Saturday, December 3, 2016

Some (reluctant) thoughts on postdoc pay

Update 12/7/2016: (first follow up here, second follow up here)

I have generally steered well clear of the issue of postdoc pay, which engenders pretty heated conversations that I'm SO not interested in getting into publicly, but one thing I'm seeing is really bugging me these days: people bragging on Twitter about how much they pay their postdocs above the NIH minimum. Like this:

I don't mean to single these folks out—it just happened that I saw these tweets most recently—but I've seen a few such statements over the last year or so since the announcement that the mimimum for salaried workers would be increased to ~$48K or so (which was just recently reversed).

Why is this irritating? Well, first of all, in this funding climate, and given many labs that have to make many tough choices, it does strike me as a bit arrogant to talk about how much more you can afford to pay than many, many other very well-intentioned scientists. The implication is that people who don't pay as much as you do are paying an abusively low amount, which is I think an unfair charge. For these reasons (and maybe a few others), I just don't think it's really appropriate to publicly talk about how much you pay your people. For the record, I support paying postdocs well, and I think the increase is overall a good idea. My point here will be that there is not an obvious default "right" position on the issue of postdoc pay, and I think it is far more complex than just saying "We should pay postdocs a decent wage."

Indeed, I think the key difficulty is pinning down exactly what we mean by the notion of "decent wage". For instance, in the first tweet above, the PI is from Cambridge/Boston, and the second is from NYC. Now, the proposed federal regulation for starting postdocs is (was) $47,484, and that would apply everywhere. Including, say, Ann Arbor, Michigan (which I choose for no particular reason other than it's home to a major, world-class research institution, but in a relatively affordable location). Now, comparing the cost of living of any two places is tricky, but I found this estimate that Boston is roughly 1.4x as pricey as Ann Arbor (which sounds probably about right). Bragging about paying $60K? Well, shouldn't that be $66K? Live in Cambridge MA instead? No better, $76K. So let's stop crowing about how "decently" the Broad Institute pays, okay?

So, is $60K "fair"? Hmm. From the PI perspective: a Boston PI could say, well my dollars don't go as far, so in a way, doesn't the Michigan PI have an unfair advantage? Then again, the Michigan PI could say hey, why do I have to pay more (relatively speaking) for my postdocs? Why does the Boston PI not have to pay the same effective wages I do? Why should they not have an enforced effective minimum standard pay and have the freedom to pay effectively less?

The motivation of PIs may also matter here as well. The focus in the discussion has been on PIs taking advantage of cheap labor, and that definitely happens. But some PIs may define their mission as training as many scientists as possible, which certainly seems reasonable to me, at least from one point of view. (And I do wonder how often those who brag about paying so much above the minimum have actually had to make the tough choice of turning away a talented postdoc candidate due to constrained funding.)

From the NIH perspective: what is the goal? To get as much science as "efficiently" as possible? To train people? To create a stable scientific workforce? Or to better human health? Should the NIH even allow people in high cost of living areas to pay their postdocs more? Would it be fair to consider this pay scale in grant review, just as other areas of budgets are scrutinized? Does increasing the minimum penalize those who pay the minimum in non-Boston/SF locations unfairly, thus increasing inequity? Or does it provide a general boost for those places, now making them more attractive because their NIH minimum dollars go further? Should the NIH scale the size of grant by cost of living in the area of the host institution? To what extent should the NIH support diversity of locations, anyway?

From the trainee perspective: It's pretty easy for trainees to say that whatever they're paid right now is not fair (though you might be surprised how little many assistant professors make). So for trainees reading this post, let me ask: what would be fair? Okay, maybe now you have a number in your head. Where does that number come from? Is it based on need? Consider: should a postdoc who has a family be paid more? Wait a minute, what about the postdoc without a family? What about immigrants with expensive visa costs? Or potentially families to support in their home country? Moving costs? Commuting costs? Should postdocs be paid more when the institution is in an expensive city? Should postdocs be forced to live further away from the institute to seek more affordable housing? My point is that there is no clear line between necessity and luxury, and wherever that blurry line does get drawn will be highly dependent on a trainee's circumstances and choices.

Or should that number be based on performance? Should the postdoc entering the lab with a flashy paper or two be paid more than the one without? Should a postdoc get a raise every time they publish a paper, scaled by how important the paper is? How many grants it generates? I think it's reasonable to assume that such an environment would be toxic within a lab, but wouldn't the same be true of pay based on personal circumstance, as just discussed above? And isn't such performance-based pay already what's sort of happening at a more global level in flush institutes where PIs can get enough grants to pay well above the minimum?

As you have probably noticed, this post has way more question marks than periods, and I don't claim to know the answers to any of these questions. I have thoughts, like everyone else, and I'm happy to talk about them in person, where nuance and human connection tend to breed more consensus than discord. My point is that reducing all this to a single number is sort of ridiculous, but that's how it works, and so that's what we all have to start from, along with various institutional prerogatives. In the meantime, given how simplistic it is to reduce this discussion to a single number, can we please stop with the public postdoc pay-shaming?


  1. Over here (in the UK) postdoc pay is set and reviewed on grant applications. On your application you say "I want to hire a grade 7.3 postdoc" (which would be around $42k, and pretty much the same in all universities). The funder might say - no way this work requires a 7.3, or they might say you'll never find someone good enough for 7.3. But once you've set the level, thats the money you get, and you can't sacrifice money set aside for, say, consumables, to pay more. Consumables money is consumables money, staff money is staff money. So in a way the decision is taken out of our hands and made a nationwide decision.

  2. Arjun - thoughtful post as always. A few areas that I think are important to consider:

    * Seeing this conversation can be helpful as it may inform potential postdocs that there's a range of pay. Funding can be quite opaque for trainees, and if one is regularly told that there's a standard rate for NIH postdocs then a trainee may not realize that pay can vary between groups. When I interviewed for postdocs, I had no idea that there was variability until I got one offer that wasn't at the standard NIH rate. By broadening the conversation, it may increase awareness.

    * This awareness may also encourage people who want to pursue an academic career to give it a shot. I know at least one person who left the pipeline because he/she didn't want to go another three years struggling to get by (family, single income, expensive cost of living area). The larger the difference between "what I could be making" and "what I am making" gets and the longer postdoc periods become, the more the academic pipeline becomes only people who can afford to give up the lost wages. I think this is bad for the field.

    * The closer the pay of postdocs in a lab gets to the level that a professional staffer would be, the lower the incentive is for a PI to retain a postdoc longer than is good for the postdoc. Also the more a PI may invest in each postdoc, as the training mission becomes the primary reason for taking on someone in a postdoc position as opposed to a staff role.

    * The opportunity cost of a postdoc in fields varies dramatically. In our field, we see starting salaries for people just finishing PhDs at and above $120k with benefits, potential stock options, etc. An individual considering a postdoc at the NIH rate could triple their salary in industry. This may put further pressure on the pipeline in those fields and retain only those who can afford to stick it out, which may reduce at least socioeconomic diversity of the next generation of faculty.

    In closing, I definitely see that the public discussion of salaries is uncomfortable. This is particularly true when pay can be opaque (i.e. for those at private universities like ourselves). I'd say that I think the conversation does more good than harm. It's an uncomfortable topic, but I think that, to the extent that it raises awareness that pay varies, it may help someone keep looking who would otherwise drop out of the pipeline. Anyway, these are my thoughts in only a bit more complete form than twitter. Always happy to chat over a donut if you want to talk more on the topic!

    1. So I of course totally agree that postdoc compensation is a discussion worth having! The question is whether braggadocious Tweets by a select few preeners while many others stare at their feet uncomfortably paying the minimum (and some just do what they think is right, without making a big deal about it to others) is the right way to have that discussion. I would argue no, and I think it creates animosity that is counterproductive. I have no argument with paying more, or having minimums, or whatever our community—students, postdocs, faculty—decides. Currently, these tweets come across more as advertisements for the PI than a starting point for discussion.

    2. You just wasted hours complaining about a tweet, really pointless, considering you could be working towards improving your funding stream it sounds like mate.

    3. Yes, yes, you're probably right Anonymous, thank you for the feedback.

  3. finally saw someone written with some sense on this topic... it's putting too much pressure on us who cannot pay that much.

    my best postdoc never complained about payment, published 5 papers in 2 years, and found a job all by himself. my worst postdoc complain about salary (at $50,000 in michigan) all the time, not able to write a single line of code, telling my suggestions are wasting his time, and telling how much other labs are paying and saying that he would go to michael snyder's lab to get more payment. sounds like snyder's lab is a collection of garbage. i only know one thing, that he is the last postdoc i would ever recruit.

    1. Your argument based on your subjective experience with sample size of 2. You are already judging the postdoc's value by utilitarian concept by a performance metric that is in debate. I suggest for you to take a simple course on what it means to be a leader.

  4. There was a similar back-and-forth on many of these points (eg cost-of-living adjusted salaries) earlier this year at a meeting between postdocs and the Vice provost for research at UPenn (the office that oversees all official postdoc policies), and one of the most powerful points that was made, was when someone stood up, talked about his difficulty for covering either childcare (so his wife could work) or health insurance for his kids on a postdoc salary and said "It just has to be enough money to live off", and I think that is very true. Postdocs in NYC and SF made similar points when it comes to housing costs (ie they don't want to get rich, they just want "enough" to not have to live in a dump). Importantly, if salaries are not in this range, then academia will simply not be accessible to/lead of exclusion of people from more financially vulnerable backgrounds. There are probably ways to adjust salaries fairly (eg on-site subsidized childcare facilities or help with housing) without directly increasing salaries. But from a postdoc perspective it seems that having that (or even a) discussion about postdoc pay seems at the bottom of the priority list for many institutions (eg some universities don't even have minimal pay guidelines for postdocs), and I hope that the current "who pays how much" will at least lead to a conversation on the topic, and bring people with different perspectives to the table.

    1. @Uschi made the point more eloquently than I did. Agree with this. Twitter is almost certainly not the best conversation place for this, but I think that it's better than no conversation.

    2. Indeed, a fair point—I guess I probably wouldn't have gotten into it either (at least not publicly) without those tweets. For what it's worth, Penn does have postdoc pay guidelines. Whether those require adjustment is an important question, and one that perhaps should be getting some attention.

  5. I really enjoyed reading your post and thank you for your thoughts. All the arguments made here are valid. This is from a grad student on his way to be a postdoc next year, by the way. I think:

    0. I would never choose a university based on salary (currently I have a choice of X over Y with X paying more than Y on paper, but X being in a more expensive place to live in). Biggest thing is quality of research and success in placing postdocs in academia.

    1. Postdocs with dependents/children should be paid more. Sufficient enough to cover or partially compensate for childcare and healthcare. Being a person without dependents, I would not begrudge a colleague earning more if that person has other people to look after.

    2. I think it should be a living wage, based on where the university is located. Hence having a fair minimum wage is of utmost importance. I will not name my university, but I am aware of PI's in another department paying postdocs with families 20,000 a year. Just so that the group can be larger. Having a law will prevent any such situations.

    A bigger concern for me rather than pay, is the holding pattern that postdocs face as they attempt to be professors. Granted in engineering it is a lot less of an issue as compared to sciences as the industry is a huge sink for PhDs; however, it is necessary to sort of limit the uptake of postdocs within research groups to ensure a proper balance.

    This will lead to:
    1. The better half of grad students going on to becoming postdocs (less chance of one being a liability).

    2. Not so much of a funding crunch for PIs knowing full well that they have someone who is really good.

    All in all, I am very happy that there is some discussion about this in the community. The whole point of a postdoc is to train as an independent researcher, mentor grad students, help with grants, and ideally move on to the next position within 2 - 3 years. The last bit is the most important to any PI or postdoc, so perhaps having a minimum wage will limit the number of postdocs, move the holding pattern to some sort of metastable steady state and also provide PIs with the highest quality of graduates, which is what really matters.

  6. Thanks for the interesting post. I'm a grad student hoping to get a post-doc position after.

    My Main Point:
    Paying graduate students and post-docs enough to afford to focus on research is critical to also allowing people from diverse socioeconomic backgrounds to consider academia as an option!

    I think it's very important that you mention cost of living in your post. I just want to post from the perspective of a person who has lived in poverty, and has been lucky enough to get into a graduate program in a location where I'm paid enough to support my family (because of a good mix of low cost-of-living and competitive stipend).

    For me, it's not as much about the research as it is about pay. I don't ever want to have to worry again about where to get food for my family or if we will be evicted next month. I don't ever want to be in a place again where I have gotten rid of all modern conveniences (cell phone, car, internet, etc) and still am not getting enough for basics (food, shelter). So I love science, but I will find a way to love what I do no matter where I am, as long as I am paid enough to live a good life. This is something many people can't understand, perhaps because they've never had to deal with starvation or homelessness.

    Wouldn't it be nice to just say "I really like this PI and the projects in zir lab!"? Competitive pay (including an adjustment for cost of living) is key to that.

    Again, thanks for asking a lot of questions!

  7. Just want to remind folks that there are non-post-docs in many labs who also have wages coming from the same grant-funded pot of money.

    In some ways, *some* of us (because I generally count myself as one of them) have it worse than post-docs... Because we'd like a stable job we can live with, not just a short-term gig to use as a stepping stone to the next level.

    I've seen PIs forced into really shitty "who do I have to screw over" choices by well intentioned mandatory pay raises. Frankly, I think there are numerous systemic problems, but unrealistic expectations about both productivity and cheapness are big factors... Labs need more money to pay people actually doing the work (which certainly includes post-docs most places, but is far from limited to just them.)

  8. Sorry, Arjun. I really love your posts and insightful comments, but I cannot agree with your plea for stopping public postdoc pay shaming. It is very important to have a conversation about this sensitive, yet highly overlooked issue and public shaming is the only way which can kickstart this discussion.

    That said, I wholeheartedly agree with you that folks should stop bragging about this on Twitter. However, in all fairness most people who do so (Boston, NYC types and their offsprings), live in a very superficial world where hype and publicity for their half baked results, and a relentless pursuit for branding matters much more to them than actually caring about science or scientists (at all levels).

    I agree with your overall theme that this is a complicated issue and one size fits all discussion on Twitter, will not provide any answers. But from my experience at least this creates some awareness about this issue. I wasn’t even aware that postdoc salary levels at universities have a lower limit and that PI’s are free to supplement the salary beyond the minimum, until I read the tweets from Casey Greene and others in Spring of 2016. And yes Casey Greene did not brag about the number on Twitter, he just said that he advocates a higher amount and follows it in his lab.

    Moreover, I also completely agree with you that with reduced budgets, most PI’s will not be able to pay postdocs a higher amount. But I think they should not lie and make silly excuses that this is a university/NIH mandated salary.

    Like you, I do not know the answer to this complicated question. As an ardent believer in free market, I think that markets will adjust themselves in the long run (assuming that the government does not create a one sized regulation for all). I do not think that all postdocs should be paid the same amount, however everyone should get a basic living wage. A star postdoc should be paid better in a more efficient marketplace for postdocs.

    I want to thank you again for this post even though I do not agree with your central premise on stopping public pay shaming.

  9. Great post and discussion. A couple of points I would add:

    1. While many of us are trying to find ways to pay postdocs more, there are still plenty of faculty who object on philosophical grounds. They see postdoc salary concerns as a red flag and prefer not to hire a scientist who would be so base as to seek a higher salary. This is the "science is a calling, not a job" crowd that still doesn't get the whole texting thing. The more these folks hear about above-NIH scale salaries (gasp!), the better.

    2. We need to make science not only a viable career, but an attractive one. There are too many options for smart, creative, and ambitious people with technical talent these days outside of academic science. We obviously can't offer the same salaries as industry, but we need to stop demanding that pursuing a career in academic science must be a financial burden. I think this applies not only to postdocs, but graduate students and lab techs, too. So while it's fine to assert opinions on a postdoc might "deserve" in compensation, it is important to note that this is ultimately a labor market in which we are becoming non-competitive.

    1. Thanks for the comments! I certainly think it's worth discussing openly that there are above-NIH scale salaries, and the extent to which this perhaps forces the hands of those who are acting inhumanely, that's great. I do think that publicly talking about how much more you yourself pay your postdocs (as opposed to more generally) comes across weirdly—even if the intent is not to brag, it's clear that many interpret it that way. But if that's what it takes to start the discussion, so be it.

      I think the labor market analogy is a difficult one to fully wrap my head around. A number of folks (trainees, actually) have mentioned that they like the idea of a meritocratic based system. I think that the reality is that in such a high-stress, high-risk endeavor as science is, we have to provide a baseline that I agree must be at a level where it does not induce financial distress. To be clear, at no point do I assert an opinion on what a postdoc "deserves"; my point is merely that it's complicated, and hard to reduce to a single number, whether a matter of NIH policy or on someone's twitter feed. Whether we are competitive with industry is a related but still sort of separate issue that ties into a bunch of other "pipeline" issues that I will never, ever write about online… :)

  10. On the second point, I didn't intend to imply you had made any declaration of deserts; it was perhaps more of a reply to the discussion you've started here. Your post is definitely more thoughtful that the standard twitter chatter, and your questions are all good ones!

    But I would quibble that the labor market point isn't just an analogy; it really is a labor market, and one that operates at the career level. I doubt I'm alone in having had a number of brilliant high school and undergraduate students come through my lab, experience a little computational biology, and then show zero interest in an academic research career. They would all rather go to industry and do creative things with data, and aren't interested in 20+ years of academic CV-building before they can be recognized as a leader. Of course, salary is just one component of this, but I think it's an important one in moving away from an outdated and often unattractive "trainee" model to a professionalization of science at every career level.