Too many PhDs? That’s anybody’s guess.
And thank you for kicking things off, Chemjobber! Remember, tomorrow we go to Paul for why tenure is teh suck, and Thursday the Mighty Matt enlightens us on what science policy can do to help us fix the employment mess.
So in yesterday’s post, CJ talked about the present and future of industrial jobs in chemistry. “Chemists are facing lower-than-average hiring and an unemployment rate that is the highest in 20 years at 3.9% (according to the 2009/2010 ACS Salary Survey),” CJ said. And it came up in the comments, like it always does: how much of this can be blamed on new PhDs coming into the marketplace? Are we really overproducing chemistry PhDs?
Fortunately, the numbers of doctorates produced each year in many fields of study are tracked by the NSF, who puts out a short report in November and a longer one in December. The most recent one says that in 2009, there were 49,562 doctorates awarded in the United States in all science and engineering fields. That’s up 1.6% over 2008, an increase that’s almost totally due to a rise in women getting these degrees. This is pretty much irrelevant to this discussion, but I thought it was interesting. Moving on.
The NSF breaks down these numbers into separate fields of study. And look, I took their data and made graphs!
As you can see, the number has fluctuated a bit over the last ten years, but the general trend has been up. Last year, there were 2,398 PhDs awarded in chemistry and 859 in biochemistry. The chemistry number is up 6.2% from 2008 (2,247), while the number of new biochemists is down 4.5% from the year before (898). There was a small spike of chemistry PhDs awarded in 2006 (2,362), and biochemists seem to have maxed out (minimally) last year. Overall, the number of fresh new chemists has grown 11.0% since 1999 (2,132) and biochemists are 11.6% more popular than they were ten years ago (759).
So, what does this mean? Well, that there are more chemists now than there were 10 years ago, but I think we could have all guessed that. The real question is, how many jobs have there been available in since 1999?
There’s a problem with that question. Namely, that there isn’t an answer to it. The Bureau of Labor Statistics says that in 2008, chemists held 84,300 jobs. But that’s all chemists, BSs, MSs and PhDs together. Besides, that’s people holding jobs. How many chemistry jobs open up each year? We know that CJ tracks the ads in C&E News, but of course not everyone advertises there. And not all jobs are even advertised anywhere; supposedly, only about 60% of openings ever blink in the sunlight. And how would you count alternative careers? Does it count as a “chemistry job” if it’s in policy, communication, or law?
These reasons, among others I’ve surely missed, are probably why these data don’t exist. And without solid data about the number of jobs out there, we can’t possibly say if there are too many PhDs being awarded.
But what we can do is guess. The NSF tried to hack away at it a bit, by asking new PhDs what their job plans are. They call them “Definite Postgraduation Commitments,” and go on to say, “The proportion of doctorate recipients reporting definite commitments is an indicator of the overall strength of the job market for doctorate recipients and the availability of positions relative to the supply of new doctorate recipients.” Unfortunately, in this case they don’t separate the chemistry numbers out from the rest of physical sciences. So the data graphed below includes all kinds of physics, astronomy, math, and earth, atmospheric and oceanic sciences.
Please note the data range on the side before you get all crazy. But still. That looks like a pretty strong upward trend. To put things in perspective, the number of PhDs awarded in the physical sciences in 2004 was 3,350. Of these, 2,116 had definite commitments upon graduation. That’s 63.2%. In 2009, 2,829 of the 4,289 doctorates awarded had a job or what-have-you lined up. That’s 66.0%. Overall, 2.8% more of new PhDs had some kind of paycheck to turn to last year than five years ago. That doesn’t sound like they’re having a harder time finding jobs to me.
Oh yes, I can hear you shouting in the back there. “But I bet more of them are taking post-docs!” Well, I hope you didn’t bet too much on that, because you’d lose. It’s about the same. A tiny bit less, even.
Very tiny. In fact, all those bar sizes up there look about the same. Now I do remind you, those numbers are for all physical science PhDs. Would it look drastically different with only the chemistry data? Don’t know. Wish I did.
The BLS also puts out an employment outlook for chemists. Their 2008 report states “Job growth is expected to be slower than average for all occupations.” That’s for the next ten years, and they mean that jobs will increase anywhere from 0 to 9%. I went and dug up the same numbers from ten years ago (not an easy feat on a government website), and found that in 1998 they said that job growth would be average (meaning it will increase 10 to 20%) up to 2008. I think they flubbed on that one, since they say that there were 96,000 employed chemists in 1998. Their 2008 report says that chemists plus materials scientists together held 94,100 jobs that year. (They didn’t differentiate between the two in 1998.) That’s a loss of 1,900 jobs, or about 2%. Of course, it would have been nearly impossible to predict the slow keening death of the pharmaceutical industry, but that’s precisely why you can’t bet your bottom dollar on predictions like this.
What’s more interesting is their data on salaries. My colleague at CEN, Bethany Halford, is writing a longer article on this topic for inclusion in next month’s magazine. She suggested to me that a lot can be learned from watching how much chemists get paid. The BLS puts data out on this as well. In 1998, the median chemist salary was $46,220. If we adjust this for inflation, we get $61,051 for 2008 (I used this calculator). So if demand for chemists decreased, the 2008 mean salary should be less than this. The BSL reported the actual number as…$66,230. That’s a raise of 7.8%. According to that, chemists are still in demand. [ETA: Bethany mentioned in the comments that the starting salaries of chemists has fallen, suggesting that demand for chemists has decreased.]
So, what gives? All hard data we have suggests that new chemistry PhDs are doing okay in the job market [but being paid less, see above note], and that chemists are still being paid the same or more than they were ten years ago. But does this match reality? I know a lot of my colleagues in grad school are having a very hard time finding jobs. This is anecdotal, though. And it seems that, in the past, it’s been quite common for established scientists to point that black finger of blame to those newly anointed in the field when times get tough. Like here. Or here or here or here. People have been saying there are too many PhDs for many many years. Is it truer now that it was then? Is it true at all?
You know what I’d like to see? Some of that NSF data broken down by subfield, ie organic, analytical, materials, etc. Especially the data on definite US commitments. Because I think Chemjobber had a very good point when he said:
“Today’s hot field could be tomorrow’s old-and-busted outsourcing minefield. Certainly medicinal chemistry was the hot field, ten years ago. This is a fundamental problem with professional scientist training in the US. While the market will always be clamoring for trained scientists today, the labor supply is close to a decade behind.”
Which is similar to the point Derek Lowe was making here. Do we have too many PhDs being produced? Yes. And no. There is a glut of medicinal chemistry and synthetic organic people, probably. But there aren’t enough electrochemists or people to work on new energy storage systems, like CJ pointed out. But how can we possibly know what kind of scientists we’ll need in twenty, ten, or even five years? We can’t. It sucks but there it is.
So…what can we do about it? Well, not much. I think it’s an obvious answer that incoming students should be aware of the job market before they even think about grad school, but well…that’s likely not going to happen. Did any of us know what we were getting into when we started grad school? I sure didn’t. And I think most incoming grad students don’t have the brains or foresight or whatever to do a lot of research into this as they’re starting out. Hell, a lot of students go to grad school because they don’t really know what else to do when they get that undergrad degree.
Is it the adviser’s job to instruct a new student on job prospects? I think we could debate this until the cows come home to roost, but it’s not going to do much good, simply because it seems to me that advisers would NOT see it as their job. For one, I wonder how many professors pay attention to the job market at all. (They already have jobs, after all.) For two, apart from the student being nuts or the PI already having a maxed out group, I can’t imagine a prof that would turn students away, since we all know that students = hands in the lab = more research being done = more papers = more stuff to put in grants = more dollars = tenure = more students and so on and so on. (Please understand that I’m not saying that ALL advisers are like this, but many are.) So a prof saying, “Gosh, don’t join my group! You won’t get a job when you graduate, because the field is so tight”? Riiight.
So no to students figuring out job prospects on their own, and no to advisers filling them in. What’s left? In an ideal world, I think PhD granting programs should each have their own career counselor, at least part-time. That’s obviously not going to change the job market, but it would at least give potential grad students a heads up about what they’re about to subject themselves to. (Er, in one respect, that is.) Or even if it’s worth their while at all.
What I don’t think will work is putting a limit on the number of PhDs given out every year. While it may curb overproduction in one subject, it will likely lead to shortages in another. Even if there was some almighty overseer that would say, “We will allow blahblah number of PhDs in X this year, and some other number in X,” it would be a total crapshoot to get those numbers right. And you’d end up with out of work chemists, just like now. Or people trained in the wrong area.
So there’s one thing I kinda skipped over here. Since new PhDs are having about as hard a time getting jobs as they have in the past five years (according to the NSF numbers), and unemployment for chemists is relatively high, are new PhDs forcing older, more established scientists out of their jobs? Unfortunately, the data we have for that is completely anecdotal, therefore not citable, therefore not trustworthy. So this will ultimately end up, like most of the things I’ve brought up here, a guessing game.
Okay, Readers. I’ve said my piece. Now say yours in the comments. A note: if you haven’t commented here before, your comment has to be approved before it shows up. (It’s the software.) So please be patient if you don’t see it right away. I’ll be approving as fast as I can!
Please be sure to check out Bethany Halford’s CEN story in January. She’s been interviewing actual people, so it should be pretty interesting.
And with that, take it away Paul!