ERICOPOLY

You are correct, Chubb has been shrinking their net premiums. Their commercial lines declined just 2% in 2008 and a further 7% in 2009. But as Chubb states in their 10-k, some of that is due to foreign currency changes. Crum shrank by 16% in 2008 and a further 22% in 2009! I doubt it was foreign currency either.

Chubb is a big cookie. Cardboard has been putting up their numbers for the entity as a whole, but here are the results for their commercial lines in 2009: Chubb's 4 commercial classes of business (42% of Chubb's premiums written in 2009) Multiple peril (85.8% combined ratio) Casualty (96.7% combined ratio) Workers' comp (92.7% combined ratio) Property and marine (83.3% combined ratio) Total commercial (89.9%) Chubb's casualty business (it's worst measured by combined ratio) in 2008 and 2007 had combined ratios of 95% and 98.6%. The pond analogy might stick, or at least it might explain a lot of it -- for example, why is Chubb's casualty combined ratio so much worse than their Property&Marine? Further, if their overall commercial ratio is 89.9%, which is 6.8% better than their casualty combined ratio, are they doing most of their business in multiple peril and property and marine? And how does that compare to Crum&Forster? Is it the same pond? Does Crum&Forster write a heavy amount of multiple peril and property and marine? Or is C&F more weighted towards Casualty? Then, another topic is to notice that C&F's expense ratio was 4.3% higher than Chubb's during 2009. Unfortunately, the Chubb annual report does not mention what the expense ratio was for it's commercial lines, but rather merely states it for the company as a whole. Now, C&F had a combined ratio of 102.2% in 2009. If you strip off that 4.3% from it's higher expense ratio you get a CR of 97.9% for all of C&F compared to Chubb's 89.9% for it's commercial lines. But are the ponds the same? I think the choice of not discounting the reserves is making at least a 1% impact on CR, and that further knocks C&F down to 96.9% compared to Chubb's 89.9%. That's a difference of 7% overall, but again how much business does Crum write in the areas of Property and Marine and Multiple Peril (Chubb's best lines). Crum's adjusted CR is not that much higher than Crum's CR in it's Casualty line (2/10 of a percent difference), and it's 4.2% higher than Chubb's workers' comp number. Unfortunately, I can't find a breakdown of the CR's for C&F's different business classes.

Cardboard, If they are competing head to head with Chubb then the issue at hand is not one of underwriting discipline -- it would be either reputational or for some reason Chubb' clients have special loyalty. Chubb is much larger than Crum and Crum's volume is plummeting. There is some reason why Crum is losing business despite offering policies at lower prices. Call it the same pond if you wish, but there is something going on in the marketplace other than simply price. Does it make sense to explain the divergence as mere pricing discipline at Crum when Crum's volumes are dropping through the floor at a faster rate than that of the larger Chubb?

The numbers vary so much. Fairfax Asia always has tremendous underwriting, whereas C&F is high CR. Asia keeps growing whereas the falling volumes and high expense ratio at C&F is killing them. In the very long term, Asia will overshadow C&F. Don't fault Prem. C&F seems to be what it is -- not the same pond as Asia. I would bet that if you hired the Chubb underwriters to run Crum you likely would get the same CRs. I think it is a Carp pond where you catch carp, not trout. Now, if you double volume in a hard market you'll get underwriting profits at Crum high enough to fill in the holes you see today. I'm using Crum as a metaphore for all of their operations that earn high CRs. Not just picking on them. Generaly, their underwriting is far better than what existed prior to acquisition. So they have discipline, but I think they bought carp ponds and they won't produce trout.

Personally, I make use of the "except for" number to see whether the underlying trend of underwriting is getting better or worse. That's why it is valuable to me and I believe that is why it is valuable to Prem. You are misunderstanding his motives I think. Stated differently, Prem turns to his presidents and says, "Stop bullshitting me gentleman. I'm tired of you guys always obscuring your overall performance behind a big cat loss. You can't be a major league player unless you can consistently field and hit well. So you'll tell me how bad things are even without this big loss."

Imagine a world where they only provided the one large CR number... Can you guess what the first question on the conference call would be? My guess: "How big of an impact did you suffer from the Chile quake?"

Buybacks given out to officers for retention/awards do not help us out though. They buy shares every year for that reason. Not sure if they bought them for that reason this time, but at that pace it's probabably safe to say that they will give award at least that many shares in any given year.

Every time we have a big cat loss, somebody always grumbles that Fairfax states what the combined ratio would have been were it not for the big event. This quarter will not let you down, as there was the Chile quake and they state: The combined ratio of the company's insurance and reinsurance operationswas 111.5% on a consolidated basis, producing an underwriting loss of$122.6 million Prior to giving effect to theimpact of the Chilean earthquake losses, the company generated acombined ratio of 98.7% and an underwriting profit of $14.2 million. Okay, so it's time for somebody to step up and grumble about this again :)

6.4% gain in book value adjusted for the dividend (28.1% annualized).

The bottom of page 33 (2009 AR) outlines how much this reinsurance program costs them each year. pre-tax net impact of ceded reinsurance transactions 2009: $337.5m loss 2008: $144.3m loss 2007: $388m loss

Okay, I actually did some research (hurray!) First, turn to page 33 of 2009 annual report: The cost of reinsurance purchased by the company (premiums ceded) is included in recoverable from reinsurers and is amortized over the contract period in proportion to the amount of insurance protection provided. Next, turn to page 52 of 2009 annual report: Here on page 52 we find the effects of discounting: insurance and reinsurance liabilities overstated by $539.5m ceded reinsurance contracts overstated by $284 million The last thing we need to do is to subtract the overstated reinsurance recoverable assets from the overstated liabilities, and I get $255.5m total overstatement. That's before tax of course. Shareholders' equity is understated by $255.5m (pre-tax) due to their decision to not discount their liabilities and ceded reinsurance contracts.

So if they have been so conservative for so many years, why aren't they showing up today to offset this headwind of conservative reserving? Doesn't the conservative reserving show up in the provision for claims? For 2009, the provision is approx. $3.6B higher than in 2004, although the actual claim payments and the level of new business is approx. the same. I know this is very simplified and relies on a lot of assumptions, but would nevertheless seem to explain the different numbers and statements by the company? I get a FFH CR of approx 87 if the reserves should have been on the same level as in 2004 (like premiums and actual payments on claims). Cheers! Thank you for giving me something useful to think about. This will keep me busy for a while.

The first thing that comes to mind is that... is it illegal to deliberately inflate reserves for the purpose of reducing taxes? I can understand that being a little bit conservative is perfectly okay, but there must be a threshold where this gets abusive in the eyes of the law. Second thing that comes to mind is that... now that volumes are dropping this should be showing up as a tailwind to their CR, not a headwind. See my example above about the company that, over a span of 4 years soft market, winds up writing $50m business instead of their past $100m. If they were overreserving by 5% in every year, then they would be getting a tailwind of $5m worth of reserve releases to offset $2.5m of reserve additions. The net result is an improvement to the CR. Now, with Fairfax we know that they are writing less business -- so the reserve releases from past conservatism ought to be more than offsetting the present year loss reserving... or at least we should be getting to the point where they nearly cancel one another out. My point is that you can't just keep hiding your taxable income in a reserve -- at some point it comes out and gets taxed. So if they have been so conservative for so many years, why aren't they showing up today to offset this headwind of conservative reserving?

MKL does purchase reinsurance. The 2009 annual report claims that MKL has $952m reinsurance recoverables. That's 1/3 of their shareholders' equity. I'm not sure how much reinsurance coverage they purchase in total, but their recoverable balance must be some indication. By comparison, Fairfax has recoverables of $3,809m (about 1/2 of shareholders' equity). However, in Fairfax's case a good deal of that is held in the runoff units. Where did you hear that the operating companies of FFH have more reinsurance protection vs the operating companies of MKL? I also looked at the total leverage of the two firms. They both operate at 3.7x leverage. (total assets / total equity) However in Fairfax's case I think that's overstated because they seem to keep a much greater sum of money at the holding company just waiting for it to rain.

Just to clarify, I was ignoring all other qualitative factors and sources of earnings. MKL and FFH I like to compare to one another because if you had the same investment team managing both companies then perhaps the only significant qualitative difference would be the underwriting profits. So I took the gap in their respective underwriting profits and put a P/E of 10 on it. I found that it generally explains rather nicely the entire relative premium to book value enjoyed by MKL vs FFH. The reason why I brought this thread up is that I was once an MKL shareholder (back when I joined this board) but I sold it at a very high multiple to book. Now I am a Fairfax shareholder and don't own any MKL. I like the investment team better at FFH but from time to time I wonder if I would do better at MKL due to the relatively better underwriting profit -- this of course makes me ask whether it is worth making a switch given the much higher P/B that I have to pay in order to get that relatively better underwriting profit advantage. That's what drove me to ask how much I am paying for the underwriting profit at MKL... and when I discovered that it's about 10x earnings, then I thought well there's really no reason for me to switch because I could put that extra bit of money (the relative difference in book value multiples) in a better business like WFC for what I think would be better results. I think over time Fairfax's underwriting profit will get better because their worst businesses seem to be shrinking and their best one (Fairfax Asia) is growing. Plus they just bought Zenith and in a hard market they could always favor growth there over their other less profitable lines. Partly I'm writing out my thoughts here hoping somebody will correct me if the logic is horribly wrong.

We're in agreement that they are disciplined... look at their dropping volumes. Absent discipline, they wouldn't be walking away from business. The topic is what Cardboard asked... why is their level of discipline not winning the CRs of others who make the same claim of discipline (he mentioned by name Berkshire and Markel and Chubb)? So I reasoned that they are simply fishing in different ponds (different lines of business) -- you can be an extremely good fisherman but only catch carp if you are fishing in a pond that only holds carp. If that isn't right, then what is? The fact is that they are disciplined (backing away from unprofitable business) yet their CRs are still far higher than respectable underwriters like Markel,Berkshire,Chubb... or is it that Fairfax is disciplined but these other three companies are not? I can't buy that given their long track records.

Assume that over the past 4 years the market has been getting soft... Let's say that you write $100m of business 4 years ago and only write $50m of business today. And let's say you reserve too high all along, such that 5% of your business written in any given year will come back out as a reserve release 4 years later. This means that today you are excessively reserving only $2.5 million for the business you wrote this year, but you are getting a favorable reserve release of $5m from the business you wrote 4 years ago. So doesn't this tend to INFLATE your current combined ratio by the net $2.5m reserve benefit? And in 4 more years if you are still writing the same volume of business, that $2.5m benefit will vanish. So Fairfax might actually be seeing a tailwind from their reserving practices as they shrink their business -- but if their business stops shrinking the tailwind drops off.

Their volumes are dropping hard though. It's true that they are walking away from business, yet they're just not getting meaningfully below 100%. Could it be that they are weighted towards lines of business that are easier to enter and so capital flows in freely (aka: fiercely competitive). I don't know... you have to wonder that this might be the case. Otherwise, why do they keep giving up business and tend to write aggregate CRs much higher than WR Berkeley and Markel? It has to be that however they're set up, they're fishing in the wrong pond. Some ponds have trout and some have carp. Better theory?

I am trying to tease out why exactly MKL is priced at roughly 1.34x reported book and FFH is priced at roughly 1x reported book. Assuming they both have roughly the same allocation to equities (they do after accounting for FFH's hedge). Assuming the same investment results (let's ignore for a minute that HWIC has handily beaten Gaynor's record) Assuming they both have tremendous capacity to grow in a hard market About the only thing I can see that differentiates the two of them is the past superior underwriting profit at MKL. So, one might come to conclude that the relative premium paid for MKL is effectively the capitalized value of that underwriting profit. I know that over the long term profits are profits, but... were the profits at FFH to be just a little bit higher and to come in the form of underwriting gains, it would translate to a current stock price of $500. I think at 1.34x people are effectively paying about 10x earnings for MKL's underwriting profit. A conclusion one might make: It might be tempting for some to want to go with MKL based on the theory that the underwriting profit will drive book value gains at a higher pace (assuming identical investment gains), but I don't believe it will translate into higher investor returns when you are diluting those returns by investing 0.34x at 10% returns (unless 10% turns out to be the highest component of the return). And there are better business trading at or below 10x earnings than MKL's underwriting. Take WFC for example -- that should presently be trading at no higher than 10x earnings... and one could argue that it's business is much more enduring and scalable than MKL's insurance underwriting profits. So if one assumes that FFH and MKL earn identical returns in their business before including the underwriting profits, then at these prices I would have to conclude that for a $13,400 investment it would be better to go with $10,000 in FFH and $3,400 in WFC rather than going with $13,400 in MKL. Have I vastly miscalculated what people are paying for MKL's underwriting profits? I get about 10x.

This article suggests that US banks in the aggregate hold $16.4b in Greek debt. http://ftalphaville.ft.com/blog/2010/01/29/137341/whos-selling-greek-cds/

He does think they are expensive, but he sees stocks more likely than not going to 1500 or 1600 by October 2011. http://www.gmo.com/websitecontent/JGLetter_ALL_1Q10.pdf "the line of least resistance is a market move in the next 18 months or so back to the old highs, say, 1500 to 1600 on the S&P, accompanied by an equivalent gain in most risk measures, followed once again by a very dangerous break." Further, he is overweight emerging market stocks because (although he thinks they are expensive now) he sees them getting far more expensive.

They keep the old ones well hidden, but they can still be accessed. I can't find the 2010 one yet: http://www.fairfax.ca/Assets/Downloads/2006_AGM_Slide_Presentation.pdf http://www.fairfax.ca/Assets/Downloads/2007_AGM_Slide_Presentation.pdf http://www.fairfax.ca/Assets/Downloads/2008_AGM_Slide_Presentation.pdf http://www.fairfax.ca/Assets/Downloads/2009_AGM_Slide_Presentation.pdf

I believe Pabrai just wants to keep his message simple for the interviews, and that he is fully aware that you can go short without infinite downside. For example when Fairfax shorts the S&P500 they do not have infinite downside -- and Pabrai invests in Fairfax. Fairfax hedges the short with call options.

I don't know much about stocks yet even I can still find some cheap things in obvious places (cheap if you believe recovery is taking hold). Citigroup for example, despite the run from $1, still trades at 1.2x tangible equity (plus they are reserved at 6.8% of loan portfolio for loan losses, and their loan losses are falling the last three quarters). Berkowitz goes so far as to say that they are overcapitalized.

It's somewhat of a coincidence that I'm hearing more about Crohn's lately (after hardly knowing anything about it). Yesterday my neighbor was telling me that people are getting treatment by exposing themselves to hookworms, of all things. http://en.wikipedia.org/wiki/Helminthic_therapy