Postdoc Emeritus

An aspect that is widely perceived as “problematic”, in the way the career path of a scientist in North America is currently structured, is the relatively long period of uncertainty and precariousness between the obtainment of one’s doctoral degree and the first potentially permanent employment [0]. That time, which in some fields of inquiry often approaches or exceeds the decade, is typically spent in term, so-called postdoctoral appointments.
This state of affairs has prompted cell biologist Jennifer Rohn to write an editorial published on the prestigious Nature magazine, calling for a restructuring of what she calls a “broken system” of traineeship in the sciences.

The problem
Her point is basically this: the way the system is set up now, there are always going to be more excellent postdoctoral researchers than available permanent research jobs. The talent and experience of many senior postdocs who are unable to land a permanent position, should not be let go to waste by forcing these individuals into different professions, something which seems harsh and unfair, especially so late in their lives. Many of them are not in their predicament due to lack of ability — sheer luck has a lot to do with it, or sometimes they simply are not interested in becoming a leading investigator. However, they can and do still prove a valuable asset in support of someone else’s research program, as “senior staff members”, witness the very fact that the Principal Investigators (PIs) who employ them are willing and eager to renew their positions indefinitely, as long as the funds needed can be procured.

The proposed solution
Her suggestion is therefore that an official, alternate career track be established in the sciences, essentially regularizing a permanent staff position that de facto already exists in most places (typically referred to as “Research Associate”), except that positions ought be permanent and not subjected to the uncertainty of the grant renewal “roulette”, as they almost invariably are now (on this side of the ocean anyway).
Naturally, because science funding is a zero-sum game, the monies to make that happen would have to come at the expense of something else. Rohn proposes an overall reduction in the number of term postdoctoral positions, in order to set aside funds for those fewer, permanent (“hard”) ones. One way to attain that goal would presumably be that of cutting research grant money earmarked for postdoctoral training. According to Rohn, this would have the beneficial effect of diverting off the research track many, most doctoral recipients whose odds of landing the coveted permanent research jobs are dismal anyway, re-directing them to a different professional path when they are still young and marketable.

I am in — sorry, we are full
This is not new, of course. Calls for limiting the number of students admitted every year in science PhD programs have been made repeatedly over the past two decades, based on a supposed “glut” of science PhDs (but don’t ever show science PhD employment statistics to anyone who tells you sad tales of ex-postdocs panhandling, or driving cabs). To her credit, Rohn herself does not advocate reducing the number of PhD graduates. On the contrary, she acknowledges that they “are useful to society, and are eminently employable in non-research fields”. As mentioned above, Rohn would instead place the “gate” at the postdoctoral level, i.e., significantly reduce the size of the postdoc population, by “reducing the number of trainee postdocs infused into the system” — in other words, let fewer of them in to begin with.

Two things are worth mentioning right away: the first is that the career path of a scientist is long, difficult and precarious everywhere on this planet, not just in North America, and that has little to do with postdoctoral positions, and a lot to do with the fact that there are always going to be more applicants than jobs — it seems impossible to escape that bottom line. In Italy, for example, postdoctoral positions did not (formally) exist until not so long ago, but I do not think that a single Italian wannabe scientist would take issue with the contention that being a postdoc is way better than working odd jobs for years, waiting for an entry-level position to be advertised at some university.
Secondly, Rohn’s “solution” is ultimately no solution at all. The final outcome would still be the exclusion from the scientific career path of the vast majority of those who wish to pursue it. The difference is that in the present “cold hearted” scheme, wannabes are given a chance (not always a fair one, to be sure, but a chance nonetheless) to “show what they got”. The “winners” emerge from what Rohn refers to as the “tournament model”, wherein many are let in, in excess of the demand, and “practitioners [are] pitted against one another in bitter pursuit of a very rare prize” [1].
The “more humane” approach advocated by Rohn consists of simply denying most of them the opportunity to compete to begin with, thereby saving them from future disappointments. The fortunate privileged few allowed to become postdocs, would be essentially guaranteed a career in research — some of them as PIs, whereas the ones who should turn out to be, well, not that good after all, or not to have what it takes (apparently that still can and does happen) as “permanent research staff” [2].

Ten years ?
I have written about postdoctoral appointments already a few times, on this blog. Specifically, I have expounded my opinion as to why postdoctoral appointments are generally a good thing, both for the scientist in training as well as for the hosting investigator. I have also expressed my (strong) feelings about the need for time limits.
An aspiring researcher ought not spend too long a time on postdoctoral appointments. How long is “too long” probably varies slightly across disciplines (though I would be surprised if it really varied more than “slightly”), but I think that in any field where postdoctoral researchers operate, there exists a more or less understood, “optimal” duration, beyond which one’s employability and appeal as a PI in a research setting, decrease very rapidly with time. And that is what a postdoc is for, want it or not. It is a training position, a stepping stone, part of a selection process, not a point of arrival.
My personal “rule of thumb” is that after five years as a postdoc and no permanent research job, one should very seriously consider switching to another career [3]. I also happen to think that the supervising PI should actively, concretely try and help the postdoctoral advisee make that switch.
PIs who, for personal gain (real or perceived), inability to serve as effective mentors or any other reasons, keep on renewing postdoctoral appointments beyond their usefulness, misuse funds and do a serious disservice, first of all to the trainees themselves, but also to science as a whole. They do so by allowing junior scientists unable to make the transition to PIs to keep their lives on hold, postponing facing reality, while depriving freshly minted PhDs of the same training and educational opportunities granted to their older colleagues.
There is where the problem lies, not in the lack of “permanent temp” positions.

How does that happen ?
My repeated observation is that people “settle” in postdoctoral positions, out of
a) sheer stubbornness, refusal (or perhaps inability) to accept a negative professional and life outcome. Boy, can we all be good at ignoring the writing on the wall, at not heeding advice that we do not want to hear. I have had countless conversations with colleagues expressing their impatience and frustration with senior postdocs who act as if finding permanent employment were not their problem, as if they were entitled to have their positions extended indefinitely.
b) the natural “inertia” that sets in, especially when families and/or significant others are affected by frequent displacements. I know from personal experience that one often loses sight of the negative, long term consequences of being in a postdoctoral appointment for too long, in view of the positive, short term relief arising from stability and two extra years of salary.

What to do, then ?
The question then, is not what to do with perennial postdocs, but rather how to make sure that no one end up spending years and years in a position that is meant to be temporary only.
Funding agencies, universities and government laboratories already often place restrictions on the employability of senior researchers in positions formally labeled (or, easily identifiable) as “postdoctoral”. For example, at my institution people having obtained their doctoral degree more than five years earlier cannot be hired on temporary appointments as “postdocs” — salary and benefits must be significantly higher, which in turn makes it financially much more burdensome for the PI. What other actions could be taken to “ease out” of the postdoctoral quagmire those who no longer benefit from that experience, and simply need to move on ?
Allowing only a minuscule fraction of people to become postdocs to begin with, and then creating government-funded ad hoc, permanent research positions for those who do not have what it takes to be PIs, is not the way to go. It is not right, it is not fair, it is wasteful use of taxpayers’ money, and it does not serve the best interest of society in the long run.
I think that what might work is a specific funding program aimed at connecting PIs and senior postdocs with potential employers, similar to internships.
Rohn is right in that starting a new career when one is in the late 30s can be a difficult proposition, but we know from the actual record that it is possible and that it does happen (which is why one does not see many former postdocs in unemployment lines or flipping burgers). These are people who can become effective contributors in different settings; I think that the effort must be directed toward connecting them with prospective employers.

The current system, which includes a few years of term postdoctoral appointment as a necessary intermediate step between the doctorate and, typically, a tenure-track faculty appointment, has been the subject of criticism and the source of discontent for many years. To be sure, some of the charges that Rohn levies against the postdoctoral position have some merit, and in fact the contention has been periodically put forward by some, that postdoctoral positions should simply be abolished.
In my opinion, however, Rohn completely misdiagnoses the problem, in the process perpetuating appalling misconceptions about the scientific environment and its players.
Aside from dealing a significant blow to the nation’s research output, scientific standing and overall productivity, her proposed course of action (one which will likely and luckily never be implemented) would have the result of greatly reducing opportunities for promising young scientists, accelerate the (already observable) overall aging of the scientific enterprise, reward seniority, independent wealth and/or stubbornness over excellence, and quite likely exacerbate the existing, serious problem of “scientific castes”.
And I also think that it would altogether make the research environment a less exciting, staler milieu.


[0] To what extent, in this day and age, any job can be really called “permanent”, is unclear to me.

[1] Rohn makes the trite usual point that “to force a highly trained postdoc from research is a terrible waste”. I am not going to repeat here what I (together with countless others) have already stated (I did it in this post, for example), other than that this happens to everyone in every walk of life, that there is nothing special about scientists (oh, I know, it is so much harder, blah blah blah…), that by that token we should shut down 99% of all arts, humanities and sports programs, etc. As for the “tournament model”, it is not clear to me why people think that that is a specific feature of the scientific career only. Capitalism is based on competition, like it or not (I don’t like it but I don’t think it’s going to go away soon). We are all competing, all the time, either to get a job, or to keep a job (i.e., to keep our business afloat or be spared by the next round of layoffs).

[2] This, of course, begs the question: how would one go about picking the “very best” PhD graduates, the ones to encourage to pursue the postdoctoral path ? It is already difficult as it is to pick the best candidate for an assistant professor position, from a pool of postdocs with a fairly extensive research record. Would identifying the most promising and research-talented freshly minted PhDs, out of a much larger pool of junior scientists with necessarily less experience and (usually) no record of independent work, not be even more difficult ? I mean, at that point the only differentiating aspect would be the school where…. oh… wait… never mind… silly me.

[3] Frankly, I think that five years is a long time pretty much in every discipline. I also have to admit it, I have the sense that in some fields there is the tendency to confuse opportunity with availability. The fact that the money is there to prolong a postdoctoral appointment, does not mean that that is necessarily what serves best the interest of the trainee.

Tags: , , ,

17 Responses to “Postdoc Emeritus”

  1. Jennifer Rohn Says:

    Dear Massimo, thank you for your comments on my piece; I found your arguments interesting.

    Just two responses.

    1) When you say “My personal “rule of thumb” is that after five years as a postdoc and no permanent research job, one should very seriously consider switching to another career”, and muse about ways to enforce that, you are also advocating, as I did, reducing the number of trainee postdocs in the system. I did not state that the tap had to be turned off at the very beginning – but it does need to be turned off at some point, so there will be a net reduction in the numbers. How this could be decided is a matter of debate. The editorial process inherent in the brief op-ed format did not permit much nuance.

    (It is worth noting, however, that the sort of expiry date you’re advocating would have consequences for people who take career breaks for family reasons. It might be better, during the cull, to weigh each applicant on his or her own merits, regardless of age or experience. Uncontrollable circumstances can sometimes influence how long it takes for a person to get from point A to point B; it doesn’t necessarily mean they’re not as good as someone who got there faster.)

    2) I am very sorry that you think my point about forcing talented scientists from research is “trite”. I am not sure why you think making this important point as part of a persuasive argument is a bad thing, as it is a key one. You could argue that your entire blog point has been said before “countless times”. In my view, discussions and arguments made by a diversity of people, in a diversity of venues, should not be discouraged. Otherwise, since people cannot be expected to read and keep up to date with the millions of blogs and articles out there, no one would ever feel able to say anything. And surely this sort of rebuttal-by-disdain, complete with sarcastic strikeout, does not do any favors for productive dialogue.

    • Massimo Says:

      Hi Jennifer, thank you for your comment.

      1) Point taken. You’re quite right, but I think that this non-trivial nuance is not obvious from your editorial. I interpreted what you wrote as implying that the marketability of freshly minted PhDs being greater, an encouragement should be given to them to get out of the pipeline at that stage.

      2) Jennifer, I am sorry if the way I phrased my comment, which was meant to be (mostly) humorous, irritated you. Please accept my apology. I still stand by everything I wrote.
      I know my blog is repetitive (my wife encouraged me to start blogging because she was sick and tired of hearing the same speeches 😉 ). This is why it is just that — a blog, it’s a place for me to vent. A post of mine is not in the same league as an editorial in Nature. My readership is probably in the (half) dozens, mostly friends 🙂
      When I say that this point has been made “countless times”, I am not quoting myself. It’s a debate that has been going on for two decades, and I can assure you that much more authoritative sources than me have expressed the same kind of skepticism toward any kind of quotas. Among the various objections that are raised, “Why in the sciences and not in other fields ?”.

  2. Jennifer Rohn Says:

    Thank you for your response.

    I’d like to point out that there are quotas in some other professional fields. Check out the enrollment policies for medical and law schools in many countries, for example. You won’t find them letting in hundreds of students for every single final position available; there is an enforced cut-off that is a lot more realistic. An Italian friend of mine told me that in Italy, med school numbers aren’t restricted, and it’s been a disaster. (This is only anecdotal, however, so apologies if her assessment wasn’t accurate.)

    Also, I could turn your question around: Why do we have to do things the way other professions do, if we want to make our profession better? Isn’t it a bit defeatist? Here in the UK, the government has become interested in the question of the science post-doc career structure and there is every chance that steps may be considered to make the system more sustainable. As far as I’m concerned, it’s a problem worth addressing and I’m going to keep fighting.

    • prodigal academic Says:

      I agree with Massimo here. A quota system hasn’t protected recent law grads from un- or underemployment, and most of them have massive, massive debt from law school that PhDs and postdocs don’t have.

      As I wrote on my own blog on this issue, restricting access to PhD programs screws over the disadvantaged in our society, and could potentially reverse the progress towards diversity we have made in the sciences. I would much rather allow people to roll the dice and take their chances than to restrict research careers to people that can get into and attend Harvard.

      In almost every career there are many fewer “desirable” positions than applicants. Why is this a particular issue for science and not for say machinists (union jobs), teachers (“good” school districts), or even doctors (attending/partner in a large practice)?

      That said, I also agree with Massimo that there should be rules in place to protect postdocs, since it is scary to leave what you know for the unknown, even if where you are (your postdoc) is not really good for you long term. At National Lab, a postdoc lasted 2 years. Under some conditions, this could be extended to a third year (usually for people converting to staff, projects with less than a year of funding left, or life circumstances that meant the postdoc hadn’t really had 2 years of experience). We could only hire people as postdocs with 0 or 1 prior postdoc (so no perma-postdocs). All applicants were told these rules up front. If something similar were implemented in academia, I think that would be a good idea. But this is not the same thing as restricting the total number of postdocs, or of restricting access to PhD programs. In my field, it would not have that significant an impact except on foreign scientists, since it is unusual to find someone without visa issues postdocing that long anyway.

      • Massimo Says:

        Why is this a particular issue for science and not for say machinists (union jobs), teachers (“good” school districts), or even doctors (attending/partner in a large practice)?

        Thank you, Prodigal, that is how I feel too.
        I think that anyone advocating special treatment or protection for a given category, should first and foremost make a case to the general public as to why said category is experiencing unusual pain or suffering over others. My sense is that there is where any such effort will hit a hard wall, namely convincing the general public that scientists are going through abnormal hardship compared to everyone else — I think people would look at numbers and roll their eyes…

      • Anon Says:

        As I wrote on my own blog on this issue, restricting access to PhD programs screws over the disadvantaged in our society, and could potentially reverse the progress towards diversity we have made in the sciences.

        I disagree. Where is the data that backs up your claim? Restricting access to PhD programs is not the same thing as abandoning affirmative action efforts. Admission to top-notch schools today is more competitive than ever, yet these schools have not suffered setbacks in terms of diversity – in fact, quite the opposite.

      • Massimo Says:

        I do not have data, but I shall make a general comment. For the sake of fairness, as much as possible the decision to pick one candidate out of many should be based on actual record, rather than on subjective assessment of the person’s potential. The earlier one places the cut-off, the less the decision will be based upon actual record and the more on perception of excellence. I am wary of any decision which is largely based on perception, conventional wisdom etc (example: these two candidates have comparable publication records but one got her PhD at a more reputable institution — let’s go with her).

      • Anonymous Says:

        For the sake of fairness, as much as possible the decision to pick one candidate out of many should be based on actual record, rather than on subjective assessment of the person’s potential.

        Then I suppose you do not support any kind of affirmative action, since that is all about looking at someone’s record in context – i.e., allowing your decision to be tempered by a “subjective assessment of the person’s potential.” For example, Julio’s scores are not that high, but he comes from a low-income background and went to a crappy school with limited opportunities, so he might be as capable as the guy from rich-fancy-prep-school who had access to test prep classes.

        And I say again: Admission to top-notch schools today is more competitive than ever, yet these schools have not suffered setbacks in terms of diversity – in fact, quite the opposite. It’d be nice to see you or Prodigal explain that fact away.

      • Massimo Says:

        Then I suppose you do not support any kind of affirmative action

        No, it’s a non sequitur.
        All I am saying is that if I want to decide which one of two individuals is a faster runner, the fairest way to do it is have them run and see who makes it to the finish line first, simple as that.
        Replacing that with a subjective decision of any kind, one that does not give one of the two the chance to run, is less fair, regardless of how one does it. Now, there are cases where making a decision based on presumption of future excellence is necessary, but this is not one of those cases.
        It is a straw man argument, I could just as easily respond that you are wanting to make it all about the school where you studied, that in the lack of an extensive research record allowing a fair comparison, a Harvard graduate should always be given preference over a Florida State one. I am not following you there, I am sorry.

      • Anonymous Says:

        All I am saying is that if I want to decide which one of two individuals is a faster runner, the fairest way to do it is have them run and see who makes it to the finish line first, simple as that.

        No, it is *precisely* not “simple as that,” and I think you know that.

        I encourage you to re-read my comment above and address what I actually wrote. But hey, if you aren’t even going to do that, then really what’s the point….

      • Massimo Says:

        … and I encourage you to go make your anonymous straw man arguments somewhere else.

  3. Massimo Says:

    I’d like to point out that there are quotas in some other professional fields.

    I am aware of that, but I think the comparison with the sciences is iffy. I personally see the sciences as much closer to the arts and the humanities, and there are no quotas there, that I know of. When it comes to medical schools, where the cost of training per individual is very high and physicians receive a very specialized education, quotas can perhaps be justified (and even there, they remain highly controversial, and their efficacy far from proven). They are enforced usually by limiting the number of schools that grant degrees to begin with, and by having rigorous entrance exams. Would you support something similar in the sciences ? I think that society only stands to lose from any such mechanism.
    (As for Italy, please don’t get me started — until Italians refuse to look for employment more than 30 km away from home, nothing is going to change).

    Is it really such a tragedy if there are scientifically competent people around who do not make a living out of doing research ? Why is it OK to have a lot of philosophers, musicians, painters make a living in ways that do not utilize their skills, but scientists need to be doing science ?
    I am certainly not going to try and prevent you from fighting for whatever cause you deem worth fighting for (and how could I anyway). However, I can express an opinion too, especially on my blog.
    With all due respect, I think that if society and its policy makers were to address the issue of middle-age workers who lose their jobs, or are displaced from their profession and need support (either by way of re-training or by creating alternate employment opportunities), scientists as a category would not even make top ten, in terms of perceived (or demonstrable) distress.

  4. Jennifer Rohn Says:

    I wasn’t trying to stop you from expressing your opinion; I was just engaging in counter-arguments – i.e. a discussion.

  5. Anonymous Says:

    Wonder what advice you will provide to a “postdoc emiritus” ? Seems an average physics postdoc for such a long time has nothing “sale-able” in the current job market. I believe career for high academic qualifications (Ph.D.++) is difficult. Industry likes people with experience or younger (less school trained) section of the population, plus some connection (network, alumni, referrals, etc.). Postdocs are trapped between the culture of their lab that may have practices in variance with general expectation of outside world. It is certainly not a blame game… but there must be a way out of here.

    Last thing I remember, I was
    Running for the door
    I had to find the passage back
    To the place I was before
    “relax”, said the night man,
    We are programmed to receive.
    You can checkout any time you like,
    But you can never leave!

    • Massimo Says:

      Well, it definitely gets harder as time goes by, which is why I personally have always tried not to hire or keep anyone around who would come to be regarded as “postdoc emeritus”. Honestly I do not know what “advice” to give, and frankly I have never been impressed by that given by APS — never did seem particularly realistic to me.

      However, I believe that a way out exists, i.e., I do not believe life is significantly harder for a scientist than it is for any other displaced middle-age worker, whether that be a mid-level manager, or programmer, or skilled worker who loses his/her job, like it happens every day to many.
      I do know of few people to whom that happened, in fact a number thereof if you include some who did not get tenure, and my anecdotal evidence is that they ended up working:
      1) at government agencies such as NSF
      2) at scientific journals such as Physical Review
      3) Four-year and Community Colleges, as well as High School
      4) I know one person who started his own company

  6. Anon2 Says:

    Different Anon here. Interesting discussion. I agree with you, Massimo, (cf March 12 post) and disagree with the other anon that you were in discussion with.

    In fact, I would go even further and say that the job candidate with the same h-index but from a lesser-known university and with a lesser-known advisor is mostly likely the better researcher. I don’t have any data, but, from talking with colleagues (I’m in pure math), I find that it’s harder for people at these-lesser known universities to have their papers accepted at highly visible journals and, in general, to have their work more visible to the field. Since citations depend on (but not only on) visibility, there is an inherent bias against people who are at lesser-known universities or whose advisors are lesser-known—this bias is unrelated to the merits of the research.

    One caveat with h-index (at least in my field of math, which has a low distribution of h-indices—the top postdocs 3 years after their doctorates have h-indices of 3 or 4) is that the papers of lesser-known candidates should be looked at to make sure that she isn’t almost solely publishing with her advisor or her h-index is excessively influenced by self-citations. Of course, this caveat applies even more so to the candidate from the famous university or with the famous advisor for the same reason that I stated above.

    To use your running analogy further, the two candidates are sprinters in the 1000 meters dash. The candidate from the famous university with the famous advisor starts at the 0 meter mark; the candidate from the lesser-known university and with the lesser-known advisor starts at the -100 meter mark. The h-indices are the same; the race is a tie. Except if the hiring department is interested in the better researcher (the faster runner), then most likely the candidate from the lesser-known university is better

    Finally, one more (sad) comment. There are senior assistant professors at R1s (for example, my own university) with very low h-indices (say 2, in mathematics) and won’t get tenure. They’re 8 or 9 years after their PhD (usually from a top university and under a famous advisor). Clearly, they were hired based on potential, not on an objective measure like h-index. I think the problem here is that no matter how famous and old the advisor, her experience is limited (after all, she is an expert in the subject matter, not in people) and her judgment is less reliable than the judgment of the field, which is borne out by the candidate’s h-index or mean citations. This is an unfortunate situation for both the assistant professor and the university. While we know from investing that past performance (h-index, mean citations) isn’t a guarantee of future performance (influence on the field by the end of one’s career), past performance is still the best predictor of future performance.

    • Massimo Says:

      No argument from me. Again, my point is: if one has to pick one individual out of a pool of N >> 1 candidates, one is less likely to make a mistake if the decision is based as much as possible on an examination of actual record, rather than on predictions based on anything but record.
      For this reason, I think it is beneficial to postpone such a decision until all N candidates have had the time to build such a record. Straight out of graduate school it’s too early, in my opinion. After three years of postdoctoral appointment, it’s easier and more objective a proposition, because one has more data.

      Naturally, even with a record the evaluation remains an operation fraught with difficulties and it is always perfectly possible to disagree on a ranking and make mistakes, no question about it. Nothing is unambiguous when it comes to humans, and I am surely not advocating relying exclusively, or even primarily, on quantitative measures such as the h-index (although I do believe that it is a good indicator).

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: