"I never trust anyone who's more excited about success than about doing the thing they want to be successful at." -XKCD
Suppose there were some cultural group that made more than their share of positive contributions to society. Suppose members of this group overwhelmingly believed in evolution, liberal democracy, and [insert your favorite meme here]. Suppose, for the sake of argument, that they bicycle to work, insulate their homes, etc., making their per-capita consumption of nonrenewable resources less than average for their society. The only thing is, their numbers increase twofold with each generation, so the total consumption of resources by this group increases, despite their low per-capita consumption. Is this group a problem?
The group, of course, is people with PhDs in science. The PhD population will increase until, as Malthus wrote:
"the actual population [is] kept equal to the means of subsistence, by misery" [or, perhaps, by some form of birth control?]
That misery has arrived. NSF receives ten times as many grant proposals -- at least five times as many good grant proposals, in my experience -- as they can fund. An assistant professor who can't get a grant may not get tenure, even if his or her teaching and research potential are excellent. One professor's failure to get tenure may create a new opening, but each opening attracts many qualified applicants, eager to trade postdoctoral misery for a chance at a tenured faculty position.
Can we end this misery?
The most commonly proposed "solutions" require allocating more money to science. If only there were more grant money, every high-quality proposal would be funded, right? On the jobs front, one frustrated postdoc suggested that postdocs should be given "permanent research staff positions." These would presumably be paid more than postdocs and their greater job security might make them less motivated, but she argues that their greater experience would make them worth their higher cost. Maybe. But would creating such positions solve the problem of too many postdocs and not enough jobs?
Only temporarily. There have been more PhDs than faculty jobs for decades, but students keep signing up for PhD programs. Improving career options for science PhDs would tend to attract even more students, worsening the PhD surplus a few years from now. Similarly, increasing funding available for grants leads (immediately) to more grant applications and (in a few years) to more faculty competing for grants.
The latter point is interesting. How does grant availability lead to increased faculty numbers? Aren't faculty hiring decisions based on teaching needs? Not at major research universities. Research grants typically include "indirect costs" or "overhead", which can exceed 50% of the actual or "direct" costs of the research. Indirect costs include heating buildings, clerical staff, journal subscriptions for research libraries, etc. But the university would have to pay most of those costs, at least in the short term, with or without any particular grant. In other words, current overhead rates exceed the short-term marginal costs of grants to the university. (In the long term, getting lots of grants may require new buildings, etc., but university administrators mostly think short term. Hiring more grant-getting faculty than would be needed for teaching gives them money to support new initiatives that strengthen their next job application.)
So cutting overhead rates on grants -- but not so low that universities start rejecting grants -- would be the simplest way to limit the professorial population explosion. Assuming constant total funding, cutting overhead would also let them fund more grants -- maybe even most of the best ones? -- so professors could spend more time doing research and less time writing grants. If our only goal were to reduce the increase in faculty numbers, we could do that by cutting total grant funding, but do we really want professors to waste their time writing 20 proposals for each one that's funded?
Note that I'm not arguing against an increase in research funding. Basic and applied research can lead, directly and indirectly, to more-sustainable agriculture, better ways to protect environmental quality, more-efficient industrial processes, and improved medical care, with a much better benefit:cost ratio than such alternative societal investments as propping up corrupt dictatorships. But allocation of funds for research should be based mainly on the potential direct and indirect benefits of that research, not on how easy it makes life for professors. The only exception to this principle is that we want to keep research careers appealing enough to attract creative and energetic people. (While we're at it, any ideas for how to attract top, rather than bottom, students to precollege teaching? Sabbaticals, maybe?)
Elsewhere in this blog (or am I thinking of my forthcoming book?), I've also argued that granting agencies should give preference to labs that don't already have a grant. This is because a professor who already has a grant can usually put together a more impressive grant application, with lots of preliminary data, even if his or her ideas are somewhat less creative than someone who doesn't have a grant yet. (Now that I'm on three grants, I'm reconsidering this idea!)
Even if we slow the growth in number of professors, each professor is likely to have more than one PhD student over a career. I got criticized, at Davis, for not taking a PhD student until I had an applicant I thought would eventually be successful even in a tight job market. (She was!) So the PhD surplus isn't going to go away. If you're an individual considering grad school in science, already in grad school, or starting a postdoc, how should the scientist surplus affect your individual decision-making?
First, don't enroll in (or stay in) grad school unless grad school itself appeals to you. That way, even if grad school doesn't lead to the career you wanted, you won't consider those years wasted. Also, if you enjoy grad school, that means you enjoy working long hours in the lab and reading and writing scientific -- exactly what you need to do to have a shot at a research career. See this earlier post.
If you're already a postdoc, I've noticed a very strong correlation that may interest you. Postdocs that aren't in the lab on weekends don't get research jobs. Those I always see in the lab on weekends have gotten jobs as professors or research scientists. This may not be true always or everywhere, but it makes me question the statement (from the postdoc mentioned above) that "a lab-head position requires a strong publication record, which can be as much about luck as skill and hard work." Maybe, though, the postdocs that are working weekends are doing so because they're getting exciting results, rather than getting exciting results because they work harder.