Thomas,
Before answering your questions, I'll tell what I use
for valuation. It is pretty simple approach which
has its positives and drawbacks. :-) It may be a good
platform for your "backtesting" exercise. :-)
As the basis I take the "Buffetology" spreadsheet
available at valuestocks.net
It requires 10 years (or less if you're comfortable :-)))
historical ROE, current dividend payout, current
market cap, current equity and that's it.
There are couple other parameters: 10 yr low and high PE,
short-term interest rate, marginal income tax rate and
cap gains tax which I set and forget.
Note: I set 10 year low and high PE to 15 and 25 for
all companies - I think that using actual numbers does not
give much advantage.
Now what the spreadsheet does is really simple. It
assumes 10 year forward growth using the last 10 year
average ROE and calculates the earnings in 10 years
from now. Then it multiplies it by low, average and high
PE and gives you the future market cap. And finally
it calculates the after-tax annual rate of return
from current market cap to the one in 10 years.
So now I can answer your questions:
>In the current topic, which aspects do you see as
> requiring or [currently] relying on art?
Mechanic use of the spreadsheet does not require art.
The artistic part is:
1. Does a company even qualify to be used in the
spreadsheet? Since we want at least 15% 10 year
average ROE, not many companies qualify mechanically.
But there are some companies that may qualify
mechanically, but would have to be discarded "artistically".
BTW, G&Ks are natural candidates to include, since
GG assumes that they can maintain their positions
and -hopefully- ROEs for long term.
2. What is the forward 10 year ROE? Using average
historical 10 year ROE is fine sometimes, but not
so good in other times. Concrete example: can MSFT,
ORCL, CSCO maintain their historical
high ROE and growth from here on? I doubt it.
3. Finally, there is some art in choosing buying and
selling points. I.e. you should buy if expected
annual return is >15%. But when should you sell?
At expected return <10%? 5%? 1%? I don't know.
I would like to see this approach backtested and you
almost talked me into doing that. But then
as I said, I currently don't have time. ;-)
Couple anecdotal "proofs": when ORCL and ADBE
tanked couple years ago they were definitely in
the buy range according to the sheet. I did not buy
them though... :-(((
Jurgis - hey this thing is TOO simple... :-P |
