Scott Aaronson discusses the book The Truth About The Drug Companies: How They Deceive Us and What To Do About It, by Marcia Angell:
Like many in the US, I once “knew” that drug companies have to charge such absurd prices here because otherwise they wouldn’t be able to fund their R&D. This book reveals the hilarious truth about what drug company R&D actually consists of. My favorite examples: coloring Prozac pink instead of green, marketing it for “premenstrual dysphoric disorder” instead of depression, and charging three times as much for it. Inventing new drugs for high blood pressure that are less effective than diuretics available since the 1950’s, but have the advantage of being patentable. Proving in clinical trials that a new drug works better than an old one, as long as you compare 40mg of the one to 20mg of the other.

The book paints a picture of the pharmaceutical industry as, basically, an organized crime syndicate that’s been successful in co-opting the government. It trumpets the free market but depends almost entirely for its existence on bad patent laws that it helped write; it bribes doctors to prescribe worse expensive drugs instead of better cheap ones; it waits for government-funded university researchers to discover new drugs, then bottles them up, makes billions of dollars, and demands credit for its life-saving innovations.
This is consistent with what I hear about the pharma business from my Slovenian colleagues who have entered the medicine and pharma industries.

We know that the U.S. patent system is utterly broken for software. It may be that it is utterly broken for pharmaceuticals as well, and we are just being persuaded how invaluable it is by people who grew rich exploiting the exclusivity it gives them while really not making much of a contribution to justify those privileges at all.

I'm strongly a libertarian, but I've argued before (in Slovenian) that basic research is most effectively done as a community effort, and not by individual companies. This is because the interest of individual companies is to find a local optimum in a given market, not to disrupt the market; but the greatest common good may well be found by disrupting the market and finding a global optimum.

Furthermore, companies are unmotivated to invest in basic research because when basic research yields results, it is often difficult to control who profits from it; the research may take so long and be so expensive and uncertain that it may well be beyond a single company's or manager's investment horizon and appetite for risk; and the revelations it uncovers may be so fundamental that it's difficult to keep it under wraps long enough for a single company to profit from it.

In my belief, evidence shows that the combined individual actions of companies and consumers are good at finding local optima, but because of the Nash equilibrium, they aren't effective at breaking out of a rut and discovering entirely new global optima.

This is why I think that the most basic of research - and now including perhaps research into drugs - is best financed by the community, and the results made public for any company to use.

Meanwhile, we should probably just abolish patents, which give an unfair leverage to people who generally seem not to have earned the privilege at all. At least in software, there seems to be hardly any use for patents other than as a way for:
  1. companies to muscle out potential rivals while they're small,
  2. patent trolls to extort money out of legitimate business, and
  3. large companies to practice mutually assured destruction.
There may yet be a benefit of patents for startups in the bio, physics and pharma worlds, but there is none in software; and the overall value of patents in the big pharma business seems to be questionable, too.