I read a few places before that 90% of the cost of developing a drug occurs after clinical trials begin but I have things to do other than write HN comments so

I read a few places before that 90% of the cost of developing a drug occurs after clinical trials begin but I have things to do other than write HN comments so I didn't look it up again. I will be happy if someone else finds a reference to support or deny that and will happily admit I am wrong if I am. Of course I'm totally wrong about the 90% delay I put in this last comment, I was thinking about this on the way to the office. If we're thinking about the added cost of regulation we have to compare the cost of regulation to what drug companies would spend on clinical trials in an alternative regime, and that number is not $0.