<<distinction between biotech companies with an ample in-house drug-discovery engine, who can bounce back quickly (e.g. IDIX), and “virtual” biotech companies who have to go out and license something to get back in the game>>
This is true, of course. However, isn't there something attractive about the pure play, the clean gamble on one molecule's success / failure? Let the share price go to zero if the program doesn't work out - holding a portfolio of these bets and not putting too much in the sector are fine forms of insurance, no? If one holds that the value in biotech ultimately has to come out in a molecule / molecules that make(s) it through development and past the FDA and to reimbursed patients, there's a sense in which the bouncing back to discovery can feel like merely playing for time.
A big part of my skew on this is that it's relatively straightforward to put a probability-weighted value around a late-stage pipeline and then to see if the market cap represents a rewarding discount from that value, or not. I don't know enough to get my head around valuing discovery capability and pre-clinical and Phase I candidates.
But I'd also argue that even big biotechs are little more than collections of drug projects. Their value tends to rise and fall with the success of those projects in the late clinic or in the market. Folks want to believe there's some overall capability beyond the projects that justifies a fancy growth multiple but I'm not very convinced. Genentech perhaps, at times in its life.
Other views?
Market Data 
Markets 
AI
