The Aging of Science?

Kevin Drum reproduces the following graphic (from here via here) under the title "The Aging of Science."

The problem is that this isn't about the Aging of Science. First, most scientists don't seek NIH grants, so this would, at best, show the aging of medical research. But really, it represents an intentional and ill-considered policy at NIH of pushing for larger grants to a smaller fraction of applicants. Before you give researchers millions of dollars, they need to have encouraging results, so they need to have established research programs with solid results. So NIH funds older researchers, usually as part of large, multi-collaborator projects with relatively conservative aims.

Younger scientists exist, and indeed exist in ever greater numbers. Graduate programs keep churning out PhDs, but the number of NIH grants has declined pretty steadily. So people take post-doctoral research positions in ever greater numbers, often not just once, but several times. Where people could have begun settling into a tenure-track position by their early 30s a few decades ago, it's not uncommon for people to do 2, 3, or even more 2-year post-docs, keeping young scientists from being able to settle down and start their own research projects until their mid to late 30s.

Here's data from an NSF study of post-doctoral work by PhDs in 2006:

From this, we can more or less reconstruct the trend by reading each group backwards. The number of life science PhDs who never did a postdoc has dropped, while those with one or more has grown, those with 2 or more has increased and now may be dropping (but note that people <5 years out of their doctoral program may yet seek out a second post-doc), and there's no discernible trend among those with 3 or more post-docs. Those trends don't speak to science (or scientists) aging, but to an increase of the age at which scientists become independent researchers, getting their own grants, forming their own labs, hiring their own post-docs.

It isn't clear that this trend benefits science in any way, and lots of reasons to think it's harmful.