We're in the middle of the second day. Some of you have sat through about 20 or some odd presentations, including an accounting presentation. At this point, we like to get sort of a jolt of energy. Someone who's got a lot of enthusiasm. We asked Richard Simmons to speak, but he wasn't available. He's too expensive anyway. We found a cheaper speaker. We're very pleased to have Dinos Iordanou, who's CEO of Arch, with us today. I actually worked on the IPO for Arch, and I would say that's a decade ago, and Dinos has as much passion for the business as he had back then, and it's a pleasure to have him here. Dinos?
Thank you, Jay. You never told me I was going to be following an accounting guy. We'll try to make it exciting. Let me start a little with our fourth quarter financial highlights. I know we had the conference call yesterday. We released numbers on Tuesday evening, four minutes too late instead of 4. We were 4:04, I think. Usually like to release a minute after the market close. Pretty good numbers, but not satisfactory to us. Our operating income was $0.92 a share. Our book value grew by 2.7% sequentially. Our combined ratio was just phenomenal at 90 with the background of all the cat activity we had. It was pretty satisfactory. Investment income, it was okay at $0.59 per share. A pre-tax yield of 2.72. Unfortunately, that's the environment that we have to look forward in working over the next few years.
I will have more comments about that as to what we need to do on the underwriting side to compensate for that. Cash flow was still pretty good at $110 million for the quarter and $866 million for the year. Our capital is very strong at $5 billion or so. For the year, again, you have to take that with the events of the cats in mind. We had $303 million after-tax operating income, which is down from $491 over a year ago, where we had much lighter cats. Our cat losses for the year, there were $400 million, which is about $6 million more than the Katrina year. I went back to look at the numbers. We had $398 million of losses during the worst catastrophic year, at least for us, until this year. Our ROE was 7.2%, down from 12% a year ago. Still way below about half of what our target is.
Our book value per share grew by about 7%. In a year that we had unprecedented catastrophes, okay results, nothing really super exciting for us, but we are satisfied, based on what happened in 2011. Now, let's put a little bit into perspective, what we're about, a little bit about our risk management. I think there's a slide that I'm missing, but it's okay. If you look, what I have put up is the 2005 year, the 2011 year, the two worst catastrophic years in the business, and a 10-year mean. As you can see, even though we're a significant participant in the cat business, we look more like a specialty company with more balanced exposure to natural cats than predominantly cat writers.
As you can see, our mean is a little less than 5% of capital over the last 10 years, and our worst year being this year, that's the yellow line, we do a little bit over 12%. From a risk management point of view, how we balance the exposure with our balance sheet and how much risk we take, I think we have exceptional performance over the last 10 years because as I said in many presentations before, independent how good the meal is, we don't eat too much of it because you get indigestion. The cat business is the most profitable part of our business over the last 10 years, but still you have to do it with a degree of conservatism, so you don't expose the balance sheet to undue risk over time. A little bit about market conditions.
This is the cumulative quarterly rate index, as you can see towards the end of the curve. I think the next curve is even better. We're starting to start lifting of rates, this is the CIAB survey about three or four quarters ago. On a relative basis, the rates are still back to what they were about in 1999, year 2000, especially for large accounts. As you can see, the small accounts have done better. They didn't harden as much, but they haven't come down as much. Then medium-sized accounts are somewhere in between. This is the environment that is starting to improve.
I think the last couple of quarters, we have strong evidence that things are moving in the right direction, but you got to see it from a cumulative point of view, and we're still nowhere near the good years of 2002, 2003, 2004, 2005 in our business. There's always questions as to when really the market turns and when we get a better environment to operate. I usually follow this. This is a U.S. industry statistics. As you can see, the light blue lines is where we had market turns all the way back, for all of us that have gray hair, in 1985 and 1986. Then the 1992, after Hurricane Andrew change, and then of course, the 2001 change. Look at the predictability of operating cash flow and market trends as a percent of net written premium.
Even though this is our projection, we're getting very close to those ugly negative numbers, that every time when we hit below the 10% line or 5%, all of a sudden there is enough stress in the system that pricing improvements need to occur. I think that's what we're witnessing today. The market is starting to really adjust to the pricing. The operating cash flow forecast is one area that we look at. Another one that I look is the return from two perspective, the return on surplus, which is the blue dots, the dark blue dots, and then the underwriting return, which is in essence excluding what you should be earning on shareholders' capital. The blue dots, it's all your return, including the investment income on the float, meaning on the reserves for the business.
The orangey or whatever color that is basically what you're really doing out of the business, excluding what you're returning on shareholders' capital. As you can see again, it's predictable that usually a market turns when the underwriting return, excluding return to shareholders' capital, gets to be negative, and we're getting very close to that. As a matter of fact, as a testament that this works in the marketplace, in 2008 when we had the financial crisis and that went to zero, at the beginning of 2009, we were starting to get excited because we were seeing some price adjustments, spreads came back in, everybody's balance sheet started to heal, we went like a bunch of drunken sailors competing again and cutting each other's throat. That's the nature of the insurance business.
Too much capital chasing, too little premium, you get a lot of competition. This time, I think it's more real because this is the weight of the underwriting actions companies have taken over the last four, five years, it gets us to the point that I believe a market trend, I don't know if it's gradual or more abroad, is in the near future. What do we do as a company in a business that is cyclical? We manage the cycle, we have a very simple principle that in the good years, we try to write as much business as possible, when pricing gets much thinner, we try to write a bit less, even go to negative growth as we have done, we take a lot of underwriting actions. We reduce lines and account size. We go to where we believe there is more profitability.
We buy more reinsurance to hedge risk. For us, when we shift the portfolio both to more property and short tail because that's where we saw more opportunities over the last four, five years, also to smaller accounts because we saw the pricing with the prior slide that I said that it was more attractive in the small accounts. We're conservative also during the cycle on our investment approach. Usually have a shorter duration on our portfolio. It's a hedge against inflation and the economy. We always have a high credit quality on an investment portfolio, because at the end of the day, you buy us for our underwriting capability. You're not buying us for us to be super investors, per se.
On the capital management, we maintain very low financial leverage because it allows us over time to have flexibility to grow capital without diluting common shareholders. If we have excess capital, we'd rather return it to you than try to give you the 3%-4% skimpy returns we can get in this market. Even though these are numbers that most of you know, I just put them up over 10 years to show you that if you don't write a lot of business in the soft cycle, it will hurt your ability to grow rapidly in the hard cycle, and that's a myth, in my view. We started this company in 2002. If you can see, we got to almost $3 billion of written premium within three years. The supply and demand equations work very well in the marketplace.
We did this at a time where we still had an A-minus financial rating, and we did this when we were still building our underwriting teams and our infrastructure around the country. Given a market cycle, even though it's hard to predict what the opportunities are. We're so much better positioned as a company with an A+ across all rating agencies, $5 billion of capital, and all of our underwriting teams in place to take advantage of a better cycle. Since we run the company on bottom line profitability, we pay our people on that same basis, and we don't really worry about top line.
It's not a huge point to make, but I believe that given the right market opportunity, you can see significant growth out of this company based on how we have managed not only our resources, intellectual resources, but also our capital resources with the way the balance sheet has been constructed in order for us to allow us to have all the flexibility to write as much as possible or as much as the market would allow us to write without having the inflexibility of capital constraints. Another view is to see that we're very nimble and we're willing to change our mix significantly in order for us to find the more profitable opportunities. In 2003, our casualty book was 44%. 2011 is 19%. That's a significant change. This is on the reinsurance group. On the other side, look at our property and property cat.
It became more than 50% of our business. That was the shift from long tail to short tail. Not because we wanted to do that, it's because that's where we thought the opportunities for better profitability were, and then we were able to change our book of business. Similar strategy that we executed on the insurance group. Our casualty was 20% in 2003, is about 7% today. It didn't happen by accident. It happened by design. It happened because we always instruct our underwriters to look for the opportunities for profitability and have the willingness to either shrink or grow different lines of business based on what the market will allow us to do. A little bit about new initiatives. Our playbook is usually to grow the company organically.
We don't like to put a lot of press releases every time we hire either a team or an underwriter, et cetera. We've done a lot in the last six quarters, and I figure I'll just list them for you this way. You have an idea as to where we're going. In Canada, we have converted our existing branch of the U.S. into a subsidiary, and also we acquired a treaty reinsurance group team. In Canada now we have another underwriting center that we didn't have before on the treaty side. In Europe, we introduced mortgage insurance. We're going to do both structure and flow business for banks and building societies.
The construction of that subway.
It is a subway or they're sending me a message? We just obtained a license for mortgage insurance for us to operate in the European theater. It's an Irish company that we just got a license with an A+ rating. Of course, we have the property facultative operations that we introduced to the company about four years ago. We expanded those in Europe this year, so we have the ability to underwrite European exposures with offices in Zurich and in London. We brought in a team in Zurich to do international crop reinsurance, then on the life reinsurance sector we have a Continue to strengthen that team, which we brought in about six quarters ago. None of these initiatives is starting to produce significant number of premium. We're not premium focused. We're more capability focused and having the ability to deploy more of our capital in our business.
Over time, and given the right market conditions, I think any one of these could be a significant part and a good addition to the franchise value of Arch. I'm not going to spend a lot of time on the financial performance of the company, but I want to talk about the balance sheet a bit. As you can see, we have maintained a very low leverage, financial leverage on the balance sheet by design. Is not for us to get the A-plus, et cetera, but also the flexibility that brings us in case of a market trend that requires us to expand rapidly, we will have the ability through either debt or perpetual preferreds, et cetera, to increase our capitalization without diluting the common shareholders.
Our goal here is to create positive accretion for our common shareholders and take advantage of the balance sheet without losing the ability for us to have our profit centers grow uninhibited by capital constraints. The only inhibition will be market conditions. Given good market conditions, they can write as much of that business as possible. Of course, our liquidity is impeccable. That's never an issue with the way we run the balance sheet. Our financial performance over the last 10 years, here it is. Early on as the book builds and investment income has to flow in, not totally acceptable, but pretty attractive. Where I want to point you, even though I made that comment already early on in the presentation, look at 2005 and 2011, the two worst cat years in the business, and we still achieved positive returns and reasonably positive returns.
If you do smooth it out, you will see that our performance pretty much follows the cycle. We started the buildup, very good performance with those two adjustments, 2005 and 2011. Of course, as the cycle got away from us, our performance suffered too, along with the rest of our competitors. This is the measure that we take a lot of pride in. It's our annual growth rate for building book value per share. At the end, this is, for us, the most important metrics in our financials. That's how I get paid. That's why it's dear to my heart. The comp committee looks for me to grow book value per share for the shareholders, they look at a lot of other stuff, but this is the most important metric for me and the other people in the holding company.
We did compound 18.7% from our beginning until now. When you look at the average annual value creation for us, as you can see, we run among the best with pretty low volatility as you look at the standard deviation. Also, we are pretty proud of this performance over the last 10 years. Another way to look at it is our return on surplus. There's been a lot of studies being made about investing in the P&C sector. I think there was a McKinsey study. There were some other people like Dowling & Partners who've done studies on this, et cetera. Basically, for the companies that they are on the top quintile of performers, their returns are exceptionally good, even in a business that usually over a long period of time, the return has been around 7%-8%.
If you look at the P&C sector over 30 years, the average is in the upper mid-single digit return. However, companies that they're on the top performing quintile, they have very good returns. The other conclusion in a lot of these studies is once you are achieved to be in that elite group, the likelihood of dropping down is small. Where if you're in the underperforming group, the probability of going to the top performing pool is also very small. This is a business that is for stock pickers, and you got to pick the right horses to put your bets on because at the end of the day, that's usually what happens in the P&C world. Let me summarize before we get some time for questions. We have executed a disciplined approach across the last 10 years and the pricing cycle.
We grew rapidly early on as a company because the opportunities were there. Maybe we made a little bit of a mistake by cutting back a little bit too soon, I've said that in prior presentations. At the end, directionally, we got it right. Now as we're starting to see opportunities again, we might, depending on the market conditions, allow us to grow again. We did maintain a conservative balance sheet, both on the asset side, and also on the liability, very low leverage. Over time, even though it took 10 years, I think all the rating agencies have recognized our performance, and we went from the least capitalized startup in 2001 to one of the strongest capitalized with an A+ across all rating agencies 10 years later. We have generated very strong risk-adjusted returns over the long term with low volatility.
One of the areas that I think some people understand and some people do not, I believe it has a lot to do with our success as an organization. We align the interest of our underwriters with the interest of the shareholder. It's a pay-for-performance company that rewards its underwriting staff on underwriting results. We do it over a long period of time. Every underwriting year is being calculated four years out for the initial payouts and then six additional recalculations, and we don't close any year until 10 years later. There is no messing with the numbers. At the end of the day, it is what it is. You write the premium, you pay the losses, and at the end of the day, you count your investment income and you see if you made money or not. If you made money, you get paid.
If you didn't make money, you don't get paid. I think that's the way shareholders get rewarded, and I think underwriters should be rewarded in the same fashion. With that, we'll take your questions. I never thought I was that clear. I left everybody speechless. Right?
Looks like you did. I'll come up with a question now. I think through the fourth quarter, we certainly heard a lot of commentary about pricing on the insurance side specifically, and some of the momentum we saw during the year obviously varied by line of business. Can you talk about what you are seeing, at least qualitatively in January, and maybe even early February, are the trends that you saw last year continuing into 2012?
Yeah. I will characterize the movement a slow gradual movement. I think third quarter of 2011 versus fourth quarter versus moving slowly up. That doesn't mean that on an absolute basis all lines are there to give us double-digit ROEs, et cetera. As I said in our call yesterday with investors, we still believe that our 2012 projected underwriting year return will be in the 8%-9% basis. Don't forget, I'll be very poor if we're only producing 8%. Our incentive compensation pays 100% incentive of 15% ROE, and it falls off the cliff at 8%. If we don't do 8%, we get zilch. Zilch is not a good word. Then, of course, if we produce 23% ROE, we get 2x incentive. We're incentivized the right way.
Still the market is lifting, not enough yet to get us truly excited that we're going to step on the accelerator and try to give you significant growth. We're monitoring it. It might take another one year, it might take another two years. We'll be very patient. We'll be patient up to this time. When we find the opportunities, we have the tools in the toolkit to take advantage of it. Branch offices, fully staffed with underwriters. We didn't cut back the staff, we have suffered with maybe a little bit higher expense ratio than I would like to see. You don't destroy the factory because you can't sell a lot of cars today. You have the factory for the future. We have a strong balance sheet, we have the A+ to put that in play. Yes.
I think one of your slide one, your casualty is a higher %, now you've brought it down as a weight. Is that something you want to keep, it depends on pricing? How do I think about that? Thanks.
The casualty business, I've been in this business for 37 years or so, it's feast or famine. It has the most violent pricing movements for the simple reason people can get too optimistic and too pessimistic, the longer the duration of liabilities, excess casualty has long duration, excess workers' comp has very long duration. It causes people to fool themselves, the price adjustment becomes more volatile. Given we have no preconceived notions as to how much we should be writing in casualty or how much in property cat or other property. It's only pricing. If pricing improves significantly, you will see a mix change again, we can write a lot of casualty business. That's what drives us. It's the returns.
I'm not too optimistic on the casualty business for the simple reason that 3% or 2.5% new money invested yields that you're getting today in investing in high credit quality corporates. It's not going to carry the day like it used to with a 5% or 6% or 7% return, even on business that it has a four-year duration over time. Very hard to predict the future. If prices go up significantly, we will write a lot more casualty, yes.
That's great. Well, thank you very much, Dinos. Great presentation.
You are quite welcome. Thank you.