Re:Only in SV could Elizabeth Holmes lose $4.5B she never ha...
by dragonwriter on Wednesday June 15, 2016 @06:50PM
(#11912717)
> $9 billion is the statistically expected value
No, $9 billion is the subjective expectation of the party or group of parties willing to pay the most.
(Now, no doubt, they apply some kind of statistical formulae to conclude that that is the statistically expected value, but its also clear that other potential investors are also doing that with different models, and in the case of disagreement, the ones assessing the valuation the highest are, ceteris paribus, the ones with whom a deal will be made. Valuations derived from public market caps have some problems, and may not be justifiable rational expectations, but they are at least some kind of broad consensus. But private-deal-based valuations are inherently inflated by the way deals get made.)
> A related problem is the way startups are valued. Theranos wasn’t valued at $9 billion because it was bringing in massive amounts of revenue or held vast assets. Rather, venture capitalists valued the biotech company so highly because it had the potential to be worth that, or more. It held $9 billion of expectations. Obviously, this is not the same thing as cash.
This implies that valuations are intrinsically an overestimate of wealth.
$9 billion is the statistically expected value (https://en.m.wikipedia.org/wiki/Expected_value), which weighs outcomes by their likelihood. The best case is far, far more then $9b and the worst case is far less (assuming rational investment...ha ha).
Now, there can be trends of humans broadly overestimating value (i.e. a bubble), but there's nothing intrinsically wrong with counting illiquid or unstable assets. There's a chance that the future is better and there's a chance it is worse; the estimate takes the middle of the road.
Omitting these assets would be an underestimate.