This is off the current topic, but I just started reading this
thread today, and some questions occurred to me while skimming
the discussion between you and Mr. Michaelson on valuation of
companies.
If I understood the discussion (and I might have missed a lot
since I was reading it while waiting for code to compile), the
primary difference between your model and his is that you use
what might be called a "micro" analysis and his is more
"intrinsic". (These terms are surely incorrect in the language of
finance, or should I say Finance?)
What I mean by this is that you appear to be looking at a company
X, trying to project its value into the not-too-distant future,
and assigning a future value based on what it would sell for today
if it "were in that future state" now (with obvious modifications
for competitors and being in the some future state, etc.). That
is, you apparently don't try to assign an intrinsic value to the
company.
Mr. Michaelson's discussion appeared to attempt to do just this
in what I understand is the standard way using future dividends
and liquid value of assets (presumably with some goodwill fudging
thrown in).
There are at least two things that I don't understand from your
discussion. First, is it reasonable to suppose that a software
company has any "intrinsic" value? Any that I have ever worked
at don't pay dividends, and their hard assets are fairly
negligible (at least compared to their market valuation). And as
CA, Borlund, and Sybase have discovered recently (to name a few),
it doesn't always work to buy another company, so it seems very
hard to impossible to figure out what you could get for a
declining software company. Is part of the basis for this
"intrinsic valuation" the idea that someday MSFT (for example)
will start paying dividends, and that they will actually be able
to do so for long enough to justify their stock price?
The second thing that I definitely didn't understand is, do you
have some more fundamental model for estimating the value of a
company you are evaluating, or do you assume that the market is
sufficiently efficient that it assigns, on average, an accurate
value, or what? |
