Wednesday, March 17, 2010

The state of the astronomy job market

Image Credit: Anil Seth et al. 2009, "Employment and Funding in Astronomy"

The current job market for professional astronomers is, for those of us trying to navigate it, absolutely terrifying.  Finding jobs in astronomy is, in "normal" times, a highly uncertain and stressful process, and the Great Recession in the United States is having a profound impact on astronomy jobs, just like in most of the labor market.  But perhaps the largest source of terror comes from within, from a foreboding among young astronomers that there are fundamental problems with the astronomy career path in the United States, and that a large fraction of us currently on the job market are not going to realize our dream of becoming lifelong professional astronomers. 

The standard career path for the professional astronomer begins with four years of college to earn a bachelor's degree, followed by roughly 5 or 6 years of graduate school to earn a PhD.  Next, usually, come one or more postdoctoral appointments (think "pure research"), each lasting three years on average.  After this, the standard track has an astronomer being hired to a tenure-track position by a university (or sometimes a laboratory or observatory), earning tenure (effectively a guarantee of lifetime employment) 5 or 6 years later.  There are alternate paths, but this path is what most aspiring astronomers try to follow.  I myself have just finished my second postdoctoral position, and I'm searching for that tenure-track job.

One very important additional factor is the source of funding along the career path.  Graduate students pay to attend school, though usually they earn a stipend and tuition scholarships by working as teaching assistants or research assistants for university professors.   Postdoctoral positions are, in every sense of the word, "real jobs" (albeit with a time limit), with most salaries ultimately coming from the federal government.  This money may come through fellowships sponsored by NASA or the National Science Foundation, or it may come from research grants awarded to the postdoc's supervisor (again, usually university faculty) by NASA or the National Science Foundation.  There are a few other federal and non-federal sources as well.   Astronomy faculty, though, more often tend to be hired either by public universities, which tend to get their money from state budgets, or by private universities or observatories, which rely heavily on endowments and tuition income.  There are also quite a few professional astronomers, like those at NASA, whose paychecks come from the federal government, but a large fraction of tenure-track and tenured astronomers are not hired or paid salaries by the federal government.

Employment statistics in astronomy are not well-kept, but an attempt at a reasonably accurate trend of astronomy jobs through 2006 is shown in the graph at the top of the page.  Let's focus on just a few of the lines. 
  • The solid black line is the number of PhDs awarded in astronomy in the United States each year.  Over the past three decades, the annual number of astronomy PhDs has slowly increased, from about 100 in 1980 to about 170 in 2006.
  • The dashed-dotted lines are the number of tenure-track astronomy faculty (blue) and (quasi-)permanent staff astronomers (green) hired each year.  There is a lot of variation from year to year, but the number has been fairly flat at around 60 to 90 total positions.
  • The dashed black line shows the number of postdoctoral hires each year.  In the mid 1990s, that number was around 100, but it has tripled in the past ten years.
So, in summary, over the last decade the number of PhDs produced has been roughly constant, the number of permanent jobs has stayed constant, but the number of short-term jobs has tripled.  I've made a quick attempt at counting this year's job numbers, and it looks about the same, with about 60 permanent jobs and over 200 postdoctoral jobs advertised (postdoctoral jobs are still being posted, so there could well be 300 by the time the academic year ends).

It doesn't take a PhD to see that only about 1 in 4 astronomers in a postdoctoral position will get a permanent job, whereas a decade ago nearly every postdoctoral astronomer could find a permanent job.  This is a drastic shift in the job market in a very short amount of time, which means that even relatively new astronomy faculty may not realize the magnitude of the change.

You would think that our training in science should make astronomers valuable commodities in the high-tech industry.  After all, we are trained in designing and running experiments, in complex statistical analyses, and in computer and instrumentation skills.  So, shouldn't it be easy for those of us who don't get a permanent astronomy position to transition to careers in industry?

Sadly, the answer is no.  Academic resumes do not fare well under scrutiny by most HR managers, even in good economic times.  In times of high unemployment, like right now, there are many job applicants with a rich track record of success in industry, so it is even harder for an academic to get his or her foot in the door.  There are other ways for academics to get recognized and hired by industry, but that leads me to the next point...

Our preparation as professional astronomers tends to omit building marketable skills.  Most astronomers can write computer programs, but most of us use arcane programming languages.  Most astronomers speak disdainfully of Microsoft products, yet much of the industrial world requires fluency in Windows and Office.  We academics tend to view our research as an eternal and ongoing process, building on our past work, while much of industry operates on individual projects that both start AND end.  These differences are not insurmountable, but they are significant enough to hinder the career of the former academic, and many of us are discouraged, implicitly if not actively, from spending time on developing skills marketable outside of astronomy.

Also, let's now consider those funding sources.  Scientific research received a significant boost in funding from the stimulus package, and the current administration is amenable to increasing research funding.  But remember, federal money does not create many permanent positions!  Permanent jobs come from endowments, which were severely hit by the fall of the stock market, tuition, which has been increasing at an unsustainable rate, and state budgets, which are in crisis mode.  There is not going to be a significant increase in the number of permanent astronomy jobs in the forseeable future.

All of this is compounded by one other factor: the passage of time.  By the time an astronomer has completed two or three postdocs and come to the realization that there is no job for them in academia, he or she is in their mid- to late-30s, with little if any significant retirement savings and no easy path to another career.  We've worked our entire lives toward a goal that we believed was attainable, that of becoming a professional astronomer.  We've jumped through the hoops, we've been very successful throughout school and our early career, and now, all of a sudden, 3 out of 4 of us will have to leave the field altogether.  Is it any wonder that many of us are sleep-deprived, nervous wrecks?

It is my opinion that the astronomical community needs to take a hard look at how we train and groom academic astronomers.  Is the standard career path, the model that we've all been brought up with, really the best model?  And what if, as some people suggest, this career path does lead us to accomplish the best science, but at the cost of creating a significant population of scientists who will not find jobs in academia yet who are considered overqualified for many jobs in industry and also lacking some crucial keys to success in a world outside of academia?  How do we balance the demand for scientific progress with the needs of individuals? 

These are very difficult questions, very complex issues, and I don't pretend to understand the questions fully, let alone have the answers.  Yet there are several senior scientists (astronomers who I suspect and value their opinons) who look at those of us anxious about jobs and say, essentially, "suck it up and deal.  I also had to walk to school, uphill, both ways, and apply for 100 jobs before getting one offer."  This isn't just one senior person, nor a few. (And to my mentors and supervisors, it isn't you!  Honest!)  I appreciate their candor, and perhaps they are right and that the job market has always been this hard and frustrating.  But there is a perception of a lack of sympathy, a lack of caring, and that perception is growing among the younger astronomers.  That perception, whether correct or incorrect, only deepens the despair of those who are struggling to find a job, and that despair cannot be good for science!

Here are a few (of many) links to discussions about the current state of astronomy jobs, and science jobs in general:
Last point, just an "Amen" to a conclusion in the Scientific American article:
"Whatever model or models the nation chooses, many observers believe that the existing system of research by professors who constantly produce large numbers of scientists unlikely to achieve their career aspirations is near collapse. The real crisis in American science education is not young Americans’ inability to learn, or the schools’ inability to teach, but a distorted job market’s inability to provide them careers worthy of their abilities."


  1. raymond.l.wagner@boeing.com1:44 PM

    As an astrophysics PhD (UT, 1972), denied tenure after 4.5 years teaching at a major state university (preceeded by 2 years as a post-doc), I personally found it relatively easy (but stressful as suggested) to market my programming and algorithm development skills to enter the space systems engineering field. After 30 years in that new career, I have found it actually very much more rewarding (both technically and financially) than teaching in a university setting. I still am able to sit on thesis committees and guest teach (usually specialty topics -- Astronomy 101 or Physics 101, only one time each). The world does not end when you don't get tenure, you do have to look for the opportunities and think creatively. You still can do research and you still can keep your professional connections and interactions.

  2. Kurtis,

    One could also view this graph as showing that the overproduction of PhDs relative to the number of advertised long-term jobs has not changed much. The difference is that there are a lot more postdoc positions now.

    Sometimes people respond to this issue by suggesting that we should practice birth control by letting fewer people into grad school or hiring fewer people as postdocs. This sounds sensible for a moment but IMO, nobody who says this really thinks *they* were one of the ones who would have gotten rejected from grad school. It also doesn't address the issue that the PhD to long-term position ratio has not actually changed that much.

    A more defensible position, again IMO, is the idea that funding should be redirected to create fewer 3-year postdocs and more longish-term research positions. How this would be administered is an open question.

  3. Good comments, Ben.

    I also think that perhaps one of the root problems is how we define success in the field. From the time we are recruiting undergrads, we speak of success as tenured faculty at tier 1 research institutions. Sure, we give lip service to "alternative careers," but the coded message is not hard to see.

    Yet when I look at people who have "left the field" (as we might whisper if a prospective grad student really presses us on the issue), I see nothing but what the rest of society might call success. I see PhD astronomers who made a killing in Silicon Valley startups, and who lead R&D or quality control departments in biotech. I see colleagues who are excellent scientists and very happy and successful in their careers at teaching-oriented colleges. A good friend from undergrad with an astronomy degree is very successful as a lawyer. And Raymond's comment is yet another example. These are all successes, and yet would be swept under the rug by many astronomy programs.

    Perhaps astronomy undergrad and grad programs should be fostering success of all kinds by offering (requiring?) training in skills such as teaching, programming, management, etc., and by seeking out private industry for suggestions on how to provide students with tools for success.

    1. Amanda (?)1:13 PM

      "Perhaps astronomy undergrad and grad programs should be fostering success of all kinds by offering (requiring?) training in skills such as teaching, programming, management, etc." They absolutely should, but I don't see this ever happening. It goes back to the lack (real or perceived) of empathy. I think we has young astronomers need to take the Taurus by the horns and create our own networking. Maybe a wiki where people who have gotten in the industry door can share ideas, tips on what to put in resumes (vs CVs), etc. ?

  4. Kurtis, Raymond, and Ben

    Some people choose to stay in science research, even if it means less money. No one should ever be told that people do better leaving science, and then told to leave. Having that happen can make it impossible to develop and maintain the skills it takes to stay.

    Ben, I think those of us who believe that it would be better to have had a more orderly reduction in numbers at graduate school know that it is not certain that we would have been let in, but it would have been better to have it better determined earlier. It would have helped me know that it might be impossible to change fields later. It was the bad astronomy job market that led professors to lead me to physics, which ended up being worse. Even if I had had to reapply for a few years, it would have been better for me if the job market was not so bad after graduate school.

    There is also a huge discomfort that leads us to ignore the issue of nationality. I believe that the US citizen science community should openly discuss how many citizens versus non-citizens are given visas to start with Ph.D.s. I believe we can still be "world-oriented" and still say that the number of jobs given to others is too high. It is the unspoken part of the problem, and this bad job market has led to further discouraging American students from considering science. None of us that advocate considering national origins wants to "shut the door." But there must be a limited quota for science immigration like every profession. Otherwise, half the world would move tomorrow. Such a quota could still allow the US to seek the absolute best. I always advocate that any quota should consider "net immigration" with the intent that for every US scientist that goes to a country, one more person can come here. I have been unemployed for too long: there is nothing wrong with advocating that for mid-level science jobs, that if the US provides more jobs then US citizens should get more of those jobs.

  5. Anonymous6:24 PM

    Hey Kurtis,

    "How do we balance the demand for scientific progress with the needs of individuals?"

    Thanks for the writeup! This is a crucial problem for those of us in the early stages of the pipeline, yet many faculty with stable jobs are not as worried about individuals as about the future of astronomy in general, as in your quote above. For example, my department is growing in number of grad students and postdocs, but not at the higher, stable/permanent positions. Thank you for your post, as at least I know I'm not alone in my concerns about this pyramid scheme of more and more grad students and postdocs but the same number of permanent positions!

    Even if our training were translatable to industry, why will 75% of us switch at age 35-40 to a new career and industry, when we could have started there at 25, had we but known there wasn't a future in astronomy for us?

  6. Anonymous10:24 AM


    Our professional resume writing service welcomes you at our Top 20 Most Influential and Popular Career Blogs from all over the world. Now your blog is in TOP 3 best blogs about career and job in the Internet -


    Alex Miller,

    webmaster of

  7. Anonymous12:11 PM

    How about looking at other countries for employment? South Africa has lots of opportunities for astronomers for instance.

  8. Anonymous11:24 AM

    All of this makes sense, and the only solution to the problem as stated is to reduce the total production of PhDs to equal the total number of permanent job openings. Looking at the graph, that would be a 25% cut or so, which isn't too large. But it's difficult for universities to do that, as everyone in the university (except maybe the applicants) has no incentive to reduce the size of grad programs.

    But perhaps the problem is not so much an imbalance between PhD production and final consumption into permanent jobs but the increasing number of years spent as a post-doc in between -- in regard to lifetime financial trajectory, getting hired as a professor immediately out of grad school is hugely different from spending 9 years in 3 post-docs in between.

  9. Anonymous9:43 PM

    Do you know any professor or anyone who is doing research or working on the relationship between the stock market and the planets positions?

    Please let me know: predictdjia at hotmail


  10. Anonymous2:46 AM

    thanks for information sir. i check this regularly i suggest this blog to my friends also.

  11. "But there is a perception of a lack of sympathy, a lack of caring, and that perception is growing among the younger astronomers." - Welcome to the real world academics. Yes there is a lack of caring, a 100% lack of caring. The world isn't there to accommodate you. No matter how much you want to be an astronomer - if there's no jobs, there's no jobs.

  12. Anonymous7:33 AM

    I do not agree that astronomers aren't trained for jobs in the private sector. We learn how to present our work in talks and papers, convince our collegues in discussions, we learn how to teach, lead a group (e.g. guide PhDs), how to write applications, e.g.for money and observing time, so we understand how politics and diplomacy works and acquire general administrational skills), how to organise and plan a project, to use computers, most of us speak one or two foreign languages ... last not least, we learn how to learn and tackle problems. What else, what other skills does one need to get a good job in the industry? You say we learn outdated programming languages. True, but who wants to be a programmer? Wouldn't you like to be the one who tells the programmers what to programme?

  13. Interesting post. I'm actually surprised by how large the numbers are. There were 6 astronomy faculty positions at liberal arts colleges (so not including R1 universities) when I was applying.

  14. There are not too many astronomy PhD's for the economy, just too many for the Academy. But this has been true for a long time. PhD astronomers have many skills that employers seek, and can rather quickly acquire the few skills graduate school does not teach and become highly sought after.

    We certainly need to do a better job of making sure that our undergraduates understand the likely outcomes for PhDs when they consider applying to graduate school, and that our graduate students understand the odds of getting the job they think they want before they sink 4-7 years of their lives into a PhD.

    This is especially true because maximizing one's chances of getting that fellowship that maximizes one's shot at that tenureline job at Prestigious U requires a dedication to one's thesis that can be inconsistent with mental health (see "The Letter":

    We also need to eliminate the stigma of so-called "alternative" career paths and recognize that they are not "alternative", they are THE NORM. Using the terms "permanent job" and "permanent research job in astronomy" interchangeably is part of the problem. *ahem*

    We need to make sure that they our students are getting good career advice, and have an "exit strategy" that is more than a booby prize. We need to make sure that they have the opportunity to learn the few skills they need but don't yet have that change them from not-so-employable into highly desirable applicants.

    The tenureline faculty advising the next generation of PhD astronomers are the worst people to be in charge of this, because they generally attempt to mold their students into their own image. After all, how much do they typically know about industry or failing to advance to the next level? Some students might even be scared to even admit to their advisers that they are considering a so-called "alternative" career.

    So, bottom line, I don't think the solution is to produce fewer PhDs or try to create more permanent research jobs (though I would certainly welcome the latter!); it is to make sure that we are not making false promises to our students, and that we enable them to get the skills and contacts they will need to be happily employed, wherever they go.

  15. Jason - I agree the world is not here to accommodate anyone, and no career path is fair or all unicorns and rainbows. But my opinion is that there is a perception among pre-faculty astronomers that they are being trained for a specific job that is especially hard to get, and that the *people* further up the academic ladder don't give a rat's patootie about the stress, doubt, nor the long-term costs of pursuing a career unlikely to pan out. And there is a perception that not succeeding in obtaining a tenured position at a research institution represents a personal and professional failure. As a faculty member (now, 3 yrs after I wrote this post), I feel it is a moral obligation to recognize these concerns as being valid and do what I can to educate potential professional astronomers about the realities of the academic job market and the numerous, higher-paying opportunities outside of that rank.

    Anonymous: I agree with you that astronomers are indeed trained with numerous desirable skills. But early-career astronomers often do not appreciate this, and even if they do, they have difficulty advertising these skills. I do stand by my criticism of HR departments, but as I mention, there are ways around it (networking)