The Causes and Impacts of Overpopulation

Let’s start with a raw number. The annual production of new PhDs is occurring at about 250% the annual hiring rate, maybe a little more. So the aquarium is getting pretty overstocked. How do you make yourself noticed among all the pretty fish? (It’s not unlike bachelor’s degrees; now that so many people have one, pretty much every employer regards it as a baseline for their job, even if the job itself may not have changed in the fifty years since your high-school-grad grandma did it.)

The National Academies of Research have done an interesting (and massive) study in doctoral education, attempting to rank all American research doctoral programs in all known fields. The methodology is complex, and seems reasonable, but one of the things to consider before we get to the findings is just how large the’ve discovered the enterprise to be, with over 4,800 different PhD-granting programs located at about 210 institutions:

  • Agricultural Sciences (of various sorts): 317 doctoral programs
  • Biological Sciences: 989
  • Health Sciences: 189
  • Physical Sciences (including Math): 916
  • Engineering: 798
  • Social Sciences: 930
  • Humanities: 866

So we know from economics, for instance, that the good faculty jobs go to only those who got their PhDs from a handful of schools: 25% from two schools (Harvard and MIT), 50% from the top eight. There are 117 economics doctoral programs in the US. If your program is merely good, you’re screwed.

And that was a study from ten years ago! As the oversupply of doctorate holders increases, it’s not merely that those with degrees from lesser schools will get teaching jobs at lesser schools, because those schools too are now receiving faculty applications from graduates of the uppermost programs who can’t all teach at R1s themselves. So smaller and lesser schools all the way down are also selecting faculty who graduated only from the elite programs, not from the merely excellent. I know someone who just got a teaching job in an undergraduate biology program who herself has a PhD from a top-5% doctoral program and served a six-year postdoc with one of the NIH programs. That’s like having Thomas Keller in charge of putting Pop-Tarts into the toaster… it seems like it might be boring before long. So they have to spend money on research infrastructure to keep her interested, drawing the school away from its core undergraduate education mission. (In another decade, she’ll probably have a doctoral program of her own).

Here’s a couple of examples, let’s leave them un-named…

  • a public highest-research university (R1) with eleven doctoral programs. According to the NRC ratings, not one of those eleven programs ranked in the top half of its respective discipline; seven of the eleven were in the lowest quartile of their fields.
  • a public high-research university (R2) with seventeen doctoral programs, again not a single one in the top half of its field, thirteen of the seventeen in the bottom quartile.

Why are those schools still allowed to offer doctoral degrees at all? What exactly are they selling, and to whom? What exactly do its students believe themselves to be buying?

Compare these with:

  • UC Berkeley, with exactly fifty doctoral programs: only one is below the 50th percentile ranking, and half of them are in the top 10% nationally in their fields.
  • University of Michigan, sixty-five doctoral programs, four in the bottom half, eleven in the top 10% (go to Michigan for philosophy, psych, and math)
  • Columbia, forty-seven doctoral programs, ten in the bottom half and ten in the top 10%. There may be some productive winnowing to be done here.

You have to know these things. You have to know that a school’s general reputation isn’t the same as its doctoral reputation, which varies from program to program. And you have to know that most schools have no vested interest in telling you any of it if they want you to go there.

The simple existence of a doctoral program does not mean that the doctorates issued thereby will be recognized as equal currency. Your degree will be read on your CV in components: I have a PhD in [discipline] from [university & department], studying under [dissertation advisor]. Your job chances are dependent on each of those three terms, far more than the fact of your graduate GPA.

There’s a lot of talk about restricting the numbers of people who get PhDs each year, and most of it is framed in terms of reducing the numbers of entrants (again putting the burden onto the individual student or prospective student). Why don’t we talk in terms of putting the burden onto the institution? Why should we have 4,800 issuers of the PhD, when we know that only a few of those will offer productive gateways to faculty life? Why shouldn’t we make each department prove itself every few years to keep its doctoral license?