On the 40h work week in academia

There has been a bit of talk on Twitter and among my colleagues IRL and online about work-life balance in academia. See for example this excellent post by @TheNewPI. The gist of it is: is a normal 40h work week, which is the norm everywhere, feasible for academics? I think the answer is yes and no. It really depends on what your goals and priorities are. If you really want to develop an independent research program as PI of a lab at a decent research university or institute, then I don’t think sticking to a strict 9-5 schedule will serve you very well. The more time (up to a certain limit) you put into your work, the better you will become, the more you will accomplish. However, I think there are a few points worth elaborating on here.

The problem of quantification

I think one of the key underappreciated issues in the “hours worked” controversy is how these hours are calculated. Does the hour when you are in the lab, but waiting for the gel to run and sitting on facebook count? Will it count if instead you are on Twitter, but  your twitter feed is mostly science-related? Does it count if you are reading science-related blogs? Does it count when you are at lunch with your labmates? Does it count when you are taking a collaborator out to dinner? Does it count when you are lying awake in bed trying to figure out a work-related issue? I think the people who say they are working 80-hour work weeks will answer yes to all these questions, and exaggerate a little bit to boot. Just last week I did probably a solid 50 hours of really intensive work (mostly at the bench, some supervision), not counting any downtime, and it feels exhausting. Perhaps I could pull off an extra 5 hours, but that’s it.

Diminishing returns

When you work overtime, at some point you will reach a level of fatigue that will be counterproductive. There is a very well known paper by John Pencavel at Stanford showing how productivity drops off after the 50h work-week. Mind, this is data for a rather mindless repetitive task, so work in research may be a bit different, but the data probably more-or-less applies to any occupation. At some point you will simply not work as efficiently as your well rested colleague. This work also points to the necessity of having at least one day off in the week, which is very much in line with my personal experience. If I work every day for 7 days, my next week is very unproductive due to accumulated fatigue.


In a way, the hours worked in research, especially for bench work, is not really about whether it’s 8.5 or 9.5 hours a day. It’s all about flexibility. The way I see it is this:

In the drawing above, each colored sawtooth pattern is a task. Tasks often look like this in biomedical research, where actual work (pipetting, mixing, setting up reactions) are interspersed with periods of waiting. So if you limit yourself to 8 hours flat, you may accomplish three tasks (yellow, red, blue) with the latter part of the day only half-occupied. If instead you say, I will stay flexible and try to get those two extra tasks (green, purple) done, you manage to accomplish almost twice as much, at the cost of only two extra hours at the bench. This is where I think people who don’t adhere to fixed work hours can really get a lot more work done than people who do.

Situation outside of work

What you have to do when you get home has an enormous influence on how flexible you can be in your work hours. If you are the primary caretaker of two kids, you will have to pick them up from daycare or school at a fixed hour, you will have to feed them, clothe them, entertain them, all of which are things that sap your energy almost as much as benchwork does. If you have a newborn at home, you will not get as much sleep as someone who does not. If you have a full-time working spouse you will have to do a bunch of chores, whereas if your spouse is stay-at-home, you will be off the hook a bit more. If you are well-off, you will be able to pay people to do your chores, cook your food etc. All these things matter!

The symptom, not the cause

Here’s a big one. I really don’t think the people who spend crazy hours in the lab are successful because they put in those hours. I think their spending time in the lab is a reflection of their passion for research, their insatiable curiosity, their “I can’t wait till tomorrow to know the result” attitude. And it is that passion that results in both their success and their long working hours. I’ve seen many PIs require in more or less unequivocal terms that their lab members spend a certain number of hours of overtime in the lab. Really, if these people are not intrinsically motivated to work overtime, they will not be more productive when they are forced to – they will just pretend to work during these extra two-three hours a day. The only way I think this might work is that it will quickly weed out people with less drive, which is maybe the real underlying motivation behind such requirements.

Level of motivation

In my experience, your ability to work hard is to a large extent limited by the perceived fruits of your labor. If you keep working 50 hour weeks and all you get is failed experiments and negative results, you will not be able to keep that up for long. If your results are great, a 50h-week may feel like a breeze. To some extent this will be dependent by the objective difficulty of the problem/project and the experimental methods, and to some extent it will be the external feedback that you get – from your boss, from your colleagues, from your reviewers. So my advice is: listen to your body. If your your experiments are falling apart all around you, you feel exhausted, and work fills you with loathing, just take a step back, have a leisurely evening or two, take the weekend off, and then go back to your work with a fresh mind. If, on the other hand, you are on a roll, you feel energized and full of enthusiasm, then ride that wave and stay at work longer – you will thank yourself later that you did.

The inevitable trade-offs

So now we can go back to our original questions: is working overtime important for your success as a researcher? Well, it depends. Are you extremely efficient, very skilled at the bench, very talented, and lucky to boot? Then 40h a week may be perfectly sufficient for you to get this coveted PI position at a reputable research institution. But you must realize you will be competing against people who are all these things, and in addition work 50-60h weeks. So you may instead get into a second-tier institution, be less well known internationally, do slightly less bleeding-edge research, attract fewer talented trainees to your lab etc. These are the trade-offs you that you may have to make to have a more fulfilling life outside of work, more face time with your kids, hobbies, etc. It all boils down to your value system and what is really important to you. These people who work 70h weeks have sacrificed pretty much everything for a satisfying career and that’s their (and yours) choice to make.

And to finish off, here’s another interesting post on the subject.

Edit: Another cool post, this time from someone who has left the ivory tower, but says that old habits die hard.

Edit: Here’s a nice paper on the how many hours/days people work in academia in the UK. It appears that at any career stage working 40 hours or less per week is the exception rather than the norm (~15% of PhD students, ~20% of postdocs, ~25% of PIs). In fact, it looks like working more than 50 hours a week is more common than working 40 or less.

Leave a Reply

Your email address will not be published. Required fields are marked *