> Promotional spend has a positive ROI (or you wouldn’t do it).
That's incorrect; promotional spend on prescription drugs is the globally suboptimal result of a prisoner's dilemma where the prosecutors are going around saying "hey, that guy over there snitched, so should you". There is absolutely no guarantee that ROI is positive.
> Who do you think provides capital for drug development? The Grenadier Guards?
Most of basic drug development happens in labs e.g. at universities that are funded mainly by public sources.
Per the latest PhRMA member survey, drug companies spent about $13 billion per year on preclinical R&D, which was about 15% of their total R&D spend. Meanwhile the NIH alone provides about $40 billion per year for preclinical research. So funding for actually finding new drugs and therapies is dominated by other funding sources than drug companies.
If you look at e.g. Pfizer and J&J, more than 85% of their revenue comes from drugs that were invented elsewhere. Just to clarify, I'm not saying clinical trials isn't a necessary and expensive thing. But in large part it's not developing drugs, it's checking that there are no big side effects and that efficacy remains what you saw at small sample sizes.
For instance mRNA vaccines were invented in publically funded laboratories. Monoclonal antibodies that work against inflammation is another example. And of course CRISPR, which is very promising.