Posted by: jonkatz | November 15, 2009

Inefficiencies in the postdoc market

In contrast to the market for tenure-track jobs, the market for postdocs in our field seems woefully inefficient. From the potential postdoc’s perspective — and I have a student going through the process now — there are generally no advertisements for postdoc positions and so one has to resort to some combination of word-of-mouth (with the help of one’s advisor) and sending random emails to people asking if they might be hiring. And the inefficiency goes both ways. Two out of the past three years I have tried to hire a postdoc: the first year I was unsuccessful (and so the grant money I had went unspent), and this past year I was unable to find somebody until late May. (Strange as it might sound, part of the issue may be that it is a very strong market for candidates “in the know”: over the past 3 years I have made several postdoc offers to people who took postdocs [or, in one case, a faculty position] elsewhere.)

Surely there must be something we can do about this? Note that I’m not talking about increasing the number of postdoc positions, I’m talking about better matching people looking for positions to whatever positions already exist. How about a wiki where applicants can post their application materials, and advisors can post openings? Any ideas how best to set this up?


  1. Wouldn’t the best thing just be mathjobs for cs?

  2. Besides the lack of a centralized information source about postdocs, another inefficiency is the huge spread of application deadlines, leading to a four or five month window where one might receive a postdoc offer (at least it was last year).

    This long window contributes greatly to the problem because after receiving an offer, applicants may wait a month or longer to see if any better offers are received in the meantime. If something more suitable is received, the applicant will move to that offer and the process continues. You can’t really blame them, indeed it is natural given the scarcity of jobs. However, it leads to the effect you describe, where the prospective employers sometimes cannot find a committed applicant until May.

    A mathjobs won’t fix this, and a centralized wiki won’t fix it either. A centralized place that states when offers have gone out and if they have been all accepted would help, like that seen here:

    The above uses names, but similar objective could be achieved without using names: we could just say that x number of postdocs were available, y have been accepted, and the other x-y have until March 11th to decide.

  3. I second the nomination of a CS equivalent of mathjobs, in case anyone’s counting…

  4. I was unaware of mathjobs. But as far as I can tell, mathjobs allows postings of open positions but not postings of applicants looking for positions.

  5. When you fill out your application on mathjobs, it gives you the option of making your application available to employers. So it does have the option–it’s just not in the public way that you suggest (people looking for jobs do not have to announce that publicly).

    I think mathjobs would help coordinate deadlines (math does have much better coordinated deadlines) because you just go to one place and see that most of the deadlines are in a certain time window, putting pressure on your job offering to run on the same schedule.

    The current system definitely has inefficiencies. One of the worst such example was the CI fellows–most of whom (in theory) already had jobs, but took the position because of the extremely high (for a postdoc) salary. Of course these positions were distributed in July, and it is likely that no one filled the vacancies that opened up due to the late date.

  6. As for crypto-related positions, there is the IACR jobs page.

    It’s not fancy but informative. Maybe they can be talked into improving the page.

  7. There is also this page:

    started by the Computing Research Association.

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 )

Google photo

You are commenting using your Google 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: