It's as important to keep in mind when to sell as when to buy. Big mistake to sell low when a company is worth a lot more, just because it's gone up a lot. Here's an estimate of intrinsic value for BRK. The arithmetic should be okay because I used a calculator. Comments and any competing estimates please?
Est. BRK Value end of 3d Q 2007 after tax numbers
85,000 investments per share, held primarily in insurance divisions
$3500 estimated non-investment after-tax earnings per share (Geico underwriting profit plus non-insurance divisions)
88.5 K p/e of 1
92K p/e of 2
95.5 K p/e of 3
99K p/e of 4
102.5k p/e of 5
106k p/e of 6
109.5K p/e of 7
113k p/e of 8
116.5k p/e of 9
120k p/e of 10
123.5k p/e of 11
127k p/e of 12
130.5k p/e of 13
134k p/e of 14
137.5k p/e of 15
141k p/e of 16
144.5k p/e of 17
148k p/e of 18
151.5k p/e of 19
Ben Graham rules of thumb: p/e 12 for a normal company, p/e of 20 for a growth company with strong prospects
" I am not a big fan of the two column method. Buffett divides the businesses into 4 parts."
its two columns, but if you like you can apply your own p/e metrics to each individual biz:
-fin & fin products: 10-14x p/e
-utilities: 16-20x p/e
-insurance u/w: 7x
I the retroactive losses likely lend some conservativeness to the u/w earnings.
I always go the conservative route and ignore underwriting results and then discount fair value (based on the two column method) by 25% for my "buy" price. The problem with this method is you don't get many opportunities to buy (maybe once every 5-10 years). The good thing is I have done very well with BRK using this methodolgy (which Buffett has supported in his writings). The reason you get such a broad range of values for BRK is that some people include underwriting and also project that the cash will be deployed at a certain rate of return.
I am not a big fan of the two column method. Buffett divides the businesses into 4 parts.
However, the insurance values are all over the place. I think a normalized underwriting profit for GEICO would be about 5% or a combined ratio of 95%. The retrospective deals, including Equitas have underwriting losses built into the accounting, although it should really be broken out. I don't know how that will mess up people's valuation metrics, but I wouldn't be surprised if at least some people don't end up valuing it for more then $7 billion BRK recieved.
You are also saying that the assets associated with deferred taxes are valued at 100 cents on the dollar.
I think the growth charts in the front are there to make a point about the importance of operating earnings, not as a literal valuation tool. For example, BRK has invested billions buying businesses to increase operating earnings for manufacturing and services. These businesses aren't growing organically at nearly the rate in the AR.
Anyway, the valuation isn't a fixed number but has increased maybe $20 billion in the last 18 months. All earnings and unrealized capital gains are fully retained, so it damn well better move up at a reasonable rate. It just so happens that the last 18 months have been better than reasonable -- damn good.
i guess I have to disagree w/ those that don't include u/w earnings in the "two column" valuation method.
We all get to BRK value in different ways...and funny enough, they all result to numbers north of current trading prices.
I guess part of it is differentiating between buy-in and sell-out valuations.
Buffett was right to buy GenRe for whatever he did, but he likely wouldn't sell if for same price metric he bought it for.
Strat, yes there is value there. Dave had a good post that showed the history of underwriting profits at Geico:
That's quite a history of growth as it relates to underwriting profits. So there is value there. However, I take the eashworth definition of merely valuing the investments per share when determining IV. By doing that it provides a natural margin of safety when running the numbers. The underwriting profits themselves are icing on the cake.
And what's cake without icing?
When you ask, "But does one assign no value to u/w earnings?" there is an assumption that no value is being assigned to insurance underwriting profits in the "two-column" method. That assumption is incorrect. The "two-column" method, in effect, values the insurance operations at the value of their investments. That is ONE way of valuing insurance companies; there are numerous others, including the float model often discussed on these boards. Note that Buffett paid the value of Gen Re's investments for that company when BRK bought it--AND Gen RE had a history of underwriting profits. Those underwriting profits weren't ignored; they were just included in the total valuation when valuating it by its investment portfolio.
Hope this helps. There are others on this board a LOT more knowledgeable about investments and valuation than me who may want to add to this discussion.
imo, Geico's UW profits and float are a very significant component of Brks IV. But it's difficult to calculate this and requires a wide range of assumptions. Eventually these Geico UW profits should stop growing and start declining but they should stabilize at a nice positive number as long as Geico continues to be the low cost operator. Geico is an amazing biz.
That's reasonable based on an earnings growth rate of 30% for operating businesses, enormous energy in the form of 40 billion in cash with more money coming every day, and positive prospects for the investment portfolio. Plus, and I get it HC, he owns the mutha.
I think you're right to leave out u'writing earnings. Buffett says long-term they just break even. However, in Geico's case that's not true because Geico has a structural u'writing profit and that's how it makes its money because it doesn't generate float like BRK reinsurance and whatnot. Or am I wrong about that? Facts are elusive to me and therefore I am always looking for data to replace what's corrupted.
As BRK rises, we need to be able to see where's a level where it's priced fairly so we know when we would not be suckers to sell. Some people need to sell sometimes. So I include Geico to avoid underestimating too much.
Including Geico, BRK still has a huge margin of safety due to (1) a reasonable multiple used in the valuation (2) earnings growth from incoming cash and the cash hoard. If BRK were at the market multiple using the two-column method, which it exceeded routinely in the nineties, it still would have a big margin of safety.