Arch Capital Group Ltd.

Good day, ladies and gentlemen, and welcome to the third quarter 2020 Arch Capital Group earnings call. At this time, all participants are in a listen-only mode. Later, we will conduct a question and answer session, and instructions will follow at that time. If anyone should require assistance during the conference, please press star, then zero on your touchtone telephone. As a reminder, this conference call 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 assessments 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 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 8K 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. Mark Randerson and Mr. François Marin. Sirs, you may begin.

Good morning Liz and welcome to our third quarter earnings call on Halloween Eve. You're in for a treat. In our results, you can see tangible evidence of the advantages of the ARCH model. By protecting our capital during the soft market years, we are well positioned at each of our segments leans into improving market conditions. Our underwriters are making the most of the hardening property and casualty market, while our mortgage insurance segment is benefiting from record mortgage origination activity this quarter. This year, for the first time in nearly a decade, we've been able to grow significantly and deploy more capital in our PNC businesses that provide acceptable expected returns. And due to our strong financial position, we have accomplished this while maintaining a strong presence in MI, which continues to deliver meaningful returns. Our ability to continually rebalance capital amongst our diverse businesses enhances our total underwriting returns. It also should decrease earnings volatility over time. Since our inception, we have believed in cycle management and this strategy brings an added margin of safety to our collective underwriting activity. Allow me to elaborate on our third quarter results by touching on three key themes. One growth, two margin improvement, and three capital allocation. First, let's talk about growth. In this quarter, net written premiums in our PNC units grew 25% in total, 17% in insurance, and 38% in reinsurance over the same period a year ago. This growth was driven by rate improvements, but also reflects our ability to increase our participation where clients needed additional capacity. In the insurance segment, we continue to obtain strong rate increases in areas like property, DNO, and casualty. Group-wide, our rate increase for the third quarter averaged over 11%, and we believe this trend of increasing rates should continue through 2021. At Arch, we have always followed a simple rule. Our participation in the business should follow the direction of premium rates. As rates improve, we write more business. When rates decrease, as they did over the past several years, we write less. This strategy takes courage. You will often appear an outlier to the market, but being intellectually honest, disciplined, and applying our cycle management techniques is what we're all about at Arch. Obviously, the PNC market is broad, and all opportunities are not created equal. There are areas, such as workers' comps, where premium rates or conditions are not improving to the levels we believe are needed for an adequate return, and in those instances, we manage our appetite accordingly. Despite headwinds from the pandemic, our growth and insurance lines like ENS property, casualty, and professional lines are a great example of our platform's ability to flex into improved underwriting conditions. Our reinsurance unit has been able to lean into this hardening market both earlier and with more vigor than our primary operations. There are two main reasons for this. First, when a market transitions, needed rate increases compound up the insurance supply chain. Reinsurance is often a leading indicator of what's to come more broadly. Second, reinsurance can provide capacity quicker and in larger amounts since it can put capital to work through clients' platforms. Of course, sheer growth is only one part of the equation of growing returns. Let's turn now to margin improvement. We all know that, mathematically, rate increases in excess of lost trends lead to margin improvement. The marketplace seems to be supporting the momentum of continued rate increases. We are in the early stages of seeing the benefit of -on-rate increases in our operating results. Simply stated, adding the two parts, growth and margin, will lead to better returns. Many factors are driving today's P&C markets. These include elevated natural-cut cash loss activity in each of the past four years, weakened reserve positions from soft market years, lower investment yields, and a rise in claim inflation. Add in a global pandemic that is still ongoing, and it's not surprising that market conditions are changing. Now, pivoting to MI. Our $33 billion of NIW in the U.S. in the quarter was a record for ARCH. Low interest rates are producing huge refinance activity and, unsurprisingly, some churn in our in-force business. However, MI premium rates remain above pre-COVID levels, and the continued high credit quality of borrowers is generally better than it was pre-pandemic. We continue to face uncertainties, such as the economy's health and how the pandemic may ultimately affect individual borrowers. However, we are optimistic that, among other positive factors, recent trends in the U.S. housing market will mitigate the effects of the pandemic. Finally, ARCH's ability and willingness to allocate and manage capital remains a key competitive advantage. We always think about balancing our capital deployment over five pillars into the insurance, reinsurance, MI, into our investment portfolio, and lastly, into our stock repurchase. Our job is to optimize risk-adjusted returns through capital allocation across these pillars. We see managing the five pillars being similar to coaching a basketball team. We're constantly looking at how we can distribute the ball, i.e. our capital, to the right players. For the past several years, we've been able to feed the big 7'7 MI guy down low and rely on him to get easy dunks. Now, as the playing field, i.e. the market, changes, we've adjusted our tactics slightly and are increasingly relying on our two hot shooters, reinsurance and insurance. MI will still score its fair share of points, but the PNC players are getting more open three-point looks and layoffs. In short, our game is becoming more complete and diversified. Our ability to adapt to the new conditions is what makes us stronger as a team. The market dynamics take me back in time. We have talked about Paul Ingray's underwriting clock that helps track and measure the phases of the insurance cycle. It's been central to our management philosophy since the beginning and is a helpful reference to understand the underwriting lifecycle and assists us in gouging our risk appetite. I recently asked our underwriting teams where we were on the Ingray clock and the most common response was, Armani, the clock. If you take a look at the clock in our most recent annual report, you'll see that it's a very nice time to be at ARCH. There's a buzz among our underwriters because we've become the first call for so many of our clients. They know that we have the capacity, the expertise and the desire to serve them. Now, I'll turn the coach's whistle over to Francois as he goes into more detail on our quarterly results and I look forward to responding to your questions afterwards. Francois?

Thank you, Mark, and good morning to all. We at ARCH hope that you are in good health. On to the third quarter results. As a reminder and consistent with prior practice, the following comments are on a core basis which corresponds to ARCH's financial results excluding the other segment, i.e., the operations of Watford Holdings Limited. In our filings, the term consolidated includes Watford. After-tax operating income for the quarter was $120.3 million, which translates to an annualized .2% operating return on average common equity and 29 cents per share. Book value per share increased to $28.75 at September 30th, up .1% from last quarter and .2% from one year ago. The increase in the quarter was fueled by the continued strong performance of our investment portfolio and good unwriting results, taking into consideration the elevated catastrophe activity in the quarter and the uncertainty surrounding the current pandemic. Our property casualty teams continued on their path of solid growth and improved performance as we continue to see strong positive pricing momentum in their markets. Losses from 2020 catastrophic events in the quarter, including COVID-19, net of reinsurance recoverables and reinstatement premiums stood at $203.3 million or 12.5 combined ratio points compared to 5.2 combined ratio points in the third quarter of 2019. The losses impacted both our insurance and reinsurance segments and include $191.4 million from a series of natural catastrophes in the quarter, including hurricanes Isaias, Laura and Sally, the Midwestern derecho, California wildfires, and other smaller events, as well as $11.9 million for losses related to the COVID-19 pandemic. The COVID-19 losses we recorded in the quarter were small, reflecting additional information that became available during the quarter and represent our current assessment and best estimate of the ultimate losses for occurrences through September 30, based on policy terms and conditions, including limits, sublimits and deductibles. As of September 30, the vast majority of our COVID-19 claims are yet to be settled or paid with close to 80% of the inception to date incurred loss amount recorded as incurred, but not reported, i.e. IVNR reserves or as additional case reserves within our insurance and reinsurance segments. As regards to the potential impact of COVID-19 on our mortgage segment, we note that the delinquency rate at the end of the quarter was .69% down from .14% at June 30. Our current expectation is that the delinquency rate should be in the 5 to .5% range at year end 2020. While we have seen many positive signs over the last few months that point us to a more favorable view of the ultimate performance of the USMI book, many of the uncertainties we identified on our last call remain. In particular, the potential impact from a second wave of infections, potential lockdowns and the lack of an additional fiscal stimulus package or risk factors that we continue to monitor and evaluate on an ongoing basis. For these reasons, and consistent with our corporate reserving philosophy, we believe it is prudent to take a cautious approach in setting loss reserves across our MI book. In the insurance segment, net written premium grew .1% over the same quarter one year ago, a strong result demonstrating our ability to achieve profitable growth in this environment. Adjusting for the net written premium decrease observed in our travel, accident and health unit, the -over-year growth in net written premium would have been 26.5%. The insurance segment's accident quarter combined ratio, excluding CAATS, was 94.1%, lower by 620 basis points from the same period one year ago. Approximately 300 basis points of the difference is due to a lower expense ratio, primarily from the growth in the premium base from one year ago and reduced levels of travel and entertainment expenses this quarter. The lower XCAAT accident quarter loss ratio reflects mixed change and the benefits of rate increases achieved over the last 12 months. Prior period net loss reserve development, net of related adjustments was favorable at 1.1 million, generally consistent with the level recorded in the third quarter of 2019. As for our reinsurance operations, we had strong growth of .4% in net written premiums on a -over-year basis, which was observed across most of our lines and includes a combination of new business opportunities, rate increases, and the integration of the Barbican reinsurance business. The segment's accident quarter combined ratio, excluding CAATS, stood at 83.1%, compared to .8% on the same basis one year ago. The -over-year movement is primarily driven by a more normal level of large attritional losses compared to a year ago and rate change activity over the last 12 months. Most of the remaining difference is explained by operating expense ratio improvements, primarily resulting from the growth in earned premium. Favorable prior period net loss reserve development, net of related adjustments, was 40.8 million or 7.4 combined ratio points compared to 4.0 combined ratio points in the third quarter of 2019. The development was mostly in short tail lines. The mortgage industry had a record-breaking quarter in terms of NIW, and we certainly followed suit with this quarter's NIW of 32.8 billion, a full 30% higher than our prior high watermark. Offsetting this record-level production was the high level of refinancing activity across our portfolio, with the net result being a slight reduction in our insurance enforce. The combined ratio was 64.2%, reflecting the lower delinquency rate observed during the quarter. The trends we saw this quarter were favorable relative to last quarter, but the game is far from over. The expense ratio was slightly lower over the same quarter one year ago, and prior period net loss reserve development was favorable at 4.5 million this quarter. Total investment return for the quarter was positive 230 basis points on the US dollar basis, as the strong recovery in the capital markets produced healthy returns across our entire portfolio. Returns in our equity and alternative investments contributed approximately 40% of the total return for the quarter. The duration of our investment portfolio remained basically unchanged from the prior quarter at 3.21 years. The effective tax rate on pre-tax operating income was .8% in the quarter, reflecting a change in full year estimated tax rate, the geographic mix of our pre-tax income, and a 10 basis point benefit from discrete tax items in the quarter. As always, the effective tax rate could vary depending on the level and location of income or loss, and varying tax rates in each jurisdiction. We currently estimate the full year tax rate to be in the 8 to 10% range for 2020. Turning briefly to risk management, our natural cap PML on a net basis increased to 918 million as of October 1, which at approximately .4% of tangible common equity remains well below our internal limits at the single event 1 and 250 year return level. The growth in the PML this quarter is attributable to our ENS property unit within the insurance segment, which increased its writings in an improving marketplace. On the capital front, the increased in interest expense this quarter was mainly the result of the issuance of the $1 billion senior notes we issued in June 2020. So far, we have been able to fund our recent growth with our existing capital base, and our balance sheet remains strong with a debt plus preferred leverage ratio of .1% that remains well within a reasonable range. As for our USMI operations, the mortgage insurance linked notes market has recovered to a great extent from the lows we saw at the onset of the pandemic. Earlier this week, we priced our third Bellamy transaction of the year at terms that are getting closer to what we saw in 2019, both in structure and price. Our latest transaction will provide .5% of coverage in excess of a .5% attachment point, both expressed as a percentage of the risk in force. Including this transaction, the Bellamy structure is currently provide approximately $3.9 billion of aggregate re-insurance coverage. The fact that this market has recovered as extensively as it has in just over seven months, with investors more and more comfortable with the exposure they are assuming, is quite telling, and provides support to our current assessment of the health of the US housing market. With these introductory comments, we are now prepared to take your questions.

Thank you. If you have a question at this time, please press the star, then the one key, on your touchtone telephone. If your question has been answered or you wish to remove yourself from the queue, please press the pound key. If you're using a speakerphone, please lift the handset. Our first question comes from Elise Greenspan with Wells Fargo. Your line is now open. Hi, thank you.

Good morning.

Good morning.

My first question is on the capital allocation. So Mark, you laid out five pillars. I just want to confirm, I guess, based off of how you were talking, it sounds like scary purchase is last kind of on the chain right now. And I guess because it seems like you have such good growth opportunities that you could put your capital to use just basically to incrementally add to your insurance and re-insurance writing. Am I understanding that correct?

So, yeah, I didn't mean to put them in ranking order. I think that they are probably all fairly equally attractive at this point in time. I think that we're investigating, as you know, on a quarterly basis as to what the opportunities are in various kinds of business. I didn't mean to name stock repurchase as the last one as a rank ordering mechanism. I think that it's also part and parcel of the discussion. I think that for the first time in a while, I would argue that the five players actually are actively making the point for receiving the ball and play it, be part of the game. So I would definitely say that. And we're evaluating, but certainly our game number one, our focus number one is to allocate capital to the unrunning units if the returns are there. And that certainly is a relatively easier place to deploy and easier way to see it at this point in time. But everything is up, everybody's playing on the field. And the court.

Okay, that's helpful. And then my second question, your insurance underlying margin, 94% in the quarter. Arch had targeted to get to kind of a mid 90s. Yes, you know, you guys are kind of there, but there's more, you mentioned a lot of rate, I think, 11% that you're getting in your book of business, which means you haven't even earned that in yet. So as we think about earning in of the rate you're getting today, plus what seems like incremental rate, you could get into 2021. Do you have a new target for that business? Or is there a way that we can think about the margin profile there, especially since you kind of hit that target that you laid out for the street?

Yeah, so let me go back to the 95 combined, which I mentioned, I think three years ago now, I think it was meant to be there as an aspirational, you know, target and to to to shoot for at that point in time based on the mix of business and the opportunities that we had in the marketplace. I think this has changed, right? I think that now that we're obviously going in that direction, and very nicely in the market, certainly helping us, I think we still look very heavily into line by line, and are we byline the business, and I would argue, Ali, that sometimes the business would need less than a 95 combined, and some might be still okay at above 95. But certainly what I would tell you is a combined ratio was aspirational. We have to keep in mind that there's still, you know, COVID-19 is still ongoing. And secondly, interest rates have also been decreased, have decreased significantly since three years ago. So rate increases might be needed beyond a combined ratio just to make the returns equivalent to what they should be or would have been at a 95 historically. So it's really an ongoing process of reflecting current investment yield. So there are no targets on the combined ratio, but certainly targets on the return basis, and I would argue with with the 140 to 150 BIPs decrease in interest rate that the combined ratio would, you know, presumably mathematically need to go down to make an equivalent return.

Okay, that's helpful. And my last question, you mentioned taking a cautious approach to reserving within your mortgage book, and then I think you also laid out that the default rate gets to five to five and a half by the end of the year. And so I'm just trying to understand, like, as the default rate, I guess, I guess your expectation is it'll go up a little bit from where it sits today. And I'm assuming that you'll continue with the same conservative approach that we saw in the third quarter. So how should we think about kind of the combined ratio? You've been giving some metrics for how that business could trend, and you've obviously come in better than expected in the second and the third quarter. Do you have an expectation for how that could ultimately trend in the Q4, and I'm not sure if you want to provide initial color in 2021?

Yeah, I'll take that, Elise. I mean, just to be pretty clear, I think I've, as you guys all know, I, you know, I'll take the blame. I did a pretty poor job of forecasting combined ratios or delinquency rates over the last two calls. So I think we're trying to minimize those kind of forecasts. But listen, there's, as we know, there's a lot of uncertainty out there. Yes, we're extremely pleased that the delinquency rates have come down. Right? We had talked about, even just last quarter, we had said, like, somewhere around 8% by the end of the year. It doesn't look like we're going to, it's going to be as bad as that, based on what we know today. Still, you know, a few months to go, but, you know, definitely saw some encouraging signs this quarter. So that's all well and good. Does it translate to, you know, what kind of combined ratio does it translate to? I mean, it's hard to know, because again, what we're facing is really, we just don't know how long this pandemic is going to last. And, you know, the fact that the forbearance programs are, at this point, scheduled to end, but who knows if they get extended or not, how do people convert from, you know, forbearance to a regular delinquency? There's a lot of unknowns at this point that we feel are extremely hard to predict and estimate. So, listen, we're taking it one quarter of time. We're happy with where things are at right now. Again, it's reassuring, but we're not, as I said, the game's far from over. And who knows, maybe hopefully even my 5 to .5% forecast or expectation for delinquency ends up being a bit high, but we'll find out in a few months and we'll reassess at that point.

Okay. Thank you very much.

I appreciate all the color.

Sure.

Our next question comes from Jimmy Buller with JPMorgan.

Good morning. First, I just had a question on how you're thinking about pricing, reinsurance pricing as a buyer of reinsurance and how should we think about your sort of overall exposure, especially to cats, as you're entering 2021? Are you given better pricing, are you going to hold more exposure or maybe should we assume that you'll keep your retention sort of similar and be a buyer of record regardless of prices?

The short answer is we don't know yet, right? I think we'll have to see what the one-one brings to our reinsurance folks on the inward side and certainly on our ENS property. We'll have to evaluate and assess what kind of exposure and what kind of margins we're getting there. And then look back and say, okay, now what is the, because buying reinsurance, as you know, is like raising, it's like using capital. It's like we have to pay for that, so we're going to go through a very straightforward analysis of capital usage and what we pay for it. And we take a very, very, you know, first economic view of the whole world and specifically as it relates to the insurance exposure, we take on a property cat exposure specifically. But I think that all this, we also have, you know, balancing it or adding to that information process is that we also are careful and not over-stretching the capital. So we'll always be buying reinsurance to some extent. The question is, you know, what level and how much and at what price. But clearly we like stability and we're always trading stability for sometimes lesser margin and that's going to be part of the mix. But clearly both markets, you know, where we can be on the advantage, we can gain both sides, but we can actually get rate increases on the insurance side, find a way to buy reinsurance in a good way. And on the assumed basis, we actually are benefiting from the improvement in the market there as well. So it's really a holistic view of the cat exposure. Too early to tell what exactly is going to transpire, but everything is always up for discussion.

And then on Watford, you raised the price. There's pressure on you to raise it again. And at what point does the deal become sort of uneconomic and or are you already there or any comments on how you're thinking about that situation?

Well, we didn't raise the price. We did not raise the price, right? We have an agreement. We have a signed agreement with Watford that, you know, we're starting the process to get regulatory approvals, et cetera. But yeah, I mean, there's another party that has come in that Watford feels they have to look into their potential offer and what it all means to them. But at this point, you know, what's been announced is what is still valid. Depending on, you know, what they end up deciding, we may choose to do something different. But on this point, we can't really say much more than that.

Okay, thank you.

Yep, sure.

Our next question comes from Mike Zaremski with Credit Suisse.

Okay, thanks. Good afternoon. Focusing on the reinsurance segment, you know, robust growth, thanks for the commentary, bullish commentary. If we look at the underlying accident year loss ratio, I mean, actually, expense ratio too. So just a lot of improvement. Anything we should be thinking about? Anything you want to call out? Or is, you know, a lot of this just market conditions and operating leverage that you're benefiting from?

Yeah, so the reinsurance group has grown tremendously. So the expense, you know, expense ratio going down or, you know, what it did certainly is explained by that in large part. I think in terms of, you know, the loss ratio, what I would like, and we said that on prior calls as well, is we tend to look at reinsurance performance on a 12 to 18 month basis. There's a lot more volatility in that segment and that sector. There's also a lot of shift in the mix over time. Our reinsurance folks do not tend to have a, you know, they have stable relationships obviously, but there's a lot of things moving on in terms of taking season, the opportunities in the marketplace. So what I would look at, the loss ratio is more like, you know, this is some quarters is much better than others and it's really due to volatility in the reinsurance results, I would say.

Okay, okay, got it. So the expense ratio, some of that's going to grow. Okay. Follow up on the mortgage side, I think I heard you say earlier that ILN pricing kind of had improved a lot. I guess I'm looking at, we've been trying to track overall ILN pricing and we kind of still see it as double what it was pre-pandemic, but maybe I'm just looking, you know, maybe Arch's pricing is better or maybe I'm just incorrect.

Yeah, I mean there's two parts to it, right? There's the structure meaning, in particular, attachment points and then the pricing. So, right, so this is our third one, our first one of the year. Really, we were the first ones out of the gate after the pandemic. The reality at that point is, you know, appetite from the investors was not as good as what it, you know, as it was in the past and our attachment point was much higher. Second one, attachment point came down with slightly better pricing and with this new, latest one, we're effectively back to the same, you know, the two and a half percent attachment point that we had seen pre-COVID. So that's certainly a very good sign. Pricing, you know, you risk adjust for everything, it's still up, no question. It's not to the same level that it was pre-COVID, but it's, you know, it's not double for sure. So it's, you know, much, you know, much better than that, much lower than that. And, you know, quality to book is better. So how does that get factored in? And so there's a couple other things that it's hard to make it perfectly apples to apples, but directionally on an absolute basis as well. We're very happy with where things are going.

Okay. I got it. That's helpful because I think that you're describing it. It's better than what we have thought.

Okay.

And lastly, sticking on mortgage, you know, hopefully, you know, you said you, you know, the delinquency rate doesn't tick up as much as you thought, but are you, it's almost November now. Are you, have you seen over the last few weeks delinquency rate tick up?

So last few weeks, it's been, it's been actually keeping in the same general direction that we saw in the third quarter. So it hasn't picked up, flatish, I'd call it. And, you know, we got a couple of months to go. We'll see how things play out, but that's kind of

what we're seeing. Just thanks for the color. Yep,

sure.

Thank you. Our next question comes from Josh Shanker with Bank of America. Yeah,

thank you. Two questions. One, just to understand accounting and the other one is about Watford. On the accounting, you had a decline in delinquencies due to cures of 6,000 approximately, but you took up reserves. You know, I know you can't really reserve for a loss that hasn't happened yet, but it looks like you're reserving for these new claims at about 12, $13,000 per claim. Compared to a historical average of $4,000 to $5,000 per claim. Am I doing the math correctly? And can you explain sort of how you think about that in the books? I think in the first quarter, you also took up reserves more than typical because you can only take them up when you have claims. But can you walk through that a little bit?

Yeah, no question that we bumped up reserves this quarter on effectively the delinquencies we saw in the second quarter. So your math is correct. I mean, the reserves per delinquency, you did the math right, but what you need to adjust for, I would say, is call it a reserve strengthening exercise that we went through just to reassess where we were. You know, took a hard every quarter, we take a hard look and our view is that might as well not, you know, be cautious knowing what we know and knowing what we don't know. So think of about a $45 million adjustment on reserves in the third quarter for effectively Q2 delinquencies. So once you adjust for that, I think you get back to claim levels or severity that you would be more in line with what maybe would have expected.

So along those lines, is there a reason to believe that the severity of the losses are going to be different than historical severities? I can understand because there's a lot of frequencies and you can't really take frequencies until you get a claim, but is there a reason to be more cautious around severity in this pandemic?

We don't see it with the only adjustment, obviously, that what we're reporting in our supplement in terms of paid severities is lower than what we're seeing from the new delinquencies, right? New delinquencies, we mentioned it last quarter at about a $65,000 or so per NOD. This quarter, it's right around $60,000. So I mean, it's certainly higher because it's more recent loans that are going delinquent versus what we the loans we paid on in the quarter. So that's the only adjustment. But in terms of percentage of the insured value or the insurance enforce or the risk enforce, we don't at this point don't have a reason to think that it's going to be materially different than what the weight's been in the past.

Okay. And then on Watford, we don't know how it's going to turn out, but your own stock trades around book value, the offers for Watford around 0.8 times book. I'm not sure. And maybe you've some thoughts on whether Watford is a better investment at 0.8 times book than Arches at one time book. But if you don't buy in Watford, it would suggest that you have a chunk of excess capital that you were already allocating towards financial uses. Can we expect that your interest in buying business you already know is attractive even the market opportunities here?

I think we made the point clear by putting our offer up there. And that's pretty much what we'll leave it as Josh. I think we're we think still we still think that Watford is a good platform, valuable platform.

But clearly Arches is a more valuable platform than Watford. If you're willing to buy in Watford stock, should you be willing to buy in Arches stock as well?

I think the answer is always yes. We're always looking at the possibility of buying our stock. And certainly, like I said before, on the five players on the court, I think that the share repurchase is really attractive, as you would expect me to say the CEO of the company.

All right. We'll keep watching. Thank you.

Thank you. Thank you.

Our next question comes from Yaron Kinnar with Goldman Sachs.

Hi. Good morning. Thanks for taking my questions. I do want to start by thanking you for giving basketball analogy so I can actually understand what's going on. It was a word you'd give a hockey analogy. My first question goes to the slight increase in COVID losses in the quarter. Can you maybe talk about what drove that specifically? I guess specifically what I want to get at is does it have anything to do with the FCA court case over in the UK and how you're thinking about your overseas business interruption exposures?

Yeah. I mean, it's really in two parts. I think on the insurance side, it's all basically related to our travel book in the US, so no connection to the FCA ruling. And a little bit we added in the reinsurance side is around property exposures, mostly out of Europe. So it's a little bit of a BI angle to it, but that's pretty much it. So I mean, I call it a bit of just new information coming in. No material new information that came through that would have caused us to revisit our picks. And the FCA, as we said, we didn't change it. And the where it stands, we're still very much, we had fully reserved for it. And again, the ruling doesn't change our position on that.

Okay. The rest of the portfolio, yeah, for what it's worth. And we mentioned that on prior calls is that we have the vast majority, almost the totality of our policies have the exclusions that would protect us somewhat from a deviation. So this is nothing in there to really think about.

Okay. That's very helpful. And then my second question, when you talked about the improvement in the accident or loss ratio for insurance, you didn't talk about feeble frequency. So is it because you didn't see that or is it because you didn't book that?

It's a little bit of both, actually, we saw some of it, some improvement on the frequency of claims possibly in the second quarter, that was definitely a difference. But if you neutralize for the obvious lines of business where claims would actually naturally, go down such as travel, for instance, with a lot less exposure or go up, then if you neutralize all this, we didn't see a discernible change through the nine months of this year. It's hard to see what the natural rate would be, but we don't see a discernible change in the frequency to speak of. And really we put our reserve pick and our loss pick based on long-term averages. We're not looking at specific one quarter at a time and one even accident year at a time. We do get in the mix of multiple years and project out much more on a longer term expected. So it will take a little while before we would recognize any significant improvement in frequency of claims, if at all.

Got it. And then maybe one final very quick one on Watchford. If the deal does go through, I just want to confirm that the company wouldn't lose the fee income if Watchford's consolidated, right?

Correct.

Okay. Thank you.

Yep.

Sure. Our next question comes from Ryan Tunis with Autonomous Research.

Hey, thanks. Good morning. The one question I had was just on NII and thinking about low rate headwinds. It looks like you guys have almost 30% of your assets in essentially government bonds. It looks like your portfolio yield is sub 2. We just heard another Bermuda competitor today say that their new money yield is 2%. It would seem to me, given how you're allocated and given your current portfolio yield, that there actually wouldn't be that much incremental yield headwind moving forward actually. Arguably, if you wanted to move out of some of the gubbies and take some credit risk, it actually seemed that maybe you could expand your portfolio yield. It doesn't sound like that's right, given your commentary, but I'm just kind of trying to understand why is ARCH continuing to be incrementally exposed to this low rate environment, given the fact that you guys have already really taken on the chin in terms of where your yields at now?

Yeah, I mean, no question we've been defensive on credit. So, yeah, we have a bit more on the Treasury side that you're right to mean. Unless rates keep going down, but they seem to stabilize, I guess, for the time being. Our play in the last few quarters has been trying to, as many others, I'm sure have done the same, is try to find other opportunities that are a bit more in the eye of alternative space that provide us still some level of investment income without taking on too much risk. So, it's something we keep looking at, but it's a bit of a challenge. There's not a ton of opportunities out there. We're not worried a little bit. We're defensive on credit still, and that's something that we look at carefully. We don't want to overshore ourselves there either.

Do

you guys have a new money investment yield you'd call out handy that you did in the third quarter?

Well, the last call in the last month, what we've been putting our money to work is just about 2%. So, that's kind of where the latest information that we get from our investment guys.

Got it. Thanks for the answer. Thank you.

Sure. Our next question comes from Jeff Dunn with Dowling and Partners.

Thank you. Good morning. Morning. First question, and I appreciate the added color around the EMI reserving. If we make the adjustment, is the math correct that you change your incidence assumption up to about 9%?

Close enough. 8% to 9%, yes. And our view on that is that we feel like the more recent NODs may be prone to a bit more stress. So, the fact that these people took a bit longer to go into delinquency may tell us that they're maybe, again, subject to more stress, but again, time will tell.

And is that 9%? Is that just an overall number or is it a blend of, just as an example, 6% on forbearance and 13% on non-forbearance notices? Do you delineate between the two or is it just more of a holistic approach?

It's more of an aggregate approach that we take. We don't separate the two as cleanly as you're suggesting. We do look at all of this. Jeff, I want to

make sure you understand that we also look at prior events, prior CAD events, prior programs of the sort. But as you could appreciate, Jeff, it's probably more art than science at this point in time.

With the new ILN, a little confusing in the documentation. It looks like it applies for policies after Jan 19, but it looks primarily more recent loans over the last three, four months. What is it designed to do? Is it really the last quarter or so? Or is it part of the design to come underneath the 20-1?

The latest, the 20-3, is very much, the vast majority is covering June, July, and August. That's three months of production. There's a little bit of spillage of the 20-2 and also some 2019 loans that had been late to being processed, whatever. It's just a little bit of a catch-up on a few small things, but think of it as really June, July, August production.

All right, great.

Thank you. Yep.

Sure. Our next question comes from Meier-Shields with KBW.

Thanks. Mark, I'm a little confused by something that you said. You were talking about the five capital deployment opportunities. You cited the investment portfolio. I guess I would have thought that if you're going to be writing more P&C or mortgage insurance, then the supporting capital will be in the investment portfolio anyway.

Yeah. You're saying that we should just repeat the question, Meier, please.

I'm trying to figure out how capital deployment in the investment portfolio is distinct from capital deployment into either P&C or mortgage.

Okay. Now I see what you mean. The way we look at the underwriting returns is we attribute to the units the treasury return or the risk-free return. We try and separate church and state, what we call it internally. We actually credit the float on the insurance on the underwriting pieces at the treasury rate level. Then we allocate the capital for the access over that to the investment portfolio. This is how we separate the capital usage between the underwriting units and the investment units.

Okay.

That makes

sense. Got it. I'm trying to get a general sense of how you see either profitability in property catastrophe in retro or your view as to whether the vendor catastrophe models are conservative enough based on recent years' CAD activity and the underlying trends.

Yes. A good question, Meier. I think that our position on this, we have weather scientists upstairs and we have a lot of lengthy experience and information to look at. Listen, we actually augment and modify the vendor models as we see fit. The vendor model represents to us a solid starting point from where we can definitely have a lot of science into this one. We've been historically using them as a starting point and augmenting it and modifying it with our own view of the world. I would caution everyone by saying that, yeah, we do have a view of ferris perils and various exposure around the world. We also try to factor in shorter term versus longer term, possibly modification of what could happen out there. It's hard to predict those things. What we tend to do is to hold ourselves to a higher return for those risks that are inherent in the way we underwrite the business. That also helps explain why we've taken possibly somewhat of a more conservative approach to property CAD exposure over the last four or five years as you've seen in our numbers. It's a year on year analysis, but we will look at it and we'll evaluate. I think in some of our discussion, we have El Nino, all these various discussions happening all the time. We're not the largest writer of property CAD, but we do have a very solid team looking into this. I think we try to make it up with holding ourselves up to a higher level of returns. Okay, that's very helpful. Thank you so much. Sure.

Our next question comes from Phil Stefano with Deutsche Bank.

Phil, good morning. Most have been asked and answered. One follow up on the MI piece and thinking about the increase in the default rate. I guess in my mind, there are three mechanisms to get there. It's either new defaults are increasing, cures are slowing down, or the denominator policies enforce the shrinking. I'm just curious how you're contemplating which of levers is contributing to the increase in the default rate. I think it helps us think about does this come through as losses or premiums or maybe development if the cures are right.

Phil, let me try and take this. It's a big question, a lot of parts. First and foremost, the denominator is not changing a whole lot. It's about a billion, a million, 250. That's what it is in terms of policy. The DQ rate has decreased this quarter. It went from 5.1 to I believe 4.7 this quarter. This is what the number of policies or the numerator that are in default and you divide by a similar number. It did go down. I think what we're saying in terms of reserving, you reserve for the new notices that you've received in the quarter. We received 58,000 in the second quarter. We received 20,000 this quarter. What we're telling you is we take the book of these 20,000 new notices, the cures are the cure. They go away from the overall balance, if you will, that we just mentioned, which is the 4.7. The new notices, we reserve for those in the quarter. We have to reserve in the quarter on those that were newly notified. This is what you would do on a reserving. What we did in addition to this is we modified our view of what we should have booked for the second quarter for the new notices in that quarter. I'm trying to help give you a bigger picture, hopefully that helps you understand how to reserve.

That makes sense from the reserving side. The default rate is loans in default, that population, over policies in force. If the policies in force isn't changing and the default rate is going up, then presumably there's an expectation that new defaults is going to increase this quarter. I just wanted to make sure that I was thinking about it correctly because to get from 4.7 to 5.5,

there

must be an acceleration in new default or cures are slowing down. I don't know.

That's a big top-down view. It's hard to know. We can't predict. You don't have the crystal ball on whether cures are going to accelerate or not or slow down. Same thing with new notices. Again, we're trying to provide a bit of color that we think obviously, or knock on wood, we're not going to be at 8% delinquency rate by the end of the year. We're certainly well below 5%. Does it go above 5% by the end of the year? That's what we think could happen, but again, we don't know. If everything stays the same, the odds are probably that it won't be at 5%.

You're right on your assessment that if we go from 4.7 and we tell you it's going to be 5.5 at the end, that we'll have to have an increase in NODs or lesser cure. I'm not sure we have made enough assessment as to what she's with. I would argue that we have tended to, since May, we have tended to, at least the last two quarters, we've overstated the number of new NODs that we received. That probably is what would drive the potential. We still don't know. Francois said that we haven't been as good as we would have liked to be in predicting. I think the new NODs would be a more modern place to look at. Having said this, Phil, you also know that we had two things going on, less new NODs and higher cure rate. They both actually helped to manage the amount and they helped accelerate and actually help explain why we had a projection of about 10% and nearer in the year to much less right now. They're two moving parts.

Yes, understood. I guess in my mind that default rate might continue to come down. As I said, the crystal ball is very cloudy at this point. Thank you, guys.

Thank you. Our next question comes from Brian Meredith with UBS.

Thanks. A couple of them for you. Just quickly, Francois, I'm just curious, given that you're getting 2% new money yields and your book yield is below that, should we just expect that your yield on your investment portfolio fixed income to be stable here going forward?

In the short term, I think it's reasonable.

That's helpful. Second question, on the reinsurance business, I'm just curious. Your property mix is definitely increasing as a percentage of your overall book. How much of the underlying combined ratio improvement that you're seeing there is actually driven by that mix shift versus just better loss picks?

Well, then, boy, I don't think we parse out this way until early in the game to see what it's going to look like. I would say that it's probably more mixed at this point down because our growth is largely in property other than the CAD Excel. We have a lot of play in that sector right now. It's probably more mixed, I would say, Brian. Yes.

It's not so much the rate activity that we're seeing, but it's the mix shift that's driving it.

The way we are, as Brian, for all these years is we'll tend to go where the rates are the better ones and we have a better increase in margin. That sort of goes hand in hand.

Gotcha. That makes sense. Another one, just quickly on Watford, I'm just curious, is it your guy's intentions to significantly increase your ownership in that ultimately, or are you going to buy the whole thing in just because of the volatility of the investment portfolio? It's just a different strategy?

Well, what we said publicly is that we are certainly talking to other parties to bring into the fold to support us in this vehicle if it moves forward, if we close on the acquisition. The answer is we would most likely increase our participation, but not to 100% and be much less than that. So increase a little bit, but recognize that it's a different model that others would want to participate on with us as well.

Gotcha. So the net capital outlay for this transaction wouldn't be as big as it could be. Okay. That makes sense. Correct. Gotcha. Then last question, I'm just curious, any updates with respect to the Coflis investment? Where that stands, are regulators going to ultimately make a decision on this?

Yeah, I think it's in process. We've gone through the regular process. It's on track. We're hopeful that we could get something done potentially in the first quarter or second quarter of next year. It takes a while, as you know, in this COVID environment. Government and regulators take a little while longer than otherwise, but yeah, it's on track. They produced, as you know, results, I believe, earlier this week or late last week. So it's also looking much better for them, which is good for us.

Great. Thank you.

Thanks, Mike.

Thank you. Our next question comes from Jimmy Buller with JPMorgan.

I just wanted to clarify your comment on not losing the fee income on Wartford. Is that just because of the five years remaining on the contract? So there's sort of a finite life to it, or was there something else that was behind that comment?

Well, I mean, the question was asked, if we buy the vehicle, do we still retain the fees? The answer is yes. I mean, I think...

Yeah, that's it. It's an internal allocation then, but in case it goes to somebody else, then that fee arrangement stays intact and through the length of the six-week contract, which is five years. Yes,

correct.

Okay. And then does your approach to underwriting overall and how much sort of capacity you have and take on on a given case change if you own more of Wartford versus maybe if it does end up being bought by a third party at a higher offer?

I think, you know, in a third party environment, Jimmy, I think it's pretty clear that we need to take care of our brethren as well as we could take care of ourselves. So our view is always to do the similar underwriting. Wartford had a different investment profile, which allowed us to do slightly different things, but at a high level, the underwriting is very, very similar. And then I would remind everyone that whatever we do on Wartford, we take 15% of it on a quarter share basis in the back. We also are participating on the capital. We have like, we're up to about 13%, I believe, in our shareholding. So we collectively own 20% of the of the underwriting. So we do eat our own cooking there as well. And it's really important to us.

Okay.

I'm not showing any further questions. I would now like to turn the conference over to Mr. Mark Grandison for closing remarks.

Thank you, everyone. Looking forward to the last remaining couple of months in the year, and I hope you have a good one.

Ladies and gentlemen, thank you for participating in today's conference. This concludes the program. You may all disconnect.