Constantine P. Iordanou - Chairman, Chief Executive Officer, President, Member of Executive Committee, Member of Finance & Investment Committee and President of Arch Capital Group (US) Inc Mark D. Lyons - Chief Financial Officer, Principal Accounting Officer, Executive Vice President, Chief Risk Officer and Treasurer.

Kai Pan - Morgan Stanley, Research Division Michael Steven Nannizzi - Goldman Sachs Group Inc., Research Division Vinay Misquith - Evercore Partners Inc., Research Division Jay Gelb - Barclays Capital, Research Division Jay Adam Cohen - BofA Merrill Lynch, Research Division Ryan J. Byrnes - Janney Montgomery Scott LLC, Research Division Ryan J.

Byrnes - Langen McAlenney Joshua D. Shanker - Deutsche Bank AG, Research Division Ian Gutterman - Balyasny Asset Management L.P. Meyer Shields - Keefe, Bruyette, & Woods, Inc., Research Division.

Good day, ladies and gentlemen, and welcome to the Third Quarter 2014 Arch Capital Group Earnings Conference Call.

Before the company gets started with this 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 8-K furnished to the SEC yesterday, which contains the company's earnings press release and is available on the company's website.

[Operator Instructions] As a reminder, this conference is being recorded for replay purposes. I'll now turn the conference over to your host for today, to Dinos Iordanou and Mark Lyons. Gentlemen, you may begin.

programs, travel and accident, E&S binding authority and construction and national accounts business. Most of our organic growth is coming from small accounts with low limits, which should have lower volatility.

On the other hand, competitive conditions in the property sector have negatively affected primary property rates and accordingly, our premium volume in those lines.

In the primary markets, in which our Insurance Group participates, we continue to obtain rate increases in most lines of business, albeit slightly below the levels that we observed last quarter. Mark will give you more color on this later on.

On the reinsurance side of the business, we have seen a continuation of softening in terms and conditions that we noted in prior quarters.

From a production point of view, net written premium was down 16% in the quarter for the reinsurance group, primarily as a result of a nonrecurring large NM [ph] premium portfolio, which, as we described in the prior call, was booked in the third quarter of 2013, and is not recurring in 2014.

Gross premiums rose in the reinsurance segment by nearly 5% with the growth in the segment arising from business we produced for Watford Re.

Adjusting for the reduction of the NM [ph] premium portfolio from last year and the business produced for Watford Re, reinsurance premiums would have declined by approximately 6% in the quarter if you compare like quarter-to-quarter Our mortgage segment includes primary mortgage insurance written through Arch MI in the United States and reinsurance treaties covering mortgage risk, which is written globally, including the U.S., as well as other risk-sharing and structure mortgage business.

As you may recall, our mortgage insurance business in the U.S. serves 2 major markets, credit unions and banks and other mortgage lenders. Net written premium for the third quarter of 2014 was approximately $58 million, of which $32 million was written by Arch MI U.S. who tenders to the credit unions, banks and other mortgage lenders.

As we discussed in prior calls, Arch MI has a dominant position in the credit union sector and is in the process of building out its client base with bank lenders. This past quarter, a substantial amount of the growth came from single paid premiums. These are loans where the mortgage insurance premium is paid up front.

As of today, we have approved more than 400 master policy applications from banks, and more than 100 of these banks have submitted loans for our approval. Of these approvals, 29 represent national accounts and the balance are regional banks. We define national accounts as the top 100 producers of mortgages.

Of the top 25 originators, we have approvals from 15 of them. Group wide, on an expected basis, we believe the ROE on the business we underwrote this year will produce an underwriting year ROE in the range of 10% to 12%.

As on a present value basis, improvements in the Insurance Group and the addition of the mortgage segment were more than offset by lower expected returns in the reinsurance segment. Before I turn it over to Mark, I would like to discuss our PMLs.

As usual, I would like to point out that our cap PML aggregates reflect business bound through October 1, while the premium numbers included in our financial statements are through September 30, and that the PMLs are reflected net of more reinsurance and retrocessions.

As of October 1, 2014, our largest 250-year PMLs for a single event decreased slightly in the Northeast to $658 million, or 11% of common shareholders' equity, while Gulf PMLs also decreased slightly to $608 million. Our Florida Tri-County PML now stands at $451 million.

I will now turn it over to Mark to comment further on our financial results and as customary, after Mark concludes his remarks, we will take your questions.

Mark?.

Great. Thank you, Dinos, and good morning. Before we get started, I'd like to just say a few words to define how I'm going to speak about things for the balance of the call, so just bear with me on this.

First, as a reminder, the financial supplement shows consolidated financial statements that include our other segments, which is Watford Re, as well as providing other financials that excludes the other segment, and those are footnoted as such on each page of the supplement.

Furthermore, within the segment information of the supplement, we have provided a subtotal of Arch's core segments, that is insurance, reinsurance and mortgage without Watford Re, and have also separately provided Watford Re's results posted alongside in order to arrive at an Arch consolidated segments income statement view.

The investment information section of the supplement, however, is completely shown excluding the other segment. So my comments to follow today are on a pure Arch basis, which excludes the other segment unless otherwise noted.

For further clarity, although today's discussion will exclude Watford result as a segment, however, when Watford is used as a reinsurer by any one of our core segments, those sessions are reflected in Arch's underwriting results.

Furthermore, since the definition of the word consolidated includes the results of Watford, I will not be using that term, the term consolidated. Instead, just be aware that my comments refer to our core segments of business, insurance, reinsurance and mortgage combined as just discussed.

This permits an apples-to-apples comparison with Arch's results a year ago. Okay.

Moving on with that defined, the combined ratio for this quarter was 88.5% with 1.6 points of current accident year cat-related events net of reinsurance and reinstatement premiums compared to the 2013 third quarter combined ratio of 86.1%, which reflected 2.5 points of cat-related events.

Losses from 2014 catastrophic events net of reinsurance recoverables and reinstatement premiums totaled $14.2 million, primarily emanating from European storm Ella and Midwestern U.S. tornado activity.

The 2014 third quarter combined ratio also reflected 8 points even of prior year net favorable development, net of reinsurance and related acquisition expenses compared to 8.2 points of prior period favorable development on the same basis in the 2013 third quarter.

This results in a 94.9% current accident year combined ratio, excluding cats, for the third quarter of 2014 compared to 91.8% accident quarter combined ratio in the third quarter of 2013. In the insurance segment, the 2014 accident quarter combined ratio, excluding cats, was 98% even compared to an accident quarter combined ratio of 97.1% a year ago.

This increase was primarily attributable to certain larger attritional losses emanating from our European operations and in particular, mostly from the aviation war line of business.

These large attritional losses account for 170 basis points more of net earned premiums in the third quarter of 2014, than a large attritional loss in the corresponding quarter of 2013.

The reinsurance segment 2014 accident quarter combined ratio, excluding cats, was 90.6% compared to 94.8% in the 2013 third quarter but represents a sequential improvement from the 92.1% combined ratio last quarter.

As noted in prior quarters, the reinsurance segments results reflect changes in the mix of premiums earned, including a lower contributions from property catastrophe business.

The insurance segment accounts this quarter for roughly 16% of the total favorable development, excluding the associated impacts on the acquisition expenses, and this was primarily driven by shorter-tailed lines from the 2006 through 2012 accident years.

The reinsurance segment accounts for approximately 83% of the total favorable development in the 2014 third quarter, again, also excluding the associated impact on acquisition expenses.

With approximately 40% of that due to net favorable development on shorter-tailed lines concentrated in the more recent underwriting years, roughly 8% was attributable to medium-tailed lines based throughout many underwriting years and about 49% due to net favorable development of longer-tailed lines, primarily from the 2002 through 2010 underwriting years.

The mortgage segment accounted for approximately 1% of the total favorable development in the quarter, stemming mainly from our U.S. primary operations concentrated in 2009 and prior.

As has been consistent in the past, approximately 69% of our core $7.2 billion of total net reserves for losses and LAE, our IBNR or additional case reserves, which as I've mentioned has been consistent across time and across the main reinsurance and insurance segments.

The expense ratio for the third quarter of 2014 was 33.5% versus the prior year's comparative quarter expense ratio of 32.4%. The increase in the operating expense ratio reflects the addition of our U.S.

mortgage insurance operations, which is operating at a higher expense ratio until business hits a steady state as well as the effect of incremental expense due to certain platform expansions in both our reinsurance and insurance businesses.

The insurance segment improved to a 31.7% expense ratio compared to 33.1% a year ago, primarily reflecting a lower net acquisition ratio driven mostly by materially improved 3D ceding commissions.

The reinsurance segment expense ratio increased from 30.7% in the third quarter of 2013 to 32.6% this quarter, primarily due to a higher level of operating expenses and by higher ceding commissions incurred. On -- our U.S. insurance operations achieved a 2 point -- a positive 2.6% effective renewal rate increase this quarter net of reinsurance.

A key point that can get lost in the averages is how different the pricing environment is for short-tailed versus longer-tailed lines. Our short-tailed lines of business had an effective 3.5% renewal rate decrease for the quarter compared to a 4% effective renewal rate increase for the longer-tailed lines both on a net of reinsurance basis.

These longer-tailed line rate increases continued to be above our view of weighted loss cost trends.

Looking more deeply, some lines incurred rate reduction, such as nearly 8% in property and 4% in high capacity D&O and others enjoyed healthy increases such as plus 10% in our lower capacity D&O lines, 7.5% increase in contract binding and nearly 6.5% increase in excess workers' compensation.

Also certain lines continued their momentum of achieving strong cumulative increases. For example, our admitted loss sensitive businesses, as well as our lower capacity D&O lines have achieved 13 consecutive quarters of rate increases.

The private not-for-profit D&O unit has in fact, secured double-digit increases for 9 of those 13 consecutive quarters. The ratio of net premium to gross premium in the quarter was 75.5% versus 80.9% a year ago. The insurance segment had a 74.2% ratio compared to 73.5% a year earlier.

This quarter's Insurance segment net to gross became 76% when the impact of the Sparta renewal rights transaction discussion discussed on last quarter's call is removed. This continues to show the insurance segment's emphasis on rating smaller lesser volatile business and keeping more of it net as a result.

In the reinsurance segment, the net-to-gross ratio was 75.8% in the 2014 third quarter compared to 94.6% a year ago, primarily reflecting sessions to Watford Re as a reinsurer and other third-party retro purchases protecting the property book. The mortgage segment posted an 86.6% combined ratio for the quarter.

The expense ratio, as expected and as indicated earlier, continues to be high as the operating ratio relating to our U.S. primary operations will remain elevated until proper scale is achieved. The net written premium of $58.5 million in the quarter is driven, as Dinos noted, by $32 million from our U.S.

primary operations and $26.5 million of net written premium from our reinsurance mortgage operations, which includes the 100% quarter share of PMI's 2009, 2011 underwriting years as part of the acquisition of the CMG companies and the PMI platform. At September 30, 2014, our risk-in-force of $10.1 billion includes $5.5 billion from our U.S.

mortgage insurance operation, $4.5 billion through worldwide reinsurance operations and $136 million through our risk-sharing transactions. It's important to note that U.S. operations utilize policy-specific covered ratios to determine risk-in-force from insurance in-force figures.

As you may recall, insurance in-force represents the aggregate amount of the individual loans insured, while risk-in-force incorporates the insurance coverage percentage.

Outside the U.S., we followed market practice to estimate risk-in-force on a similar basis, while for risk-sharing transactions, risk-in-force reflects our percentage participations with inbound layers, as well as the impact of contract limits.

That is, risk-in-force on risk-sharing transactions does not exceed the contractual limits of liability involved. Our primary U.S. mortgage insurance operation down just shy of $2 billion of new insurance written during the quarter, which represent the aggregate of original principal balances of all loans receiving coverage during the quarter.

The weighted average FICO score for the U.S. primary portfolio remains strong at 733 and the weighted average loan-to-value ratio held steady at 93.4%. No states risk-in-force represents more than 10% of the portfolio and our U.S. primary mortgage insurance company is operating at an estimated 9.1:1 risk-to-capital ratio as of September 30, 2014.

The other segment i.e. Watford Re, reported a 100.2% combined ratio for the quarter on nearly $100 million of net written premium and $34.8 million of net earned premiums.

As a reminder, these premiums reflect 100% of the business assumed, rather than simply Arch's approximate 11% common share interest due to the variable interest entity accounting election.

Our joint venture Gulf Re produced the $7.8 million loss for the quarter due to an unusually high frequency of large technical risk losses stemming from the Middle East. This is reflected on the income statement within the other income line.

The total return on our investment portfolio, as Dinos noted, was a reported negative 51 bps in the 2014 third quarter, reflecting negative returns in our fixed income and equity sectors along with the impact of the strengthened U.S. dollar on our foreign denominated investments, while our alternative investment portfolio continued to perform well.

Excluding foreign exchange, total return was a positive 21 bps in the 2014 third quarter. On a trailing 12-month basis ending September 30, the total return has been a positive 3.30% and excluding foreign exchange, the return has been a positive 3.76%.

Our embedded pretax book yield before expenses was 2.21% as of September 30, 2014 compared to 2.175% at serially, at June 30, while the duration of the portfolio lengthened slightly to 3.28 years from last quarter's 3.14 years. Fixed income duration can fluctuate due to tactical investment decisions as opposed to long-term strategic shifts.

The current duration continues to reflect our conservative decision on interest rates in the current yield environment.

Reported net investment income in the 2014 third quarter was $72.2 million or $0.53 per share versus $72.5 million in the 2014, second quarter, which was also $0.53 per share and $66.1 million or $0.49 per share in the corresponding quarter in 2013.

As always, we evaluate investment performance on a total return basis and not merely by the geography of net investment income. Interest expense for the quarter was $4.2 million, which is a significant reduction from the last 2 quarters.

This reduction is due to a favorable adjustment involving a certain loss portfolio transfer entered into effective January 2013. This deposit accounting transaction had a downward reevaluation in the quarter of the underlying ultimate loss, which resulted in an $8.2 million reduction in interest expense.

The run rate interest expense, which includes approximately $12 million from Arch's senior notes and a variable amount of Arch's revolving credit borrowings and some other items is expected to be approximately $13 million per quarter for the foreseeable future.

However, reevaluation of the underlying ultimate loss attributable to this [Audio Gap] used to give us significant financial flexibility. We also continue to estimate having capital in excess of our targeted capital full position.

Book value per share, as Dinos noted, was $44.04, up 0.7% relative to June 30, up 10.6% relative to year end 2013, and up nearly 15% relative to 1 year ago at September 30, 2013. This change in book value per share this quarter primarily reflects the company's continued strong underwriting performance offset by unrealized losses on investments.

So with these introductory comments, we are now pleased to take your questions..

[Operator Instructions] That question will come from the line of Kai Pan with Morgan Stanley..

The first question is on the buyback. It looks like you saw buyback actually in a big way in more than a year.

I just wonder what changed your thought process on that? Is that less opportunity for organic or acquisition growth, or do you find yourself, prices are more attractive now?.

Well, it's a good question. Let me give you the flavor and the decision-making process that we go through. Our excess capital was building rapidly and at the same time because of the environment we have in the property cat area with additional capital coming in et cetera, it caused us to reduce our exposure into that sector.

Usually, we don't buy shares back in the third quarter because of the potential storm activity but with $450 million PML in Florida, that risk was a lot more manageable for us. So it cleared the way for us to look at our excess capital and also look at the prospects of us deploying the capital in the business.

We are not as optimistic in finding transactions for us that are reasonable for our shareholders. We had a few that we attempted and we were in the mix. None of them materialized being conservative in our approach of acquiring business.

And if I'm not able to deploy the capital into the business, then the next option is to returning to shareholders, and that's the thought process we went through, and we decided to initiate the buybacks even in the third quarter, which is not a normal activity for us for the reasons that I just mentioned.

Plus we felt the price that we were buying the shares, it was pretty attractive in reinvesting in our business..

And I would just add to that Kai. Following up on Dinos' very last point, from a valuation and attractiveness standpoint, we bought these back at $125% of book value.

So -- and you heard Dinos earlier in his comments talk about our forward look of our underwriting years, accounting years but underwriting years being 10% to 12%, so the arithmetic works..

And so if the current condition persists, in terms of both of your stock price as well as the prospects of the business, would you fear to say that you can return 100% of your operating earnings? And would that be a catch-up in the fourth quarter, in particular to match up the first half of the year?.

Yes, that's not an unreasonable assumption..

Okay. That's great. Then I just want to struggle with this question. If you look back last 10 years, your combined ratio averaged about 91%. Yet, your ROE is about 16%, and this year, your combined ratio is actually pretty good 87%, yet, ROE is low double digits.

I just wonder -- I mean, you mention you probably play an important part of that but just -- is the current environment even with this pretty solid combined ratio, that you will be able to only achieve like a low double-digit ROE.

Is that the prospects at least in the near term?.

No. You're doing comparisons without making adjustments for the E. There is periods of time that our shareholders equity is in the right place, so in essence, we don't have much excess capital and there is periods of time for different reasons that the equity we have is about or what we need to run the business.

Of course, when you have that, it affects your -- the ROE because the R is the same and the E is a larger number. It is going to give you a smaller percentage on that.

It's management's responsibility, and I'll take that one as my responsibility, as the CEO to make sure that we have a good balance between capital needed, the protection we need to have excess capital for various reasons that we explained in many of our prior calls and take the excess, excess capital and return it.

And that was the activity you saw in the third quarter potentially continuing in the fourth quarter et cetera. So when you look at ROE, you also got to look at the capital at the same time, and then you can make those comparisons about the quality of the business that we generate. We're happy with the quality of the business that we generate.

Unfortunately, in a competitive market environment, especially in the reinsurance side, we don't think we're going to have as much of it as we would like..

So just clarify on that? Is that your 10% to 12% current ROE expect -- sort of expectation, is that on allocated capital or on the sort of your -- sort of what's currently inside your balance sheet?.

It's on allocated capital, and we allocate capital to the business units at 2 notches above our financial rating. Our financial rating is A+, and we go to AA in the models to allocate the capital to the units..

Your next question will come from the line of Michael Nannizzi from Goldman Sachs..

Just a few questions on the MI business.

Can you talk a little bit about how much of that NIW, the $1 billion that didn't come through the credit union was through the bank flow channel?.

There was quite a bit for the bank channel, but it was on what we call prepaid single premium revenue and you're focusing just on the U.S.

primary MI or the global?.

Yes. I'm sorry to interrupt. No. I was thinking U.S.

MI primary business, how much of that, the $1 billion that isn't the credit union business would fall into that sort of flow business, flow bank channel?.

The flow bank channel is still at its infancy. It was about, I would say, 10%, 15% of the total. The other 85%, it was single premium prepaid MI..

Got it. Great.

And then as far as that bank channel is concerned, I mean as you continue to sort of scale up and you've mentioned that you have relations with the large banks, how do you see that progressing? And when you think about like a target market share and kind of where you are now, what sort of glide path do you anticipate to get there?.

Well, it's -- I'm an impatient guy. So if you ask for my perspective. [Audio Gap].

if I have one. And again, I apologize. Just couldn't -- I can't hear.

In terms of capital that you've allocated to that business do you anticipate sending more capital down to the MI, or do you feel like at this point with that 9.1:1 that you're at a comfortable place to pace the growth that you expect for the foreseeable future?.

Okay. At 9.1 we're probably the most conservative from a capital point of view from any one of our competitors in that space, and I don't anticipate sending more capital to MI until it's needed. We're not going to get through the steady-state with PMIers for quite a bit of time. Mark, do you want to elaborate on that..

Yes. It's -- the wildcard is the PMIers and that market place is dynamic too -- similar to the PNC side. It depends what the demand turns out to be, and we'll do it at that time but it could be that we need to put some additional capital over the next few years, but we have to see what materializes here..

And your next question will come from the line of Vinay Misquith from Evercore Partners..

First question is on the reinsurance side. Looking at the accident year loss ratio, cat about 58% for the reinsurance division. That's lower than the 60%, 61% you got in the first half of the year.

Is the business mix change, that's the removal of the Tower's reinsurance premium is having a positive impact on your loss ratios?.

Yes, and yes. But I'll give it to Mark who'll give you more details. Yes. We did book that at a higher loss ratio and the mix is changing, but the mix change is actually goes a little bit of against us because we're not having as much cat business, which usually the expected loss ratio, it's in the 40s. You have less of that, so there is a mix change.

But Mark, do you want to add more color to it..

Sure.

Vinay, I seem you're talking on an accident quarter basis?.

Yes, correct..

Well, other to reiterate, I think you've heard us talk about mix every quarter and how mix can affect things, and I think this is a great quarter of how mix has done that, not just by line of businesses but by the operations within the reinsurance group.

We always talk about the treaty side and the property cat side but we have a very profitable property facultative side, that really had outstanding results this quarter and helped with that loss ratio..

So we shouldn't take this quarter's number as a new run rate for the future?.

It's coming down to mix. I hate to do the forecast that you're asking for. So it'll depend on what materializes, and the reason I say that is because the property facultative book is not cat-dependent. They don't lead with cat. They pick it up in some more risks, but they're really looking for attritional underwriting, so -- which they're very good at.

So their results are not going to be impacted by what you're reading in headlines on cat business..

Okay, fair enough. And on the primary insurance, both a similar question.

Now was the -- were large losses about 1.7 points that you have currently?.

Well, the large losses emanate from the world risk book. You know what they are. You got the unlucky Malaysian Airlines hit twice and then you have Tripoli.

And now for that specific book, it was much more a loss ratio point, but for the entire insurance group, I think the effect is how much, Mark?.

Yes. Total was about 3.3 loss ratio points of worldwide premium..

Right..

And today, it was 1.6 at the corresponding quarter last time was similarly valued attritional loss sizes. So the difference between the 2 is 170 basis points of net earned premium that I referred to in my comments..

And what -- I mean, it's -- you can never tell, but these things happen. I mean you get unusual losses occasionally. We don't view these recurring but what I don't know in this world if you -- you could. If they continue to recall, we're going to reevaluate our underwriting posture..

That's in a -- taken from 10,000 feet up. I mean, this is a clear example to me of a large attritional loss..

Man-made..

Yes. Smoothing, if you will, because this is a cat cover at the end of the day, and you're going to go 10 years of 5% and 6% loss ratios and there's going to be a 500% loss ratio moment that is going to return back to 5% or 6% loss ratio point. So take it from that perspective, you don't really view that as a recurring item..

Right. Fair enough. So if I take that out of the numbers, the accident year loss ratio x cat and the PNC primary insurance business, was around 64.6 this quarter. And looking at the year ago quarter, it was 64.0.

So you're seeing a slight uptick in the loss ratio? Was that business mix? Because I thought that there was some margin improvement coming through the book..

Yes, there is some business mix, but we don't evaluate our businesses on loss ratio alone and then look independently at expense ratio, and then look independently at the duration aspect and the investment income side. It's all in totality. That's what return is.

So you can't really look at a loss ratio without looking at the outstanding improvement that they've gotten on ceding commissions, which finds its way through the net acquisition ratio, and by the way, it earns its way in.

So that's on a forward sense, because that's already baked and written, those ceding commission improvements are going to continue to be baked in firstly, and secondly, not every treaty renews at the same time. Those treaties are renewed at periodic points throughout the year, so additional gains are possible..

And of course, the duration of liabilities and investment income associate is part of the mix when we make a total return determination. And you're not going to view high excess workers' comp with very, very long duration the same way you're going to do E&S property who has no durations..

Sure.

And this quarter, the acquisition expense ratio went down, and the primary insurance division was that because of ceding commissions? And should we expect that to continue for the future?.

That's the point I was really making. You can't see it but the direct commissions paid up front really didn't change appreciably at all, so that whole balance is mostly, I can't say exclusively, mostly driven by the increase in the ceding commissions..

Your next question will come from the line of Jay Gelb from Barclays Capital..

I just want to follow-up on the buyback. It's a nice surprise to see the buybacks coming in the quarter and then following through the fourth quarter.

I believe there's a comment saying that buybacks could be equivalent to annual operating earnings, but I just want to make sure, before people start building that into their models that -- I just wanted to confirm whether that's sensitive to things like valuation, market opportunities and anything else?.

It's all in the mix, Jay. I mean, we -- I think we can chew gum and walk at the same time. So -- we don't make a decision and that decision is permanent without looking at the environment. We look at our share price, we look at prospects, work when it got deployed, capital, we look at the business opportunities, et cetera.

We look at our excess capital and how big that amount is, and then we make judgments. And beyond that, we make also judgment in not only how much we need to return to our shareholders but in what form we want to return it. So it's not only share buybacks.

If our share price gets to where it's appropriately priced in the marketplace, we might then do a dividend. So we take all that into consideration.

And your next question will come from the line of Jay Cohen from Bank of America..

Question on the mortgage business. You guys suggested that a lot of the premium, some of the premium came from the single premium transactions. I'm just not as familiar with the mortgage business.

So can you describe what exactly those are? And will the accounting and earned bases be different because they were single premium transactions?.

Yes, Jay. For clarification, yes, that will be the case. To the extent that they are single bullet, they will be earned over the ratable life. Think of it that way. And to the extent that they are monthly, they'll come in and as written and earned really....

On a monthly basis..

A monthly basis. So you're going to build up on a premium reserve on the single bullets..

Do you guys care which form it comes in? Does it matter to you?.

Yes, and no. It depends on the price at the end of the day. Single premium usually is not as attractive to us because it's competitively bid et cetera.

On the other hand, depending where the mortgage cycle is and interest rates, if you get a reduction in mortgage rates and you have a significant amount of refinancing going, then that's an attractive product because [Audio Gap].

And next question in line comes from the line of Ryan Byrnes from Janney..

The call has been kind of breaking up a bunch, so I'm not quite sure what we were just talking about. I was wondering if we could get, it also broke up earlier when we were talking about interest expense, the pressure from the loss portfolio transfer from January 2013. I just wanted to see how we should think about that going forward.

I think you may have answered it but the call broke up..

Okay, well, Mark will take it. Yes..

Actually Ryan, we really -- there hadn't been a question on that, so you didn't miss anything. The $8.2 million that we reflected in our comments and in the earnings release is a portfolio transfer that was effective January 2013 on basically, losses from 2000 to 2012.

It was reevaluated subsequent to that, and it was a reduction in the ultimate losses, and the accounting requires that you go back and retrospectively look at it as if it has always had that view.

So you should think of the $8.2 million since it was recorded as a reduction in interest expense as an outlier and that piece is not recurrent, unless we decide in the future to have another valuation, which changes the ultimate liability.

Assuming that there's no future changes to it, all you're going to see in future income statements is the accretion associated with that liability over its expected payout period..

Okay.

And how often do you -- you do a deep dive on that piece?.

It's periodic. It could change, it could be semiannual. It could be -- it depends on what we're observing in the underlying information and the data..

Okay, great. And then I apologize if this has been discussed as well, but the growth at Watford was very strong in the quarter, probably better than that most of us were anticipating. Just wanted to get your thoughts there on how close were you guys getting to a run rate there..

Well, we don't really know, because I can't predict the future. There has been a lot of interest on the facility and the facility competes in the marketplace. And at the end of the day, we've been pleasantly surprised about its acceptance, so far. If it continues, if -- it continues, we can get to a steady state within 2 years instead of 3..

And again just remind us, what kind of underwriting leverage that vehicle can get, obviously, thinking that it gets a little more aggressive on the investment portfolio side..

Well, I think it's a question you got to ask them, et cetera, but I'll give you my flavor. I think on a company with $1.2 billion of unencumbered capital, you can go into the $500 million, $600 million worth of premium written 0.5 to 1, and be extremely safe and pretty conservative with that.

And that's what I would call steady state, and it might take a couple of years to get there..

Your next question will come from the line of Josh Shanker from Deutsche Bank..

Don has already cascaded me a little bit but I don't know my accounting very well. I wonder if you can help me understand how the Watford noncontrolling dividends and that contributions work at the bottom of the P&L..

Sure, sure. It's a good question because it gets a little murky. I mean you can tell by my opening comments that we got to phrase things in a certain way. Here's what I think you're missing, Josh, is that in any company, the common shareholders are going to be basically subsidizing the preferred costs.

So there always deducted before, you have net income available to common shareholders. So think of it as -- we're looking at it as what's available to Watford's common shareholders and then you got to make the next step on what's available to Arch's common shareholders.

So you're effectively going to have at the 100% level down because of consolidation whats their net income, round one. Round two, then is the further deduction for the preferreds. Okay. And then you're at the point where you're taking the controlling interest into new account, which is roughly 11% and 89%.

So we need to take out 89% so that we're left with the 11%. But since you already subtracted the preferreds to begin with, you're implicitly doing at 11%, 89% on these preferreds when you make that adjustment.

I know this sounds a little confusing but suffice it to say, those adjustments, the $10.3 million that's there is the combination of the preferred costs and the net income split back 11%, 89%..

So the preferred dividend of $4.9 million is included in the $10.3 million?.

Yes..

Okay. Or is net of the $10.3 million, or is net of? So like maybe was there a Arch's share of Watford. . ..

Arithmetically, forget conceptual for a minute. Arithmetically, the $6.6 million, which was the net income at the 100% level in combination with the $4.9 million on the preferred that lump together. If you take 89% of that, comes to the $10.3 million, which is a sign reversal because we're undoing it.

So you're left with $1.2 million of net income available to Arch after taking out noncontrolling interest..

Okay. I think that after -- I think about it for 20 minutes, I'll understand it perfectly..

It took me 20 minutes. Don't worry about it..

The other thing that I've done wrong apparently is I've also miscalculate the operating tax rate.

It seems to me that you add about $158 million of pretax operating income before dividends, and after I make those adjustments for the preferred dividend, the dividends to Watford and Arch's stake in Watford, I'm still at about $158 million but operating income was $142 million after tax, which seems like you had a big tax bill on the operating side of the P&L, but maybe I'm wrong about that..

Yes. I can't figure out how you got there, quite frankly, right as I said....

But we know what our tax rate is..

We know where our tax rate is. We're at 2.5% on pretax operating..

Of course, operating tax rate versus group tax rate..

Yes. That real differential is the difference between 2.5 and 2.8 on net income versus pretax operating. So maybe we can have a cycle, I guess, but I'm kind of -- I don't know what you did..

That just shows I should go back to school..

Just call Mark and then go through it..

And your next question will come from the line of Ian Gutterman..

I guess my first one, Mark, do you have the spread of how much of the Watford premium, gross premium in the quarter was essentially Arch source business you did to them versus third-party business from....

We really never talk about that explicitly. What I can tell you is that it continues to build momentum, and we're happy with where it is. And this vehicle is really set up to be third-party reinsurance. So over time, this is going to continue to be growing towards being Watford dominant..

I'm surprised that being so smart, you can't figure it out. It's in the numbers..

I think I figured it out but I just want to....

Okay. All right. If you figure it out, if you see gross versus net and then the 15%, all that we do is reverse engineering and you figure it out..

Okay. But I was just trying to confirm my math. but I will take....

Okay. All right. I guess -- by the way, your math is going to be predicated on fundamental assumptions that may not be true in the future, but I'll wait till the future quarter for that..

All right, all right.

The other question on that is there is picked up somewhere and I can't remember who is the Insider, or some place else there suggesting that they're going to hit -- Watford is going to hit $400 million by the end of the year, do you have any comment on that, or?.

I believe what the article said it was that on an underwriting year basis, there is $400 million. So that doesn't mean by the end of the year, another quarter so. But it points to you that it's going to be approximately $100 million a quarter for 4 quarters..

Okay, great. Any thoughts on the satellite, the NASA loss the other day? Is that a industry event, or is that not have much....

It's $200 million or something. So the only thing I know I'm not on it, so I'm okay..

Perfect, perfect. And then my last one, this is where the audio cut out a little bit so I might have missed a part of this. But back to the single premium MI, I guess just to understand that a little bit.

Is that -- essentially is that kind of a bulk business and not GSE bulk but it's essentially bulk from the banks, or is this stuff you're writing?.

Yes. It's an originator. It's an originator. We will take maybe 1 or 2 weeks of production and put it out, and it says give me a price for you to write the mortgage insurance on this block. Here is all the underlying loans that we have, and we're going to prepay it up front, single premium instead of month.

And a lot of this sometimes is paid by the bulk because they include it into their -- into the price of the mortgage and sometimes it is paid by the lender..

Okay.

And is this just sort of something that makes sense given where you are building out the company that it's a good way sort of get premiums on the books quickly to offset some of the expenses, and probably over time as you get more flow, or is this?.

We don't think it as such. The way we think about the mortgage business is that you have flow business, you have this single premium business, you have other transactions we're doing in the reinsurance, the stacker and transactions, et cetera.

So we don't -- we're not trying to -- when we look at the entire marketplace and if we like a transaction, we go after it. And we don't have a preconceived notion. We have to have 3 of these and 2 of that and 5 of these. That's not the way we operate. You know us better than that..

Okay, okay. Just checking, and then just my last clarification on that topic. Mark, I think when there is a question about how this flow through the accounting. I understand the earned part honestly but on the written, it sounded like you said this still flows over 12 months. I guess, what I thought it was -- I think was single premium written.

The written would've all hit this quarter to be earned over 12 months..

A single is hit all-in-one accounting period and then earned over the ratable life..

Ratable life, not 1 year. It goes....

Got you. Okay.

But the point is from a written basis, it's more up front than traditional MI that, that comes into the written 1/12 each month?.

That's correct. The rest of it is monthly and is earned. It's written and earned all in the same month..

And your next question will come from the line of Meyer Shields with KBW..

Last question on the mortgage insurance side.

Is the expense ratio different from the single premium business, the acquisition expense I mean?.

No, I don't know if it's different. So you have your personnel, your underwriters, you got a fixed base of expenses. So at the end of the day, it is what it is. If you write it and you book it up front, you still have to service it over 6, 7 years. So I haven't thought of it as, is expense ratio difference between a flow business versus that..

Okay. And coming back to the insurance segment, I think you've started talking about some of the small account business where you're growing being less volatile.

Does that cost anything in terms of the anticipated underwriting margin?.

Actually no. I mean, it's not that it's lesser volatile, as Dinos pointed out before. It -- as a general rule, it comes with a lower loss ratio and a higher acquisition cost..

And at this time, we have no further questions in the queue. I'd like to turn the call back over to Mr. Dinos Iordanou for his closing remarks..

Thank you, Francis. Thanks everybody for bearing with us and we looking forward to talking to you [Audio Gap].