Only in Canada

I encountered his thought provoking article on a science blog:

Gordon, R. & B.J. Poulin (2009). Cost of the NSERC science grant peer
review system exceeds the cost of giving every qualified researcher a
baseline grant. Accountability in Research: Policies and Quality
Assurance 16(1), 1-28.

Basically, the authors point out that (given the denominator and the success rate of NSERC grants) that we might all be better off with just awarding a fixed award to all applicants. It seems to make sense only in the Canadian context where high success rates, low indirects and a lack of soft-money positions might actually make this approach viable.

There are definitely some well founded concerns about this modest proposal, but the mere fact that it has such a strong argument underpinning it focuses us to reflect on the peer review system. Peer review is a massive burden and can lead to a "herd mentality". Some peer review is essential as you can't evaluate top experts in any other reasonable way (at least in the context of awarding grants).

Is there too much redundancy in the system?