I disagree -- the funding isn't ROI based at all. Heck NIH doesn't even really audit how well the funds were spent, how could they? They don't even really assess if the research had impact, save for counting journal articles and impact factors, which are in themselves poor proxies for quality of work (and easily gamed).