Pharma research does not have network effects
When they do get corrupt what do we do, since it's the same guys with the guns and court houses?
There are some very messed-up public works, but most of the time it works pretty well.
FWIW unlike 90% of this thread, this suggestion doesn't sound insane to me. We could look at the number of people with a disease, and the QALY swing caused by the treatment, and then pay the team which designed the drug for the patent for (expected number of treatments) * (expected gains in QALYs).
This would allow the drugs to be any price the government liked once created, but still convince financially motivated smart people to work in medicine. Bonus points if the solution included ever-rising-prices for long untreated diseases (kind of like an auction).