>Funding research through banning a buyer from negotiating with its suppliers is pretty much the definition of a Rube Goldberg device.
It's exactly how normal (non-monopsony) markets work in every context, including markets for intellectual works/IP.
Note: in this context, "negotiating" means "turning the national market into a monopsony". It's generally recognized as not-a-pure-good-thing when e.g. Walmart comes close to doing that and forces sellers to nearly take a loss in return for access to Walmart's market.
(And before anyone makes the obvious point, yes, I know "health care ain't a can of beans etc etc etc". But with respect to the dangers of monopsonies, it's close enough.)
I don't think anyone would dispute that that's a major way that research is currently funded, but it's fundamentally stupid. You've got a bunch of free riders, and it obscures where and why costs are actually incurred.
Allow negotiation, and then figure out clear, rational funding mechanisms.