and they are now rebuilding themselves and being amortized into income.

Total investment return for the quarter was a negative 32 basis points on U.S. dollar basis and a negative 40 basis points on a local currency basis. These returns were impacted by the effects of higher interest rates on investment-grade fixed income securities and the overall equity market decline, partially offset by positive returns on alternative investments and non-investment-grade fixed income. The investment duration was 2.6 years at the end of the quarter, down sequentially from 2.83 years at December 31 and down from 3.36 years a year ago in anticipation of rising interest rates. Also during the quarter, fixed-income investments, which represent approximately 76% of investable assets, saw a tactical shift away from municipal bonds, which were reduced by 28% in the quarter, and into corporate and AAA- asset backed securities securities due to improved relative valuations.

During the quarter, the company incurred $111 million of pretax net realized losses primarily as a result of the already referenced investment mix shift. And included within that realized loss was $18.4 million of unrealized losses in equities under a new accounting principle that requires recognition in net income of changes in the market value of equities rather than in other comprehensive income.

As you know, our investment portfolio continues to be managed on a total return basis and not by component of total return. The corporate effective tax rate in the quarter on pretax operating income was 9.9% and reflects the benefit of the lower U.S. tax rate, the geographic mix of our pretax income and a 0.5% benefit from discrete items in the quarter, mostly stock-related. As a result, the pure effective tax rate of pretax operating income, excluding these discrete items, is 10.4%. As always, the actual full year effective tax rate could vary depending on the level and location of income or loss, the level and location of catastrophic activity and varying tax rates in each jurisdiction.

On a GAAP basis, at March 31, our total debt-to-total capital ratio was 18.7% and total debt plus preferred to total capital was 25.7%, down 70 basis points from year-end 2017 and down a nice even 300 basis points from year-end 2016, when we acquired United Guaranty. This leverage reduction was due to our growth in common equity and the redemption of the remaining $92.6 million of the Series C 6.75% preferred shares that took place. Associated with this redemption was a $2.7 million nonoperating charge to expense in the original issue costs of the remaining Series C, which had been held as additional paid-in capital.

As for share repurchases, at the end of the first quarter, under our Rule 10b5 plan, we implemented our share repurchase program during our closed window period and repurchased nearly 40,000 shares at an aggregate cost of $3.3 million. Additional share repurchases have continued into the second quarter and cumulatively totaled $80 million, with an average price to March 31 book value of 1.33x. Our remaining authorization, which expires in December 2019, at the end of March was $443 million and, considering the share repurchases made through April 30, now stands at $366.5 million.

Also during the quarter, AIG completed the conversion of all their remaining convertible preferred shares issued as part of the UGC acquisition, resulting in the issuance of approximately 5.7 million common shares. You may recall that these shares were considered common stock equivalents in 2017, so the conversion in the quarter had no impact on earnings per share or book value.

Operating cash flow on a core basis increased to $370 million in the first quarter of 2018 compared to $122 million for the same period in 2017, reflecting the growth in premiums written in 2018, a smaller level of operating expenses and UGC transaction costs and a $52 million tax refund received.

With these introductory comments, we're now prepared to take your questions.

(Operator Instructions) And our first question comes from Kai Pan from Morgan Stanley.

My first question is on the MI business. Given the competitive pricing cut as well as with new pilot program, what do you think about the return on the business going forward versus your prior expectations?

So the current returns are in excess of about 15%, which we indicated in the past and still believe it is the case. After we look at the price cuts that were announced by one of the major competitors, and they actually sent around a sheet that explains how they get to -- how they factor in the tax changes, the returns are still in that area, still about 15% despite those rates. So that's what we would expect it to be. Having said all this, not everything is created equally. We're going to be looking very carefully at our RateStar framework and see whether we need to make a few changes and -- as well as looking at the rate card changes that took place. It's still a very, very good marketplace overall. The credit quality is still very, very high. So we are not changing, fundamentally, the level of return, especially risk-adjusted return as it compares to other lines of business that we would have in our portfolio.

And I would just add to Marc's comments that we really look at this as a segment, not as just the U.S., which is where our competitors are kind of vertically focused on. So our view of CRT transactions, our other businesses that we have, the fact that we lay off, we have reinsurance structures and Bellemeade structures that all are very additive towards the net ROE.

Okay. That's great. My second question is on the P&C side. What's your pricing outlook for Q1 renewals, EMEA renewals? We heard some countries that pricing actually would not be as strong as general renewals. And do you see the same thing? And how do you position your portfolio?

Yes. So we have heard -- we went through, internationally, renewal of the Japanese -- for instance, at April 1, I'm sure you heard on another call that pricing was tepid, fairly stable to slightly down or slightly up, depending on the layer -- or the types of risk. So we were expecting sort of that reaction. But the most important piece, I think you're asking, is what will the U.S. reinsurance market look like at midyear? And the initial indications are that it's not going to be as good as the rate increases were at January 1. A lot of it is also posturing. There's a lot of early -- it's still pretty early. June 1 and July 1, there's a lot of renewals taking place, so people are jousting for positioning and arguing their case as we speak. But the early signs are that the price increase is going to somewhat go down. So the second derivative is negative to the rate change. It might be still a rate change, but it's not going to be as good or as healthy as it was at 1/1.

One other thing, Kai. On the underlying businesses to which the property cat attaches, we're certainly seeing some uplift at Arch insurance group, and we believe those uplifts are happening on the quota shares and the XOLs that Arch Re attaches on top of them.

Our next question comes from Elyse Greenspan from Wells Fargo.

To start, a couple of questions on mortgage, and then I do have a P&C question as well. In terms of mortgage, did you say about how much of an earnings it was in the quarter? Is it still about that 60% level you had provided us with in the past?

I think that's -- well, yes. That's with the allocation of our investment income that we show in the corporate segment. If you allocate that back, I think it's roughly about that. It might be a couple points north.

Okay, perfect. And then in terms of -- a lot changed in the quarter in terms of the mortgage environment, with IMAGIN, the CRT deal in relation to -- with Munich as well as the price cuts in the industry. And you guys say you still see this business is generating about 15% ROE. But what about -- how about -- do you think about it in terms of the overall earnings? Because obviously, some of these different components can either increase or decrease the forward earnings that you can generate from mortgage. Can you kind of help us think through the moving pieces and how the profile has changed with these new developments, whether -- obviously, it's not just this year but more thinking about the earnings a couple years out.

It's a very good question. I think -- and it speaks very well to our ability to pick and choose where we're going to allocate capital depending on the return characteristics, on the CRT, for instance, or is it IMAGIN. If that program takes off and becomes bigger even in the future, that will also allow us to participate there. We also have U.S. primary in mind, as we've mentioned. That's also a good lever for us to utilize. It's very -- the way we look at the MI business is very similar to the way we look at any other business. You have to tell me what the marketplace looks like as we speak, and I will tell you what we -- how we will be reacting. So depending on the relative returns between the CRT, the IMAGIN or other types of the structure of its sort and/or primary MI, we'll be allocating capital as we see the returns get better. For instance, this quarter, a good example is we have allocated less capital to the CRT transactions. We saw the spreads tightening to a level that we believe is not as acceptable as we would want and not meeting our threshold return. But it doesn't mean that we need to deploy capital in some other areas to cannibalize the other segments. It's really just a deal by deal, area by area, looking at transactions, making sure we're maximizing the returns. It's really hard -- I guess the short answer is I do not know until we get to what the market is going to give us in the future.

And Elyse, I'd just like to -- regarding your next question, just to also caution, this is a pilot. It's extremely early. We don't even have a lot of visibility yet into how it's going, so -- which we'll certainly talk about in future quarters. But also, just be cognizant that IMAGIN is towards U.S. MI, whereas the relationship with Munich is more towards the CRT transactions. So when you picture Marc's comments about cycle management levers this creates between using working capital versus risk capital, this innovation that the mortgage guys came up with allows that cycle management to really take effect.

Correct.

Okay. That's helpful. And then my last question. In terms of -- you guys returned to buying back stock in the quarter and subsequent to the quarter. Can you help us think through your excess capital position, how you would kind of balance other -- continuing to return capital with your shares at this kind of 1.3x book value level? Or if M&A -- potentially, a deal on the P&C side might be something that you would want to conserve capital for? How are you thinking through that decision-making right now?

Yes. Great question. And it is kind of the (inaudible) of a lot of the things that you just mentioned. When we did the 10b5-1, we didn't expect certain things to happen from our competitors that kind of weighed in depth and down on our stock. But what we've done historically with that wavy, not quite straight, lines for your paybacks that we've talked about, that's an ingredient into the mixture. A view of the off-balance sheet embedded value is a -- helps inform but doesn't drive some of the decisions. But there's other things that we have going on. There's always things in the pipeline that we're entertaining, firstly. Secondly, we still are steadfast towards reducing our financial leverage that emanated from the UGC transaction for a couple of reasons. One, the more we do that, it gives us dry powder for other things in the future. We've made commitments and in discussions with rating agencies. And it's also an aspect of our GSE relationship that will be helpful to us as we delever. So there's a lot of usages for cash, some of which might go towards -- depending on what is the highest return and value that we see outside of some of the benefits that might accrue from the deleveraging.

I would also add that this is the competition on the allocation of capital and how we deploy it, and certainly, we felt, when we did the 10b5-1, that this was an appropriate relative allocation of capital. And to Mark's point, there was no insight on our part as to what the markets -- how it developed. We're going to have a board meeting next week, and we're going to have all units sitting around and discussing through what projects or what it is they're working on, and we're going to have a more detailed discussion next week and determine what we're going to do going forward.

And one other thing that I could add is compared to if it was 3 years ago, with the volatility of PC and so forth, we have a lot more clear visibility down the next couple of years of mortgage earnings and the quality and strength of them because of the way it operates with the (inaudible) and the persistency attached to it and so forth. So that also helps inform our decisions that way.

And our next question comes from Amit Kumar from Buckingham Research Group.

A few quick questions. Just first of all, going back to the opening remarks, I think you mentioned rate card serves 20% of the business. Is it fair to say that the competitors, it's close to 100% or so? Or what is probably the number?

The competitors are not doing any RateStar as far as...

No, rate card.

Sorry. The rate card, yes, it's 100% for everyone. We're about 18% rate card for the production in first quarter of 2018. Yes, that's the answer, yes.

Got it. Now that's what I wanted to be sure. So clearly, the stock overreacted on the news last month. The second question I had was on the time line. So you mentioned that you're looking at what to do following the pricing discussion. Do we have any idea -- I mean, is this going to be disclosed very shortly? Or does it take a few months? And then I guess you are heading to the PMIERs capital discussion. I just wanted to be clear on the timing of your decision.

I think it's going to be like -- the way we look at the pricing and the way we deliver products to our clients, having RateStar as well as the rate card and having, I would add, different distribution, community banks and credit union, for instance, we need to be very careful and thoughtful as to how we homogenize, if you will, the way we're delivering the pricing and the product to our clients. Right now, what's happening is that there are discussions as we speak, and the discussions started 2 weeks ago in Greensboro about how we're going to juggle or put together in a cohesive way our reactions to the -- on the rate card and what it means for RateStar, if it means anything at all. So I would -- the June 4 or thereabouts, the first day that the pricing will be in line for the [Magee] and, I believe, Genworth as well. So we will have to come to conclusion with the rate card in shorter order. The RateStar changes may take a little bit longer to implement because, as we have mentioned before, it's over 1.3 million different (inaudible) in decision-making. It's not as easy at it looks. It's a lot sturdier, but also being as granular as it is, it's probably less impetus to draw very quick conclusions to it. We can let this work itself even as we speak and even after June 4. But we'll definitely be proactive in making that determinations. I would fully expect by early June, we'll have full -- total, complete picture as to what we're going to do on both rate card and RateStar, if any.

Got it. That's actually very helpful. The only other question I have is going back to the insurance segment. And if I look at Page 12 of the supplement and you look at the reserve development number, it's very close to sort of 100%. And I'm trying to think, is there something deeper going on in terms of there is a moment in terms of certain lines which might be seeing adverse, and hence, the net number is just modestly positive? Maybe just help us better understand what's going on and why is it hovering so close to 100%.

So it's really -- the rate of reserve developing any 1 quarter, it's haphazard. It could be some negative in one area, some positive in some other area. A quarter change is very hard to pin down. And sometimes, you may wait 1 or 2 or 3 quarters before you take action in certain lines of business. You may want to do a catch-up on one area. So the short answer is the sum total is the sum total and is really a result of individual business units, which we have 14 of, where we go through each individual one of them and we say, okay, this one needs a little bit more adverse development. Because some losses were reported, we're going to expect some other goes down, so it's really just a -- what you see on our financial result is really the bottom-up approach of our reserving analysis at the individual line level. And it's really a quarterly exercise that you go through. And sometimes, you tend to be more proactive in certain areas because you might think that it's the trend. It's going to go against you a little bit further down the road. And some others, you're going to wait and see whether this is only a one-time off thing. I think the short answer to you, unfortunately, there's no real grand design. It's really a bottom-up approach to reserving. And then I would say that some lines showed us negative or adverse development. Some showed positive development, depending on the quarter.

And I guess what I was trying to ask is, where does the combined ratio eventually settle based on the performance of this business?

I think our accident year combined ratio that I've mentioned, that Mark mentioned, 98%, 99% is roughly in the range of what we would expect the mix of business to be. And again, I would just caveat that by saying there are some trends happening in the marketplace. Some rate changes we see or we hear have happened, will they find their way to the bottom line over time has yet -- remains to be seen.

And I would say Marc, I think, was pretty clear on his prepared comments. The consistency between insurance and reinsurance is where capital is and is not deployed. And that's the function of the rates and relative to loss trend. So there's absolute returns and then what are the market conditions doing. Is it helping or hurting that absolute return? And that's how capital gets deployed. That's how the business mix shifts. And if that's successful in the shift, it could have an even more beneficial impact.

Yes. And now we even have -- to add more complexities to this, if you have the same book of business this year that you renew, what, a 2.75 5-year treasury versus last year, 1.8, you could have a very similar accident year combined ratio but a higher return in equity. So just to add this to the mix, if it's not configured enough for you.

Our next question comes from Josh Shanker from Deutsche Bank.

The travel, Accident & Health business, is that growth based on the company getting in place the right infrastructure to be able to handle that business? Or is that business seeing a difference in terms of its profitability, which makes you more hungry for it? And I guess, third on that, where is that business coming from? Is the pie getting bigger? Or are you taking that from competitors?

Okay. So I'm trying to get any answer. So it's coming from -- yes, we have a couple of programs that we won over the last 24 months, which helped us -- and we had the relationships that we had developed for a long time internationally as well as in the U.S. So it's really growing with new relationships that -- one of them is actually growing, it's the large reason why we've grown in travel over the last 12 months. The first question, yes, we have an integrated model, we have claims, we have pricing, we have portal. We also have RoamRight. As you know, we have business-to-consumer bent with business-to-business as well, which would be more wholesale or retail -- through a retail network actually, a little bit like having a program -- no, not a program but sort of a relationship with a couple of producers to really be their go-to-market in that segment. In terms of returns, this is not a very -- a super high-margin business. I think you'll see other people talk about it in terms of combined ratio. But in terms of capital usage, it is very, very effective in terms of capital usage. So we are trying to get into that segment. It's also very, very sticky. As you know, as you might expect, Josh, if you get the relationship going with the pipe and work with the product development with the guys who sell the product, it could be beneficial for a long time. So this has been going on for at least 4 or 5 years of growth.

That's very thorough. And then on the UGC 2014 to 2016 premium, I've been sort of guessing that the decay on older accident years lose about 20% of its premium annually. I don't know if that's right. But maybe how much of a net premium written growth tailwind is the UGC quota share years going into the past giving you?

Josh, I'd say you're in the ballpark. I think you're a little heavy on the degree of decay.

Yes. A bit lower than that.

And our next question comes from Geoffrey Dunn from Dowling & Partners.

Like yourselves, it seems like a number of the MIs continue to evaluate the recent BT monthly changes. But there's -- in the context...

Geoff, I'm sorry, we can't hear you.

Yes.

Is that any better?

Much better, thanks.

All right. So again, like yourselves, it looks like a number of the MIs are evaluating the recent BT monthly changes. But in some of the commentary, it suggests maybe there is some evolution going on in terms of how some companies are thinking about approaching pricing. What are your thoughts on the competitive environment if the industry started shifting to your approach where the rate card was available for the lenders that want it but a shift to more granular or even black box pricing for those that are looking for that?

So in a way, for us, it's music to our ears. It means that our model is the right model. And if you start having a 75 cell, you develop now a multiple of 200, 300 cells, as we see some of our guys developing, sort of refine, sort of rebuild, if you will, their risk-based pricing within the rate cards phenomenon. You're going to start multiplying these cells very dramatically. It might create issues for their -- the same issues that I think -- the large bank, for instance, who said they are not really willing to entertain at this point, which is the ability to cater to all these various permutations of pricing. So as much as people are finding RateStar, it seems like it's evolving into that direction. So to us, it's a little bit music to our ears. It sort of confirms that our model -- and a couple of our competitors made comments as such over the last week or so that this is probably more longer-term beneficial. There'll be some disruptions in the short term. I think that is probably your point that you're trying to make, and I think, yes, that is possible. But we do believe that it doesn't -- the more you multiply the number of cells, the more complexities you introduce in the delivery and pricing of the product at the loan origination, the desk level, so...

And I would just say on the boring side of it, but an important operational aspect, is the response time of something this complicated to return to the lenders in the manner in which they expect it and kind of shield this from that. But the response time has to be fast. So there was major investments that the guys did in that regard. So there's just -- it's as just important to have the 8 knives of the iceberg under the water as the 1 knife that you see above the water.

Yes.

Okay. And then you've had an interesting approach on some of the innovations that are effectively introducing a capital-light model with your Bellemeade deals, with the MRT, the Munich CRT. How much -- do you view your capital allocations independently of all those? Or do you view the return on a segment basis where maybe those capital-light opportunities give you more leeway on the capital-heavy opportunities?

So to us, it's really -- I mean, we -- for reasons that are -- the unit that's called MI, which is a global MI company, so their bonus plan and calculations of their performance is based on the overall segment's result. So they are -- they can fish in broader MI market whether it's insurance, CRTs, utilizing more Bellemeade transactions today. So true, if it makes sense from a return perspective, it could diversify. We're in different areas around the world. And at the end, they're all internally making sure that they're optimizing their returns. So we're really looking at it, Geoff, on a totality at the unit level and making sure that they -- having said all this, we have self-imposed guidance. There's so much capital waiting to expose for the shareholders' perspective to MI. But within the confines of those, that constrained, they have a vested interest in maximizing, optimizing their returns. We look at it holistically, if you will.

Which we don't view any differently than how the reinsurance group does it and how the insurance group does it.

That's right.

And our next question comes from Jay Cohen from Bank of America.

Yes, just maybe a small question on the MI. With the amortization of DAC now being part of the expenses, can you give us a sense of where you think the expense ratio will end up by the end of this year?

Well, one other ingredient that I didn't put in the prepared remarks was that there was some bonus catch-ups. I mean this is highly profitable. So to the extent that bonus throughout the year was under accrued and had to be made whole with the more recent year-end calculations, that gets reflected in the first quarter. And that's what happened. I mean, not only the profitability of the business but the excellent execution on the integration that they've done all filters into that. So rough -- it's going to be marginally better, and it could be lumpy on 2Q, 3Q, 4Q. But I would say on the balance of the 9 months, it's going to be marginally better.

And our next question comes from Meyer Shields from KBW.

I think we've had a couple of quarters now where you've been more cautious or sounded more cautious on loss trends. I was hoping you could dig a little bit more into what's driving the increase in service, if that's the right way of phrasing it.

I think, well, we are looking at trend -- I'm an actuary by training, I'm a recovering actuary, I'd like to say. If you look back at loss trend historically, it's a historical phenomenon, right? You have not developed data, tried to make an adjustment for what you think the overall CPI and, unfortunately, the insurance trend and inflation typically lacks CPI pickup. And to the extent that we've seen some inflation pick over the last 2, 3 years, they will not find its way through the projection of lost trends for a little while. And we've had a combination of things, right, audit premium on most of our segments that are datable. Pretty much we're always up on the upside, so there's more activity in the industry. So whatever you think your pricing, and whatever is happening in the industry, there's always been a mismatch. It's ongoing. It's subsiding a little bit, but we've had sort of a pickup in activity in the broad economy in the U.S. So the more there's activity, the more there's friction, the more I believe there is a possibility that lost trend could go and develop adversely against you. It's probably more of a prudent phenomenon. I think that the problem that we have in our business, mostly casualty, is that you're pricing on a forward looking, looking back at loss trend. And I think we've had undue benign loss experience over the last 8 or 9 years. I think it's larger as a result of the economy slowing down so much as a result of the great financial crisis. So it's just -- we're not saying it's going to go crazy. We're just saying that the likelihood of this being above what we believe and what we are coming out of our actual model, I think these are more likely than not. We tend to be more prudent when we factor in the loss trend. Mark, anything else?

Yes, I would just kind of echo what Marc again said where capital isn't going, where it's being allocated on D&O and, say, to excess casualty. There's loss trend in the actuarial arithmetic, and then there's trend not explained by actuarial arithmetic. D&O, for example, is incredibly lumpy year-to-year. There's no real projection about it, it's -- each -- and we don't always have a primary excess book. On casualty excess, actuarial arithmetic never works. Never works. It's always terms and conditions that really drive it. So when we say loss trend, we're also recognizing the slippage in terms of conditions.

Correct. Yes.

Okay. That's very helpful. And I guess, a second unrelated question. At [AFA], I think we talked a little bit more about -- I'm trying to think of the right way to phrase it. Pursuing individual opportunities in the insurance segment a little bit more rapidly than in the past, I was hoping you could update us on how those opportunities are bubbling up?

There's always possibilities around. I'm not here to tell you what we're going to do next month. I don't think it's fair on the call of that nature, Meyer. But I think that -- I think our comments at [AFA] also had to do with we're also going to be a bit more proactive in reacting to either adverse or positive reactions. For instance, property was a great example, right? Property rights increased a little bit. And I think we -- Nicolas took it upon himself when he took over in October, he said, listen, even though the rates, as I was saying in the third, fourth quarter, you need rates of 30% to 40% to really start pushing the envelope and do, make a significant commitment in capital to property. We still have some rate increases, and Nicolas said, well, we need to be a bit more proactive in positioning ourselves in that marketplace. I think that would have been not necessarily the way -- it's not necessarily the way a traditional insurance company would think all the time. And I think we're trying to bring -- which is more of an opportunistic way of thinking, which, I think is brought upon largely as a result Nicolas is bent on what he's done so well on the reinsurance. And our insurance group has taken up to it like fish and water.

Our next question comes from Ian Gutterman from Balyasny.

So I had one -- well, why don't I follow up real quickly on Meyer's question before I get to my main question. This is maybe as much of an observation as a question. Marc, how is it that some of these lines where we're seeing adverse development, every year, most of the companies seem to take them to a 65 every year in mid- to long-tail casualty?

Boy, asking the question is -- that's a very, very, deep question. I think that if you overlay what happened -- and we were on the receiving end of this, and Mark can attest to that as well when he was running insurance. When you were looking at results in 2012, '13, '14, pricing that ongoing business, sort of looking at the results over the years, you would look at '08, '09, 2010, and even '05 or '06, you would have lesser development than the actuaries. Who are they kidding? So we do a loss reserve analysis. Pretty much everything comes down below the expectation. So -- and this has been going on for a while. Actuaries or loss reserves specialists lose a little bit of their credibility after a while because it's kind of hard to deviate from anchoring yourself at a long-term level. It's very hard for people reserving to think that this is really a 30% loss ratio. And it's also the same way, very difficult to say it's not running 65, it's running 80. But since we're looking back and then you look back after 5 or 6 years, if you're an actuary right now or if you look at loss reserve development, you'll say, well, I think it's really at 75. These are the same people that we're saying, it should be booked at 55, 68, 5 or 6 years ago, and things have developed to be 56, 57, 58. So there's a little bit of a mismatch. It's not easy for people to reconcile the way the reserving is made. Most people, and I think we can be guilty of it ourselves as well, people tend to think of insurance as being not cycle-affected. But there is such a thing as cycle-affected. It's not a linear plus or minus 2 or 3 points, especially if you're specialty insurance companies like ourselves. So many moving parts, it's really hard to pin it down. And you have history as a guide. And the loss ratio around the long-term expected varies wildly. Unfortunately or fortunately, I think I like it because it creates opportunities for us in the future because people keep on booking 65% or 66%. When it turns out 85, 88, you have to recognize it. We'll be able to seize the opportunity of people, deemphasizing that line of business precisely. But it's going to take a while.

Agree, and that's helpful. It's how I remember things from the early days when we were first meeting on the island when you guys are being formed. This just feels like a similar story. But so my main couple of questions, one on the mortgages. I get obviously the advantage of having RateStar versus the card when other people are cutting rates. But how should I think about -- and I'll try to call out an example, it's probably not the best and -- but you can hopefully get the spirit of it. If there is a cell under RateStar that maybe was priced -- tended to -- because you looked at it in a better way, maybe it was a 20% discount to most people's rate card. And maybe you had, I don't know, a 50% hit rate or something on that cell. If everyone else is cutting rates, does that hit rate go from 50% to 25% even if you don't change anything? So even though you're not using the rate card, you become less competitive and need to maybe reconsider some of your pricing in the RateStar cells.

Yes. So we've been thinking about this. And I think the best way -- let me try to make an analogy from property cat exposure. Most of our analysts are P&C people. So let's think about 2 types of risk: hurricane in Florida and California quake. If you think about writing a line of business or pricing -- you price yourself at [0.2] online for a layer in Florida, attaching a $10 billion market loss for the overall event. 20% is the current pricing. You have an excess of $60 billion quake exposure in California. And that current pricing is 10%, right? RateStar might say that the current pricing for Florida -- I should be getting 22%, but the RateStar is saying 20%. So I'm going to not necessarily get -- win a lot of that business. At the same time, because of inefficiencies in the overall card, I can tell you that our RateStar pricing for that California risk, which is much higher, much as likely to be hit, is 10% when the rate card is 13%. So right now, what I'm going to be doing is focusing more my capital on the one that is at 10%. Two things will be evident to you is that I'm having a lower rate than the average person that writes something in Florida. So that's why for us, the average rate is a very, very misleading way to think about it. Now the rate card is 20% in Florida, and it's 13% in California. Next year, somebody cuts the rate card by 10%. That 20% goes to 18%, that 13% goes to 11% and change. What am I going to be able to write next year? My RateStar hasn't changed. I'm still at 10% in California, and I'm at 23% in Florida. So what's going to happen? I'm going to get even less of the Florida business and presumably the same or it's not a bit more of the business in California. So that's sort of what RateStar does for us. Does that make sense to you, Ian?

That makes perfect sense. I totally agree with that. I was trying to think if there were sort of cells in the middle were you would have gone from maybe something that was a 20%, and you were at 19% and you were getting business. And now you're over it, now you're 19% versus 18%. I don't know how big of a book that is, maybe that's just on the margin, it's not that big a deal.

Yes, well, to your point, I made two -- I made one extreme example about 2 different risks. But you're right, there's a lot between the spectrum that goes, and that -- we have David Gansberg team, Allan and then John Gaines, spending an amazing amount of time dynamically connecting with clients and looking at production on a daily basis and try to figure out what will work. RateStar is sort of a floor of sort and we sort of over and -- we put the pricing that we think will sell in the marketplace that gives us a return obviously, but we're not trying to leave money on the table, but we're trying to be competitive and take the best risk, as in the example I just mentioned. There's a lot more going on. And you're quite right, I mean there's 1.3 million cells, 17-ish, different -- it's a very arduous process.

Yes. I'd just -- I would just say -- you probably heard some of the other questions. Geoff Dunn talked about what others have for RateStars and have for refined pricing. Then I think your question is a lot more relevant. I think right now, it looks like we got an old buckshot -- they have an old buckshot musket, and they're trying to hit an ant where they need a scalpel. And over time, I think your question is going to have a lot more relevance.

Yes.

Got it. And if I could just quickly on the Catalina transaction. Just can you give a little color on what U.S. lines of business were in there? I guess, I don't really think of you guys having a runoff book, so I'm a little confused of what exactly you mean and what went into this transaction.

Sure. Well, actually, we kind of view this as our third action because some of these programs, as we talked about, we took -- terminate it. Yes, we talked about in the past call, past years, of terminating programs, and you're stuck with the runoff. We terminate it because we didn't like the results, we didn't like the emergence of claims, of the underlying coverage that allowed that to happen. And similarly, on the specialty casualty runoff, it's predominantly old New York labor law issues, California residual -- residential contractor business where you get -- you think you're done in 10 years but then they do repairs, and the clock starts over again and those kinds of things. So that's really it. So there's no ongoing customer continuity issues, things of that nature. So given that those decisions were made, and I think they were the cracked ones, and then we still wound up, you'd see it in our 10-Ks, still having some issues with it, we said, let's look across the board on capital management. Let's just try to solve it once and for all.

Okay. And then even though it backdated to 1/1, any -- since the deal was written in April, is any financial impacting in the [shelf] in Q2? Or is it already up and accounted for in Q1?

Well, what you wind up happening, it's going to be Q2. So - but remember, ultimately, there's a difference between statutory and GAAP if there -- if the adverse development covers ever hit, Lord knows if it will be, but it gives us sleep insurance. But to the extent that it is statutory, then you get 100% of the recoverable immediately. But on a GAAP basis, it's kind of amortized. Then think of it similar to the Berkshire, AIG ADC and the way that works.

But we don't expect much, much change in quarter 2, not much impact.

And our next question comes from Ryan Tunis from Autonomous Research.

I guess, just following up on the ADC. Can you give us some idea of the amount of adverse development, I guess, maybe what you took last year, you've taken in the past few years in the lines that were subject to that?

I do not have that on my fingertips. Programs -- I would say -- I can tell that you on programs of the last couple of years, I believe the majority of the adverse is associated with these terminated programs. So I think that's the best color I can give you.

And it's the same on the casualty. If there were any adverse developments, the same on the specialty casualty book that Mark referred to.

Yes.

So it's -- and that's a pretty big -- it's a pretty good size of adverse, I think, from -- looking at the 10-K, right? So now that's a lack of headwind going forward.

That's correct.

Right.

Okay. And then I guess just -- I had a couple of bigger picture ones, I guess, on the MI conversation. And I guess, the first one is the whole discussion about mid-teens ROEs. At what return level would you guys proactively start writing less?

I think that you've seen and, as we speak, I think we have threshold, risk-adjusted. I mean, like I said, not everything else is created equally. But you saw some changes already in the 2 quarters where we -- well, probably deemphasize singles for a little while because we don't think the returns are there. This is -- it was low-teens, it's now, unfortunately, going a little below 10%, which is not acceptable to us. It's also portfolio thinking, right? You kind of think about it, Ryan, in terms of portfolio. So not every transaction is these could be accretive, they could bring diversification credit, even within the portfolio U.S. MI. But having said all of this, you got -- you heard about the singles, and you can -- the 2 other riskier areas that we've looked at, the high LTVs and the high DTIs, we have tended to go away because those returns went below the threshold that we have in RateStar -- embedded in RateStar. So for now we don't see any reason to start thinking about other than these 3 areas I mentioned in terms of riskiness. I think it's very -- it's a very ongoing, on a quarterly basis, on a weekly basis, actually, just reviewing what pricing is out there, what kind of risk is going on. And the other thing that I would tell you is now, as everything else being equal, different products could come at some point down the road in the future, so we'll be reacting to it when we see it. But that's all -- that's the best I can tell you.

And Ryan, I'll just add that the -- and Marc referenced it his CML discussion, he talked about the property cat and he talked about the RDS, which, I'll emphasize again, we're the only one with an RDS. But that is an important heavy forward focus and executive management focus. So we're 16.4% intangible. So it could be a combination of things. It could be a combination of front end, which we think we sculpt pretty well through RateStar, and it's a risk management tool on the front end. But because of the -- I think the excellent way the MI group has integrated their front-end pricing, the back-end, the RDS on the risk management side, they all inform each other, they're all -- in an integrated basis. So if Bellemeade, of that programmatic session, winds up becoming too expensive or you can't sell it out, what -- this is what I'd tell you, the outside world is having a different view of mortgage credit risk and, therefore, our net could go up more even though the front-end hasn't reacted yet. So it's a combination of all those factors that -- and the [effort] this team to take into account.

Got it. I guess, my follow-up is just I'm sure you guys have thought about this but just trying to think about the floor on pricing with MI in general. I guess the analogue I'm thinking of is the fact that property cat used to be a mid-teens ROE business. And your alternative capital and capital-light models, that just feels sort of familiar here. All of a sudden you got several years where all you're talking about is negative pricing, and all of a sudden you're back to pre-Katrina levels. And so that's obviously not been really much fun. So I'm hoping further you guys can cut me off the legs a little bit on that analogue. I mean, is there anything that, I guess in terms of the structure of the market or anything like that, that makes this you think less susceptible, I guess lower cost capital entering or, over time, competitors accepting sub-10% ROEs?

Right. So unlike property cat, right, you've written business for the last 5 years at a rate level that is pretty healthy. And that business continues producing returns and results for you as we go forward. On the basis and on the backs of, I would argue, very healthy increase in house prices, we have LTVs -- our current LTV is not an origination book, our current LTV in our portfolio, way south of 80, so it's -- not way south, south of 80% overall. So it's pretty healthy, a lot of equities, a lot of collateral in front of us. So we're actually in a very, very still -- we still have wind in our sails, if you will. So now if we play the tape going forward, right, the rates are probably 2 to 2.5x what they were precrisis. I mean, there's been a significant amount of price increase, and I'm not even talking about the kind or the types of product. So the kind of -- and then property cat is a 12-month -- it's a one 12-month commitment. It's a lot easier to change price, you change price on the fly for the whole 100% of your portfolio every single year. On mortgage, you always have this portfolio as it unwinds through time. So if I overlay this healthy house price index, lack of products, the bad product took place in mid-2000 that really created a lot of the issues, I look at a borrower FICO as an all-time noise, as high as it's ever gotten, there's a lot of room to give over time. But the question that you're asking, which we are -- we're never really asking ourselves because we're going live it together with our unit, it's going to take a while to erode that huge increase in quality and pricing that we went through after the prices of 2009. So we're not as -- so news of my -- of our debt is -- are greatly exaggerated. It's not -- it's going to take a little while before we get to a threshold of being dangerous -- too dangerous for us to stick around, if you will. But it will come, I just don't know when.

And Ryan, one other thing, back to your analogy. If it was -- this is longer duration, right? I mean, on property cat, capital comes in because pretty quickly, you can know how to exit. If longer-tailed liability streams were that comfortable for alternative capital, there'll be tons of casualty vehicles out there, but there's not. So this has a mortgage, as you know, has a lot of duration outflow and duration inflows that have varied interest rates and macroeconomically-sensitive. So I think that's quite a ways off. So I'd open that window and get back in the room.

And our next question comes from Mike Zaremski from Crédit Suisse.

I'll try to be fast, given it's past the lunch hour. Could you elaborate on what type of economics Arch receives from running Imagine?

We're not in a position to do those. I mean, we have strong NDAs, and there's a lot of communications that we need to keep to ourselves. But we are -- the returns are comparable to what we would get in a general business sense after we factor in our managing and operating and risk management and bring oversight to the cell.

Okay, awesome. Maybe we can follow up in the future quarter.

Yes.

And could you remind me, just staying on MI, have combined ratios from the rate card generated business been materially different from RateStar business?

That is one great question. And the answer is no as of yet, right, and we believe that the risk-adjusted pricing framework that RateStar gives us is going to be tremendous in more of a stress scenario. And we haven't had really -- we have some localized stress scenarios, but we haven't really gone through that exercise of analyzing it. And I think, if you look our loss ratio, when the wind doesn't blow or when the quake doesn't shake, everybody has a 0 loss ratio. So we're sort of in this relatively benign claims environment, and it's really hard to see it. And that's probably one of our biggest frustrations, like I said, as managers at Arch is that the fact that we're looking at the way we look at cat pricing and so where we -- for instance, in the way we structure our portfolio, when there are no losses, we don't look very good because we look like we should have done more. But the way we think about this, and we thought about it internally all the time, is we are very honest about analyzing the underlying economics and risk characteristics, fully recognizing that we could be wrong for a while. And MI is pretty much like a cat line of business in a lot of ways. But the short answer is no. We haven't done it, we don't expect it to be much of a difference on reported loss ratio.

Okay, that's helpful. And lastly, a follow-up to Josh's question earlier on travel and A&H, given it's -- it continues to grow at a nice clip. So felt like you guys were alluding to it being driven by a few relationships. I'm just kind of curious, are these relationships longer? If that's correct, is this -- are these relationships sticky, longer-term in nature? Or will this be kind of a line that's classic Arch, which will ebb and flow over time depending on the return profile? Like, I guess, is this a -- I know it's a short-tail liability, but is the distribution stickier?

We believe it is stickier. These are smaller items, there is more connectivity to the pricing, claims adjustment. And because a lot of travel is claims adjustment, right, you need to be able to pay the person that cannot go through their place or repatriate some of them. There's a lot of stuff you need to be able to do. So we believe it is sticker. But having said this, everything is stickier. But in the long run, everybody is there, right? I mean, in the long run, everything is variable on cost, if you remember your microeconomics. So at some point, if the pricing gets too out of whack, I'm sure everything is fixable. But it's relatively sticky in the short term, short to medium term.

And we do have a follow-up from Jay Cohen from Bank of America.

On Catalina, can you talk about the assets that get transferred over? I'm trying to get a sense of the impact on investment income.

Well, tell you what, let me answer the question more broadly than you asked it because I think you're trying to update your model, right? So look at it this way, we did an LPT at year-end, which was a bullet, I mean, the quota share has quarterly cash payments, right? So this is a bullet cash payment to our Bermuda operating company. We did the cancellation of the quota shares, which brings UPR back onshore with associated cash transfer. And then there's the Catalina which, in the scale of things, is not large. So it's effectively close to a wash between the investment income that you might get onshore or offshore because of all those flows back and forth. So Catalina, the reason I did that, of the 3, Catalina ranks third in size compared to the LPT, first; the UPR cancellation, second; Catalina, third.

And I am showing no further questions from our phone lines. I would now like to turn the conference back over to Marc Grandisson for any closing remarks.

Thank you very much, everyone. Happy quarter, and on to lunch now. We'll see you next quarter. Thanks.

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