What a great debate. Thanks to all of you for your insights, and some potentially very interesting ideas. As for the Buffetology book, I think it is the most poorly written of the Buffett books, but the most useful. The mistake people are making is focusing on the quantitative. That is the LAST step in Buffett investing, not the first. You don't screen for Buffett stocks. First you have to screen for Buffett businesses, and there is no shortcut. If you know what you're doing, they jump out at you immediately - there are so few of them. Then you ask what is the price I should pay based on that quantitative model (which, frankly, you usually have to adjust to the specific situation). If it still looks like a buy, then you do meticulous research to find out if its too good to be true. That is Buffett investing. The author of Buffetology presented the method fairly, but her error (or rather many of her readers' error) is that she made it sound way too easy. That is the fatal flaw of most investment books. This is an art, not a science.
And to go out on a limb, when a service company (Cambridge Technology Partners CATP) with a 3 year history, up 900% in three years at a P/E of 60 comes up on a Buffett screen, you're doing something wrong. UST and Oracle and a few of the others could be wonderful ideas though, if you can understand the risks (I can't on Oracle, but maybe others can).
Financially, just understand what you are doing when you use that model. You are assuming that the company can continue to invest its retained earnings at the same P/E for 10 years. If that is not an assumption that scares the pants off you, you are in for some bad surprises. David Dreman's writings on regression to the mean will put the fear of God into you.
Respectfully,
Jim |
