Sunday, October 22, 2006

Overreaction: Chapter 1

I have noticed often is the case that stocks get hammered for no good reason or jump to sky for no justifiable reason. I have grown to believe that when overreaction is on the positive side, you really can't bet against it, but on the downside, well, there are occasions that a patient investor can make bundles.
For instance recently, I come across CAT and SNDK. Both dropped by more than 14% and 20%, respectively in one day, following their quarterly announcements and guidance. Take for example CAT, the company came short on earnings, 1.14 instead of 1.35 and cut full year from 5.25-5.50 range to 5.05-5.30. I use moneychimp calculator for a back-of-envelope type computation (see http://www.moneychimp.com/articles/finworks/fmvaluation.htm). So with 5.50 esp, and analyst expected 5-year growth of 12%, 5 years of that growth, and 12% discount rate, the price is $73 (price two days before the skydive was $71!). Now let's see what should happen to get a $59 price. Assuming discount rate is 12%, and esp of 5.0, then growth has be 9%!!! So the worst-case earnings has to happen and the worst-case growth has to happen for price to be where it is now. I am not saying that CAT's price won't fall anymore, I actually suspect it will drop some more, possibly another 10% at some point in the next month or so. But after all the dust is settled, I think the price would come back up. Now if you add to this the fact most of the "mis" this quarter was legal cost and the firm might be managing end-of-the-year expectations, I dare to venture that three to six months from now the price is around, say, $80, give-or-take couple bucks. [I am going on record, so we'll see how foolish I will look in 3-6 months!]
As for SNDK, I am bit more cautious, because they came short of expectations due to increased expenses and growth was negative besaue of price-cuts. But considering that they acquired FLSH and consolidating market share and growing into flash music business, I venture a guess that stock might see price as high as $110 in 9-18 months, but be warn that in next 6-9 months its price may fall to a low of $38. I say this because given pre-announcement forecasts, SNDK was supposed to have 2.4 eps. To get the $62 pre-fall price, either growth had to be 30%, or years of growth 7+, or discount rate at 9%. Each one of these is quite unreasonable, so for now the fall is just a reversion to normality, then comes overreaction to irrationality and that my friend is signal to buy. Even if SNDK, as is said, loose 25% top-line pricing (they're marking down the price to compete!), then at 1.8 eps (this a big hit!) and growth of 20% (this seems in line with industry and analysts forecasts) then price should be $32 and P/E at around 17. Historically in past 10-years a P/E of 15-17 has been almost always the trough for stock cycles. So if we get there, it is a BUY!

Dirt Cheap Stock Picking: Chapter 1

I am a cheap b... and so I follow that philoshephy to heart. So over the years I have devised my own method to find "good" stocks (i.e., hisrorically robust earnings, growth, some recent indications of momemtum, etc.) that have gone thru some sort of correction. I admit that filters are intially all quantitative but they give me a head-start and I go from there. Here are my recent picks: CNS, CTSH, LM, BHI, SLB, VLO, RIG, and URBN.