Biotech Stock Valuations (a black art?)

May, 1999:
One often assumes (too often?) that the price of a stock is some multiple, say 10, of:
 the company's Annual_Sales divided by the Outstanding_Shares

For example, by May/99, Amgen had annual sales of \$2.86 billion and with 512 million outstanding shares, then one might expect the stock to trade at 10 x 2,860,000,000/512,000,000 or about \$56 per share. At the time of this writing (May/99), AMGN trades at about \$63 (see chart, below) ... so this 10 x Annual_Sales/Outstanding_Shares ain't bad.

Stock_Price versus Annual_Sales per Outstanding_Share* (as of May 7/99)
*meaning the ratio: Annual_Sales divided by Outstanding_Shares    All numbers were obtained via the profiles obtained HERE.))

The straight line is the "best fit" to the points (identifying several biotech companies; the larger dots are the larger companies).

If we take as the equation of the straight line: y = 7.6 x + 25 then (since y = Stock_Price and x = Sales/Shares) we get

 Stock_Price = 7.6 Annual_Sales/Outstanding_Shares + 25

Try it on for size:
Stock_Price = 7.6 (2860/512)+ 25 = \$67,      for Amgen.
Of course, it gives a great value for Amgen cuz AMGN lives almost on the "best fit" line. But it also gives a reasonable value for the other biotechs used to determine the best fit ... except for Immunex (IMNX); why's that?

For example:
Genentech (GNE) has a (May 7/99) stock price of \$87 whereas the magic formula gives \$97.
On the other hand, the (May 7/99) price for IMNX is twice the valuation provided by the magic formula (\$105 vs \$53). Does that mean that Immunex is overpriced? who knows?

Another interesting point: the wee companies (Alkermes:ALKS and Aviron:AVIR) had negative after-tax income, yet the magic formula gives a reasonable estimate of their current price(s).

Now, on to something else: Let's assume that, in five years, some biotech (let's say Negus Inc.) will have annual sales of \$200 million and 20 million outstanding shares. Our magic formula would suggest that (in five years):

Stock_Price = 7.6 (200/20) + 25 = \$101

If one assumes an annual gain in the stock price of, say, 25% per year, then the current price should be about

\$101/(1.25)5 = \$33
Why not?

Another thing: if our hypothetical company, Negus Inc., was currently one of those wee companies (with negative after-tax income) we could still plug into the magic formula for an estimate:
Let's see, with current annual sales of \$17.5 million and shares outstanding of 16.7 million, the magic formula gives *:

current Stock_Price = 7.6 (17.5/16.7) + 25 = \$33

Is it an accident that both figures agree ... or divine intervention?

* As of May 7/99, where does Sugen (SUGN) lie, on the best fit chart, above?
Right on top of Aviron (AVIR)!

January, 2000:
Now, for the new millenium ... what with all the Biotech Frenzy ... we get a different picture:

If the chart gives some mechanism for comparing stock pricing - and that ain't certain(!) - then ... uh ... who's overpriced?