The FDA wouldn't be picking winners and losers. It would be picking scientifically valid lines of research, based on proposals from private companies. These are

The FDA wouldn't be picking winners and losers. It would be picking scientifically valid lines of research, based on proposals from private companies. These are government employees; they're not buying lambos. Do you see CDC scientists getting rich? Today's drug startup company, which thinks it has a product that could be successfully developed, would write up a proposal to apply for FDA funding. If the FDA scientists decide the proposal has merit, they get funding. If the FDA thinks the company is nuts, the company can still self-fund and do the R&D on their own. Either way the normal testing phases are needed for final approval. After that, if the FDA funded the R&D, it gets ownership and licenses out the manufacturing. The startup company can move on to work on something else, or having already been paid by the FDA to do the R&D, they can close up shop. If they self-funded the R&D, then they keep ownership and get to license the manufacturing, just like they do today. (Getting bought-out is just a transfer of manufacturing rights from the startup to the pharma, so that can still happen if the startup retains the rights.) I'm imagining lots of startup companies developing nutritional-based vitamins and herbal products, getting paid to take them through the testing process, and then after approval the startup just moves on to another compound and/or disease. They could run as profitable little businesses for a long time doing this, and we'll get lots of non-patentable but fully tested products on the market that are cheap as dirt for the consumers. We can't get these products today because no one will pay for the testing process, so instead we have unregulated snake-oil products with little to no quality control and vague unverifiable claims about what they're good for.