We just completed the two-day kickoff symposium of the Rice Center for Quantum Materials. It was a good meeting, and the concluding panel discussion ended up spending a fair bit of time talking about the public policy challenges facing basic research funding in the US (with some discussion of industry, but largely talking about government support). Neal Lane is an impressive resource, and lately he and Norm Augustine have been making the rounds in Washington trying to persuade people that it's a dire mistake to let basic research support continue to languish for the foreseeable future.
Over the December/January timeframe, I'm spending time on several grant proposals. Three of them have a priori odds of success (based on past years, dividing awards by the number of initial proposals) less than 5%. Now, obviously longshots have their place - you can't win if you don't play, and there is no question that thinking, planning, and writing about your ideas has utility even if you don't end up getting that particular award. Still, it seems like more and more programs are trending in this awful positive feedback direction (low percentage chance per program = have to write more grants = larger applicant pool = lower percentage chance). Many of these are prestigious center and group programs that are greatly desired by universities as badges of success and sources of indirect costs, and by investigators as sources of longer term/not-single-investigator support. When yields drift below 5%, it really does raise questions: How should we be spending our time, one resource that we can never replenish? Does this funding approach make sense? When the number of potentially "conflicted" people (e.g., coauthors/collaborators over the last four years for every person affiliated with a big center grant) exceeds 1000 (!), who the heck is left to review these things that has any real expertise?