Toward More Comfortable Bottlenecks

Jessica at Bioephemera posts a provocative quote about the way we train and employ young people who are seeking careers in academia:

They’re doing exactly what we always complain our brightest students don’t do: eschewing the easy bucks of Wall Street, consulting or corporate law to pursue their ideals and be of service to society. Academia may once have been a cushy gig, but now we’re talking about highly talented young people who are willing to spend their 20s living on subsistence wages when they could be getting rich (and their friends are getting rich), simply because they believe in knowledge, ideas, inquiry; in teaching, in following their passion. To leave more than half of them holding the bag at the end of it all, over 30 and having to scrounge for a new career, is a human tragedy.

— William Deresiewicz, The Nation

The author goes on to bemoan this “colossal waste of human capital” — all those talented young people spending time getting Ph.D.’s, then not eventually landing faculty jobs, when they could be going right into productive careers in some other field.

I’m sincerely unsure what to think about the occasional complaints one hears along these lines. On the one hand, I firmly believe that the grad school/postdoc/junior faculty years should be enjoyable ones, not days of peril and gloom living under a cloud of uncertainty. If there were a way to make the journey easier, I would be all for it. I can think of small ways to do so, and am certainly in favor of such incremental improvements.

But on the other hand, I really can’t think of any sensible major improvements, for a simple reason: there are many people who would like to be academics, and few available jobs. Short of multiplying the number of college professorships by a factor of three or so, I’m not sure how to address the primary cause of this anxiety — the difficulty in getting jobs. If you knew you were going to land a tenured spot at a good place, it would be much easier to bear the indignities of grad-student/postdoc level salaries for a few years. Deresiewicz says, “If we don’t make things better for the people entering academia, no one’s going to want to do it anymore.” But if that were true, why are there so many “highly talented young people who are willing to spend their 20s living on subsistence wages when they could be getting rich”? These seem to be contradictory worries.

Obviously one thing to do would be to dramatically cut down on the number of people who get into graduate school. But that just moves the bottleneck around, it doesn’t change its overall size. And I don’t want to be the one who says to a somewhat-promising-but-not-superstar-quality grad school applicant, “Sorry, I’d enjoy working with you, but we’ve decided not to admit you because in our judgement your chances of eventually getting a faculty job aren’t quite as good as some of our other applicants. So you see, it’s for your own good.” Generally the people who advocate this kind of strategy are the ones who have already been admitted to grad school. (If you’re waiting for Deresiewicz’s solution, here it is: “The answer is to hire more professors.” Well, okay then.)

Again, I honestly don’t know what should be done. I would love to improve the lifestyle and general well-being of students and postdocs in any feasible way. Not sure what that way would be.

81 Comments

81 thoughts on “Toward More Comfortable Bottlenecks”

  1. I agree with this 100%. The only only possible improvement that I can see would be mentoring at the undergraduate level. Some grad students complain they were unaware of the job prospects when they decided to enter grad school, and this sounds like a failure of undergraduate advising. However, in my experience at the U of C, we were all perfectly well aware what we were getting into at the start. As far as I can tell, the system is working perfectly, allowing all of us to compete for these awesome jobs in as fair a manner as possible.

  2. At the graduate student and postdoc level, academic departments should ensure that people are prepared for non-academic careers. This could require more breadth than is usual and developing connections with industry and other non-academic employers.

    The question is how to encourage that. Departments want to have the very best students/postdocs who can make the cut in academia, and to have good support for alternative career tracks make make them feel like they are lowering their standards.

    Tenure track faculty should know what they are getting into.

  3. Mathematician: a mechanism for turning coffee into theorems.
    Physicist: a mechanism for turning the naive hopes of young people into grant applications.

  4. I have a PhD in the physical sciences, and even though I value what I have learned during the last decade, I don’t think it was a great decision for my future prospects. I think the critical mistake that I made was believing that I needed a doctoral-level degree to do interesting work, when in fact there are enormous opportunities out there if one is willing to broaden their scope and can convince employers to invest in potential talent that may not immediately match their current needs.

    As for what academics can do to address the problem, they can start by simply explaining the situation to any undergraduate who expresses interest in graduate school. The Royal Society released a report some time ago demonstrating that (in the UK) only 3% of doctoral graduates in the physical sciences attain senior academic positions, and over 75% eventually leave science completely, most leaving immediately after graduation (http://royalsociety.org/The-scientific-century/). Anyone starting a doctoral degree in the hopes of joining the ranks of academic researchers is either extremely confident or is making a tragically uninformed decision.

  5. I think the key is gracefull exits at each stage of the progression.

    If there are real job opperunities that at least provide some level of reward for the extra levels achieved it makes it much less crushing to have to bail from the academic progression.

    Coming from an engineering background I woudl extend this even to an undergrad level. A number of my class mates did not end on goign to use their degrees directly, but it would have been helpfull for them to have a usefull bail out point where they still would have had the recognition for what they needed in their tangentally related careers. They engineers interested in design and academic work could have had a bit more in depth available then to avoid the inevitabe compromise in the curiculium design to fit both groups.

    As the 2. pointed out, allocating enough time to help the 50%+ of each level who will be bumped out would certainly help keeping the best and brightest coming into the progression, without feeling like it was a bait and switch…

  6. Sean dismisses “hire more faculty” as a silly option, but it’s not quite as absurd as it sounds. Sure, we can’t (and shouldn’t!) try to produce enough faculty positions to employ everyone who wants to do research, but there’s no good reason why the pressure on young academics should inevitably increase over time — there are a lot more Americans, a lot more of us are going going to college, and we’re a lot richer as a nation than we were when most senior faculty around today were hired. Why shouldn’t the academy expand more or less apace with GDP or population growth?

    Funding for the physical sciences has been more or less flat in real terms since the 1970’s or so, and so steadily declining as a share of GDP. Certainly a big component of the pressure on young academics must be a direct consequence of this gradual American disinvestment. This isn’t something we should just accept as an inevitable feature of academia, but it’s also clearly something that can only be changed with a shift in policy at the federal level.

  7. Some general advice and comments, mainly for young people in the US:

    1. Consider whether it might be possible to get out of physics departments and into math departments. The job situation these days in mathematics is much, much better than in physics. This makes for an all-around healthier atmosphere. Whatever excuses are made, people who think that running large doctoral programs in fields with few jobs is a good idea have their own interests at heart, not yours.

    2. From what I can tell, most postdoc salaries these days are not “subsistence wages”. Looking around, I see that the latest solicitations from the NSF for postdoc grants are $60K +$12K /year research money in math $58K + $12K /year research money in astrophysics. Simons fellowships pay $70K/year already. This is significantly above the median income for people of any age, much less young people in their mid-late twenties. If you can’t live happily on this kind of income at this stage of your life, you should probably immediately change fields into finance, which is the only thing that’s going to provide a likely income that will make you happy.

    3. Don’t spend your life in academia if you’re not enjoying it. One of my father’s favorite sayings was “life is not a rehearsal”. I’ve yet to meet anyone who unhappily spent years of their life doing what they thought was necessary to get tenure who turned into a happy person once they did get tenure. If you can’t take advantage of and enjoy what a university environment, with all its problems, offers, best to find another environment where you’ll be happier.

  8. As someone who is about to come out the other side of this (with emotional well-being more or less intact), I agree that there’s not a whole lot that can be done to increase the supply of permanent jobs, and there’s not a whole lot that should be done to decrease the demand for them. (Moving the bottleneck back to the start of grad school would just be moving it back to where it was in the bad old days, right?) Getting a Ph.D. and learning to do independent research are useful, even if one doesn’t stay in academia. It is reasonable to assume that graduate students are grownups who can make their own career decisions, and it is reasonable to assume that they can do the necessary arithmetic to work out their less-than-even chances of winding up in a tenured position.

    However, there is much that can be done to better manage the expectations of incoming students and to give them more resources for finding their way outside of academia. Grad students and postdocs should be given the clear impression that a tenure-track position is only one career outcome among many, they should be encouraged to explore non-academic career options, and they should be given the resources they need to do so. One good place for departments to start would be by holding occasional talks from former students and postdocs who have left academia and moved on to new careers. Maintaining a grad-school alumni network that current graduate students could call on (like most universities do for their undergraduates) would also be helpful.

    I do fear that some departments will find such steps distasteful. There seems to be a general sense within academia that leaving constitutes failure, and those who leave a particular field are rarely spoken of afterward. Breaking down these attitudes would do a lot to improve things, I think. Many people at the postdoc level seem to feel trapped: unhappy with their academic job prospects but afraid to leave their field and unsure how even to go about it.

    Changing present attitudes may turn out to be difficult, though. The research that grad students and postdocs do boosts not only their own academic reputations but also the reputations of the professors who advise them. If the students and postdocs no longer have their eyes on tenure-track jobs, they may be less motivated to work the kinds of extremely long hours that are common in many fields, especially the sciences. (It’s of course not clear that this will cause their research output to suffer, but one might assume that it could.) So encouraging students and postdocs to consider other career options may not seem to be in the interest of their immediate supervisors (X #3 put it more succinctly above). For that reason, tinkering around the edges at the research-group or department level may not be enough. This may be an issue that needs to be addressed at the university level.

  9. So, Sean, what, if anything, would you do for those who leave academia? If you don’t see reasonable things to do for increasing faculty job or for decreasing the number of underpaid workers/grad students you admit into graduate programs, there’s still that ‘waste of human capital.’

  10. From what I saw in the US, one reason for the academic glut is the really long time it takes to get a PhD.
    People who have stuck around for 7 plus years getting their degree feel their time was wasted if they don’t at least try for tenure. Which increases the glut at the postdoc stage and later.
    And conversely, US universities like having grad students around for long periods of time, since its cheap bonded labour.

    Why not shorten the PhD period to 3 years like the UK, with no teaching duties for grad students ? With a PhD at 25, rather than 30, people would probably feel more encouraged to explore other opportunities.
    Also, unis and profs wouldn’t be motivated to hold on to their students as long as possible.

  11. The problem is our society. We do not value education, I might even go as far to say that we don’t value human beings. Sure, politicians will spout about “the greater good” and improving science and math education but public schools are still crumbling and instead of dealing with it they’d rather pass off the responsibility to privately owned, for profit charter schools that are selectively biased against the disabled and other children who need any kind of extra support.
    If all education was treated like a federal job, and had low cost (or even free) tuition, more people would enroll, there would be more need for professors and support staff.
    If we really cared about our society we would invest in academia (as taxpayers) and the return would be advances in all fields especially technology and medicine. Education would be a status symbol, instead of superfluous merchandise that no one really needs (mansions, SUV, high heels, white iphones, etc.).

  12. Sean–

    You write: ‘Deresiewicz says, “If we don’t make things better for the people entering academia, no one’s going to want to do it anymore.” But if that were true, why are there so many “highly talented young people who are willing to spend their 20s living on subsistence wages when they could be getting rich”? These seem to be contradictory worries.’

    But they’re not necessarily contradictory. While the volume of grad students may be large, it’s likely that we’re actually losing a lot of the ones who would be at the very top. The very brightest young minds know how the academic market looks, and have instead been going into finance, generating opaquely overcomplex financial products and high-speed transaction algorithms with famously catastrophic results for the global economic system.

    What’s needed is a lot more training in career development for opportunities beyond the academy, and much better career guidance and information for current students.

    And I don’t know what Woit is referring to, but in high energy physics, postdocs at top schools run at around 50k per year. And that’s after five to seven years of getting 25k per year as a grad student. Considering the amount of time and education involved, that’s crap.

    And, yes, I say cut the acceptance rate to PhD programs in areas like high energy physics, and use the money to fund the creation of more faculty positions. Better to find out that you have no future in academia at age 21 than at age 28 — and then maybe the very brightest ones will decide their prospects are good enough and decide to stick around instead of jumping ship to Wall Street or business consulting.

  13. Why should it ever be assumed that a PhD should lead to an academic position? While doing my graduate work (aerospace engineering), it seemed to me that many of the people in academia preferred the cloistered life of the university environment, not wishing to deal with the real world. Society needs a few people with those inclinations, but it must be remembered that tenured academicians live off the rest of society. We don’t need too many of them.

    Any PhD program should prepare the student for a non-academic position. I agree with BFG: “Getting a Ph.D. and learning to do independent research are useful, even if one doesn’t stay in academia.” It worked for me. In my field, the real challenging work was being done outside of academia, with the academics only consulting and advising and were not really part of the action.

  14. Low Math, Meekly Interacting

    It should be very simple: Accept fewer applicants for PhDs.

    There is, however, a huge financial disincentive to making the process more selective. Once a graduate who isn’t suited for a long-term career in academia leaves their doctoral training, he or she is no longer the university’s problem. While in school, the graduate student does an enormous amount of work for the university while being paid below minimum wage. The funding for the stipend is primarily some fraction of a grant. It’s an iron-clad economic model for growth.

    Which is fine, so long as everyone is aware that those are the ground rules, and they are also made fully aware of the odds. Not only is the tenured position in a distant, uncertain future, it may only be available to you in a place other than the one you were hoping to settle.

    Again, as long as everyone knows what the game is all about, there’s no cause for anyone to complain. But is every prospective graduate student given that needed dose of reality? I don’t think so.

  15. The salary numbers are from the NSF web-site. Numbers in physics may be lower, with institutions able to pay lower rates there just one more effect of the ridiculous ratio between number of people and number of jobs.

    “considering the amount of time and education involved, [50K] is crap”.

    If you’re spending your years in a phd program there not because you love doing it, but because you think you’re going to get compensated later with a higher salary because of this, you’re making a big mistake.

    And if you think 50K/yr income for someone in their late 20s is “crap”, you’ve got a problem with the entire US economy, since that’s above the median income. Academic jobs are a great gig compared to what most people do, why complain that they don’t also pay much more?

  16. There are way too many professors that are past their expiration dates (at least in Physics) in the US and they are still not retiring.

    At UT Austin Physics Department, the average age of the faculty is 59, with one fourth of the professors 70 or older. Those could be positions taken by newer people and that can help with “increasing” the number of faculty jobs. The worst part of it all, is that due to economic pressures, there is no guarantee that every one of those positions will be funded again when those old professors eventually retire which means that instead of increasing the number of positions (or at least keep them constant), the number will go down.

    As far as salaries, how come assistant professors in departments like finance or accounting make over $150,000 but in physics they make about half of that? Has any physics department ever looked at maintaining ties with the alumni that goes to the industry and therefore can make enough money to donate to their departments? It looks like the philosophy in physics is the opposite: if you go to the industry you weren’t good enough to be a professor.

  17. In fact, most people choose to study math or science do not take into account of the job perspective, either because of ignorance or simply because they neglected the matter and focused on personal interest. Curiosity is a very strong motivation that can often trump everything else. As you said, there are only so many jobs and so many people want them, many apparently have to leave. I agree that there are no real easy solution, but there are many small things one can do. For instance, why can’t the research institutions, national labs and universities open administrative and other none faculty jobs to PhDs as part time support? For many people, such support, combined with access to the research environment at these institutions may be sufficient for them to work and produce interesting science. This may be particularly helpful to mathematicians and theoreticians.

  18. LMMI,

    Actually 25K is nearly twice the minimum wage. And life as a graduate student is a lot more pleasant than working at McDonalds.

    Scientist Charro,

    150K/year is an anomaly specific to finance, due to the unusually high pay scales in that field. Physics assistant professors are no worse paid than the average for all fields. Expecting everyone in academia to be paid on financial industry pay-scales is kind of absurd.

  19. Pingback: Hard for Scientists to Remember Science Isn’t Just for Them « Pasco Phronesis

  20. Re: the salary after your life as a PhD.

    Right now, in the middle of my 3rd year as postdoc, I pull around €27000/year after taxes —that’s about $38000 (I work in the Humanities). I don’t complain about money, but that’s mostly because I live in Berlin, which is one of the least expensive places to live (i.e., housing, groceries, and other basic living expenses for both me and my still-PhD-wife barely take 50% of my salary). Things would be different in a different place. I don’t think the amount of money I make would be enough in, say, Paris, London, New York, San Francisco, etc

  21. A few comments:

    1. I agree with the commenters who say there should be better preparation/exposure to jobs outside of academia. While university career centers are at least trying to provide some information and contacts regarding industry jobs, consideration of non-academic jobs is frowned upon (or the existence of such jobs ignored) at the departmental level, especially by more senior faculty. You’re seen as being “not good enough” if you so much as think about careers outside academia. This attitude needs to change.

    2. While prize postdoc positions often pay relatively well, non-prize postdoctoral positions don’t. There’s a huge disparity in pay for postdocs. Let me say that I earn FAR less than the prize fellows others have mentioned; I am pretty sure I could have gotten a much higher-paid job had I been willing to cast the net far, geographically speaking, but I have a two-body problem. I took a lower-paying job so that I have the privilege of living under the same roof as my husband. (see point 4 for a related comment) Regardless of my personal circumstances, the point is that most postdocs earn significantly less than $60-70K/year.

    3. Many top departments are on average quite old. What fraction of the faculty at Harvard or Princeton or Caltech are under the age of 50? Over the age of 60? There will need to be some serious hiring soon, but that in large part depends on the willingness of senior faculty to go emeritus.

    4. My big (possibly biggest) gripe with the current system of career trajectories is that it selects for a certain personality and attitude, and a certain taste for problems to work on. Because the competition for faculty jobs is so fierce, one needs to not just be outstanding but have some degree of single-mindedness (bordering on fanaticism for some) and minimal “distractions” outside of work to achieve the goal of getting and keeping a faculty job. It selects in some cases for “true believers” who would consider anything less than a faculty position to be failure (that said, there are many awesome people who got jobs this year who are not of this ilk). While I think both men and women suffer because of the competition, it is worse for women because the period of most intense competition is also the time in which it is really preferable to start having kids should kids be something you desire. I HATE the fact that if I got pregnant right now, it would essentially be academic-career suicide unless I basically resigned myself to not ever sleeping again and cut out everything from my life that wasn’t work or baby. Moreover, most women physicists I know have two-body problems. The two-body problem is an issue that most places can’t, won’t, or have a hard time dealing with. The University of California has some great programs, but they tend to work best for couples in different departments. Of course, the two-body problem becomes much stickier once kids are involved, which again disproportionately affects women.

    This intense competition also means that people tend to have more conservative tastes in the sorts of problems they tackle. If you are being judged largely on the number of papers you produce per year, are you going to work on a riskier or trickier problem or work on a series of smallish low-risk problems? I think that is really bad for science.

  22. I concur that the first solution is to reduce the number of openings in PhD programs. It’s much easier for an intelligent student to miss the PhD boat and to retrain towards the industry at 22 than at 35.

    In addition, we have to abolish the postdoc institution. There should be two classes of people in science: PhD students and tenure-track professors. If we can’t provide tenure-track jobs to all PhD students interested in academic science, at the very least we shouldn’t force them to postpone the industry vs. academia decision till the age of 35.

    And we should think hard to understand why it is that PhD in sciences has no value in the real world. We wouldn’t be having this discussion if science PhD’s could get industry jobs for which their education & research experience were directly relevant. We wouldn’t be having it if science PhD’s could apply their education in positions which aren’t funded by other students or taxpayers. It’s one thing that people can’t get real jobs when they graduate with degrees in particle physics and astronomy: that’s perfectly understandable. But somehow things got so broken that even PhDs in life sciences (eg biochem) find themselves mostly useless when it comes to doing any real work, because pharma companies don’t really want them.

    Re: Peter Woit: median earnings for male full-time, year-round workers in the US is $45,363. And the median male full-time, year-round worker in this country never even went to college.

    Average age at first postdoc is 28 or 29. I’d have to say that making $50,000 for a 30-year-old with a graduate degree and the kind of intelligence we expect to see in physics & math PhD’s (most likely above the 95th percentile by intelligence in this country?), even if it’s not really subsistence wages, it’s still pathetic and wrong.

  23. Alright then, all of you commenters providing some measure of solutions (or at least adding in further gripes), what do you recommend those of us “in the trenches”, so to speak, do about it? I’m about to become a first year graduate student at a university that is pretty well respected in its field and worry that the instinct of self-replication from professors will take another victim. A lot of the complaints so far are so far-reaching that it’s way, way too much for any one person to be able to change the whole thing, so what are some things that if, say, every new grad student this year started doing that would help to put us on a more sustainable path that’s fulfilling for everyone?

  24. I think Cosmonut has mentioned the fundamental problem in the US, which has little to do with junior researcher salaries and attrition rates. Neither of those would be a problem except for the fact that the system in the US has a vested interest in keeping junior researchers hanging on for a decade. If you spend 3 years getting a PhD as in Europe, you come out in your mid twenties, which is a perfectly sensible investment of time. Even after a couple more years in a poorly paid postdoc, if you find yourself job-hunting in your late twenties, it’s not a big problem (as long as you went in with your eyes open). It was a reasonable gamble that didn’t pay off.

    But the PhD program in the States that takes seven years on average is ridiculous, and the reason for it is quite clear: the cheap labor both in the classroom and the lab. Of course, it’s the students’ choice to let themselves be exploited in this way in the hope of making it to the top of the academic pile later. But as some have mentioned above, that’s not selecting for the best research talent. And of course to get rid of all that cheap labor and instead hire more profs would significantly raise the price of undergraduate education. But that’s just moving the costs around. Right now, those same costs simply come out of society elsewhere (e.g. the missing tax dollars those people aren’t paying). And one reason for the need for cheap labor is the fact that we’re sending way too many people to university due to some misconceived notion that any university degree contributes to the “knowledge economy”. (But that’s another debate.)

    So to my mind, the bottleneck in the US does need to be moved back and shortened. Things seem far more sensible in Europe, in my experience. There’s still the same competition for permanent jobs, but it’s resolved more efficiently.

Comments are closed.

Scroll to Top