Should You Go to Grad School? (Repost with slight revisions)

Congratulations! It’s your senior year of college. You’ve aced all your classes and love the life of the mind. You like writing research papers, and you’re smarter than half your professors. The thought occurs to you: I should go to grad school. Well, should you?

Here’s a very candid account of what you’re in for. I’m not trying to tell you yes or no. I just want you to have a clear account of what the risks and benefits are. I am not claiming that any of the stuff below is fair or unfair, just or unjust. I’m just saying it’s how it is. Just as it would be imprudent to buy a car without knowing what you’re buying, it’s imprudent to enter grad school without knowing what you’re getting into.

A Ph.D. is a professional degree, like an MBA or a JD. It’s designed, somewhat poorly, to make you an academic. (I say poorly because the Ph.D. emphasizes research rather than teaching, but most professors do little research and spend most of their time on teaching-related activities.) 

My advice is based on the assumption you want to get a tenure-track job after you get a Ph.D.

Here’s what you do when you get a Ph.D.:  http://matt.might.net/articles/phd-school-in-pictures/


Being a professor can be a glorious job.

I get paid to play with ideas.  If you have a passion for ideas, there are fewer jobs better than being a college professor. I have a remarkable amount of freedom and autonomy in determining how I spend my days. I spend most of my time every week thinking and writing about whatever I happen to find interesting.


However, I also have one of the most plum jobs in the academy. Most people who work in academia have much less freedom than I do. They spend much more of their time doing administrative work. They teach more classes than I do to lower quality students who don’t care about learning for learning’s sake. They rarely teach courses on topics that interest them. And they do this for much less money.


Unfortunately, it seems to be getting harder and harder to succeed. I had to do more to get my job than the previous generation did. Five years from now, you’ll have to do even more. To get the jobs your professors have, you’ll have to be better than they were when they got their jobs.

I like reading and discussing economics or political philosophy. It‘s my hobby. Should I go to grad school? You can do all these things without getting a Ph.D. You can read and discuss economics while holding down a job as an insurance agent, a lawyer, or a consultant. You might be able to maintain your hobby while making a lot more money.

I’m the smartest philosophy major at my college, in fact, the smartest they’ve ever had. Should I go to grad school? Graduate programs get hundreds of applications for a few spots. Pretty much all of the applicants were the smartest undergraduate majors in their colleges.


How hard is it to get in? At a good graduate program, the acceptance rate will be under 8%. It’s harder to get into a good Ph.D. program than to get into Harvard as an undergraduate.


Should I take on debt to go to grad school? No. Your graduate program should waive tuition and pay you a living stipend. Exceptions exist, but the following principle is almost always true: If you have to take on debt to get a Ph.D., you should not get a Ph.D. Two reasons: 1. The risk of failing to get a job that pays well enough to pay off the debt is too high. 2. If you can’t get someone to pay you to go to grad school, you probably aren’t good enough. Yes, exceptions exist, but believing you are the exception to the rule is unexceptional.

What do you do in grad school? You take classes for 2-3 years in a wide range of subfields. You’ll write 12-16 30-page seminar papers. You’ll read thousands of pages of material. After that, you’ll usually take some sort of exam to show you have expertise in one subfield and a broad range of knowledge in the other subfields. If you fail that, you’ll get kicked out. If you succeed, you’ll devise an original research project. You then take an exam–a “prospectus defense” to determine whether your project is even worth pursuing. People can and do fail that and get kicked out of grad school.

If you pass the prospectus defense, you then write a dissertation. Once again, you take an exam when you’re done. If you fail, you’re done. If you succeed, you get the Ph.D. The whole process should take 5 years–maybe less if you do econ–but most people tend to take 6 or 7 years, and in fields like history, they take even more.

The longer you take to finish, the less likely it is you’ll get a job. Job committees want evidence that you will be productive. Grad school is easy compared to being an assistant professor. If you’re bad a grad school, i.e., if you can’t produce when you have few responsibilities and having panels of experts issue-spotting research topics for you, you’ll probably be terrible at being an assistant professor…which means you won’t get tenure…which means your colleagues have to hire a replacement for you in six years. Hiring committees are looking for people whom they believe will pass tenure-review.

You can easily find yourself in your mid to late thirties still a graduate student, making a measly $20K a year living stipend, with no job, no savings, a crappy old car or no car, and a tiny, lousy apartment, while your friends own houses, cars, and make six figures. Think about the opportunity cost. You forgo many opportunities to study for a Ph.D., and there’s no guarantee of a job at the end.
That said, graduate school is a glorious time, and there are worse ways to spend your life. Many intellectuals would be happy to pursue the degree just for the sake of having the graduate school experience.

What Kinds of Jobs Are There? There are tenure and non-tenure-track jobs. Tenure-track jobs are the gold standard. You spend six to seven years on probation. If you do good enough work, you get tenure, and then you pretty much have a job for life, unless you do something really stupid or your university has a major financial crisis. Non-tenure track jobs include post-docs and adjunct positions.

There are two basic types of tenure-track jobs, so-called teaching positions and research positions. Both positions involve teaching and research, but the balance is tipped one way or another. Most people seem prefer a research job to a teaching job, but your preferences might be different.

A research job means you work at a doctoral or masters-degree granting program. You have higher status in the profession. You will teach 2 courses a semester or less. You will mentor graduate students, have an active seminar program with many visiting speakers every year, and get to teach graduate-level courses. Starting research salaries, in philosophy, are usually about $75K, give or take $15K. At the end of your career, you can make $150-200K, or even more. There are philosophy professors at Rutgers earning around $300K. Economists make more than that–starting salaries are often over $100K. New assistant professors at law schools and business schools make more–assistant professors in highly ranked law and business schools can make $200K or more. In research jobs, your primary job is to publish in good academic journals and presses. Teaching is secondary. You can be a terrible teacher and still get promoted, get raises, and succeed, but you don’t publish, you’re out.

A teaching job means you work at an undergraduate-oriented program. Most of the time, you will teach 3 courses a semester, though it’s not uncommon to teach 4 courses a semester instead. Starting salaries are much lower: you might start at only $45-55K. End of career salaries are also much lower: you might top off at $90K. You will be expected to publish a bit, but most of your job will be teaching. You will mostly teach service courses rather than courses on the topics that most interest you.

Adjuncts have no job security. They make about $3000 per course. Often, they don’t get an office. They get no benefits. They might teach at three or four universities, trying to teach enough courses to make ends meet. The stories you hear about professors on food stamps are about adjuncts. Adjuncting as a grad student for extra money is fine. But you probably don’t want to be stuck being a “professional adjunct” teaching 8 courses a year at three universities for $24K for six years after grad school.

Post-docs are temporary positions for newly minted PhDs, in which the PhDs get low or no teaching responsibilities and can focus on pumping out publications. Post-docs are a good stepping stone to tenure-track jobs. I turned down multiple tenure-track job offers at good schools for a post-doc at a better school. It paid off.

What Do Professors Do All Day? At Georgetown, officially research is 60% of my job, teaching is 30%, and service is 10%. In any given week, the balance might be different, but those numbers do reflect how I spend my time. As a professor you read others’ articles and books, write your own papers, write grant proposals, do administrative work, meet with students, prep classes, teach classes, grade student work, and give presentations at conferences and other universities. The exact mix of these things depends upon the type of job you have and how famous you are in your field.

What Are My Chances of Success? I don’t know you personally. I have no idea how good you are. But let’s assume all the people reading this post represent a normal sample of people who begin Ph.D. programs. Statistically speaking, most of you will not get a research job. There just aren’t that many of them. Maybe 1 in 10 tenure-track professors have a “research” job. Also, unfortunately, most of you will not get tenure-track teaching jobs either. Only a minority of people who enter a Ph.D. program ever succeed in getting a tenure-track job. Instead, assuming you represent a cross section of the people entering grad school, about half of you will fail to get a Ph.D. Of those who get the Ph.D., half of you will spend five to six years with short term visiting gigs, post-docs, or adjuncting jobs, until finally, at age 40, you realize you aren’t going to get a tenure-track job, and you move on with your life. The odds are low. That sucks, but that’s how it is, and it’s not looking like it’s going to get better.

Thus, you should enter graduate school only if you are prepared to do what it takes to beat the odds, or if you’re willing to bear the consequences of failing to get a tenure-track job.

How Do I Beat the Odds? Go to the highest ranked program you can get into. The reputation of your school and your advisors matters greatly. It’s much easier to get a tenure-track job coming out of, say, NYU philosophy than Utah philosophy. Start publishing right away. You need to hit the ground running, and work on professionalizing yourself the moment you enter grad school. You shouldn’t write seminar papers for your professors; you should use seminars to help you write research articles that will be published in journals. You should plan to leave grad school with 4-5 good journal articles in hand. A good rule of thumb is that starting your second year of grad school, you should, at all times, have three papers under review at top journals, at least until you get tenure.

A student once asked me, “How can I be expected to write publishable articles as I’m learning the field?” My answer: “Get the syllabus from the professor before the semester begins. Read most of the materials before the semester starts. Come in on the first day of your seminar with a draft of your seminar paper. Then spend the entire semester revising that paper. At the end, you might have a publishable article.”Isn’t this advice really obvious? Doesn’t everyone do that? No, most grad students don’t. I’m not sure why. Perhaps they don’t want to be tested. Perhaps it’s scary to amass rejection letters. Perhaps they want to forestall finding out if they’re good enough to publish, much as a person might avoid asking out his crush, because it’s better not to know than to know the answer’s no. Perhaps their professors mistakenly tell them that grad school is simply a time to gestate and develop, and that publishing can come later. (That was true in the 1960s, but it isn’t true anymore.)

If every grad student were to take this advice–have three papers under review at all times starting their second year of grad school–then it would cease to be good enough advice. Instead, the bar for getting a good job would move higher. But the good news is that most grad students ignore this advice, even if they hear multiple times from multiple people. Of course, this won’t help you if you don’t have sufficient talent. But the sad thing is that many people with immense talent get kicked out of the academy because they don’t prove themselves early enough.Let me give an example. We had a job search in 2014-2015. We received hundreds of applications. The modal amount of time my colleagues and I spent looking at a dossier was about 25 seconds. If the person didn’t have any publications, we put him in the discard pile. If the person didn’t have good publications–publications with top journals–we put her in the discard pile. Now, it’s certainly possible that the most brilliant candidate got discarded in 25 seconds. But we’re okay with that. Search costs are high. Searches consume a great deal of time. I’m not going to go looking for a hidden diamond in the rough when I have a pile of diamonds right in front of me. I’m not going to go looking to see who has the most hidden potential when I have a pile of applicants with revealed potential right in front of me. (By the way, almost all of the applicants whose teaching dossiers I’ve looked at over the past eight years had excellent teaching evaluations and the like. So, don’t count on that making you look exceptional.)

Read this essay on Academic Saints. The author might be right that he or she deserves greater consideration and better opportunities. But the fact is he or she isn’t going to get it unless he or she starts looking more saintly. No one in academia cares that you’re a good person who means well and thinks deep thoughts. Search committee members’  time is valuable. Again, they aren’t going to look for hidden diamonds in the rough when they have piles of diamonds in front of them. If you want an academic job, you need to make it obvious to everyone that you are a diamond. As I said above, hiring committees are looking for people they believe will pass tenure review.

Many people think it sucks that grad students have to professionalize right away. They long for the good old days, in which so long as you went to Harvard or Princeton, you could just read all day without worrying about publishing, knowing that your advisor would call up his buddy at Cornell and get you a job. , in which grad students could just read and write without a concern for having to publish. Those days are gone. If you want a tenure-track job, you have to come out of grad school with good publications in hand.


What fields should I go to grad school in? Your odds of getting a tenure-track job depend on what field you get a Ph.D. in. The ratio of Ph.D.s to tenure-track jobs in the humanities (modern languages, lit, English, etc.) is very high.


If you’re reading this blog, you probably have interests in philosophy, econ, and poli sci. For the purposes of getting a job, you should do political philosophy instead of political theory, and economics instead of either. There are maybe 18 tenure-track jobs in political theory, and there are about 400-500 people (including a massive backlog of under-employed recent Ph.D.s) trying to get those jobs. A former post-doc at Brown got a 4-3 teaching load, $44K/year job at a lower tier liberal arts college. He was bummed about that until I pointed out  this meant he was in the top 5% of political theory applicants that year. In contrast, if you specialize in political philosophy, you’ll probably also have a specialty in ethics, and you can expect to apply for 60 jobs even in a bad year. Econ is better because there are more jobs, and you can also work in government or the private sector. You should do empirical, math-based work, rather than normative work. You have a much better chance of getting a TT job in American government than in normative political theory. Again, this is ceteris paribus advice. Do what you love. But be aware that your chances of getting a tenure-track job depend on what you specialize in.

Summary: Academia is glorious, but the odds of getting that glorious job are low. Know the risks and plan accordingly.

  • John

    The advice here is very sound. Jason was kind enough to give me roughly this advice before I started grad school and while I can’t say I did everything as well as I could have, I did graduate sooner than average and get a TT research job.

  • murali284

    Does getting a PhD in a British/European/Australian institution give you a better chance than your peers (who do their grad school in the states) because you finish in 3-4 years instead of 6-7 years?

    • Jason Brennan

      Good question. I was thinking of American and Canadian institutions above.

      The truth is that search committees know the difference, so they know that that people from British universities finish sooner. They thus adjust accordingly–they realize that getting out of Oxford in 4 years is not impressive, while getting out of Harvard in 4 years is. They realize that taking 7 years to finish at Oxford is a bad sign, while taking 7 years to finish at Harvard is not so bad.

  • Andrew Pearson

    Many thanks for writing this.
    “starting your second year of grad school, you should have three papers under review at top journals”
    What counts as a “top journal” in this context?

    • Jason Brennan

      It varies depending on your field.

      In political philosophy, this would mean Ethics, Nous, Phil Review, Philosophy and Phenomenological Research, the Journal of Political Philosophy, Australasian Journal of Philosophy, and a number of others.

      In economics, this would mean American Economic Review, Econometrica, Journal of Political Economy, and a number of others.

      In political science, it would mean American Political Science Review, the American Journal of Political Science, and a number of others.

      Your professors will be able to tell you what the good journals are. A good way to tell where to publish is to look at the reading materials for your grad classes. You will most likely be reading articles published in the leading journals. Try to publish in the same journals you are reading papers from.

  • Craig J. Bolton

    Very thorough and very accurate.

  • jimbino

    “Hiring committees are looking for people whom they believe will pass tenure-review,” in English, would read “Hiring committees are looking for people who they believe will pass tenure-review.”

    Excuse me, but if that level of grammar is what it takes to be a tenured prof, I’m in!

    • Jason Brennan

      Apparently not. The “they believe” part of the second clause makes it a “whom”.

      At any rate, that’s what copy-editors are for.

  • Jonathan

    How does age factor into the grad school considerations? All else equal, it seems like schools would prefer candidates in their mid- to lower-twenties. Will top schools consider career changers?

    • Jason Brennan

      I’ve heard many people claim there is age discrimination, but I don’t have any data. Nevertheless, I would think it’s at least strategic to presume that if you are an older candidate for an assistant professor job, because you started grad school later than average, you will have a harder time, and so need to publish a bit more to be competitive.

  • Pingback: Let’s See What’s in the News Today (Oct. 26, 2015) | Shaun Miller's Ideas()

  • Pingback: Graduate School – Political Science Career Seminar()