Program survival bias
The first point makes absolute sense, but the second is criminal. I would believe it of some of the scams, but I don't think that the implication that a substantial number of failed trials is a consequence of CEOs pulling a fast one makes sense. It's a bit too cynical for me.
I think that a lot of the failed PIII's are due to an inadequate nummber of PII's run by small bios. They don't have the resources to conduct exploratory trials. And that is a consequence of the constant state of penury in which they exist. Thay can afford neither the cost nor the consequences of unsuccessful PII's.
One of the things that has made CTIC attractive is the sheer number of clinicals they're currently in. Despite all the legitimate knocks on the Biancos, they are serious about doing it right.
Bob