An interesting news item came up in today's Science (p. 1657) regarding a controversial new policy being put into place by the U.K.'s Engineering and Physical Sciences Research Council (EPSRC).
"After rejecting options such as charging for proposal submissions or placing quotas on institutions, on 12 March the council announced a new policy to ban from any EPSRC funding consideration, for a year or even longer, any principal investigators who have "Three or more proposals within a 2-year period ranked in the bottom half of a funding prioritisation list or rejected before panel [without review]; AND An overall personal success rate of less than 25%." EPSRC notes that this policy should exclude about 200 to 250 people and is retroactive: Letters to those excluded go out on 1 April, and their proposals won't be considered after 1 June."
Now I have to confess that I don't know much about EPSRC and the number of grants that one might be expected to submit there in a two year period, but if NSF were to put this policy in place it would completely change the way I apply for funding. As of right now, I have two different grants under consideration at NSF and I plan to submit another in July. I submitted my first one last July, before I arrived here, and that one was not funded and did not receive very high ratings because of the lack of preliminary data. If the resubmitted version of this grant (submitted in January) and the most recent grant that just went in (two weeks ago) were to be similarly ranked, I would essentially be shut out from funding for the following two years. Rather than getting feedback on different proposals that as a new investigator, I would instead either have to change my research focus and apply to a different agency (a scenario where having preliminary data would be unlikely) or be looking at shutting down the lab and finding work elsewhere.
I can understand that certain researchers might be frustrating the panels with poorly written proposals, but there is a very high chance of new investigators getting caught in the cross-fire here, unless the "overall personal success rate of less than 25%." protects new investigators. This seems like a dangerous precedent to start.