Is the economics job market worth it?

There is a new NBER paper on this topic by McFall, Murray-Close, Willis, and Chen, gated copy here.  Here are some key takeaways from the paper:

1. One-third of the job candidates in the sample were women.

2. More than one quarter of all job candidates on the market come from top ten institutions, which tend to have the largest Ph.d. programs.

3. 28 percent of new job candidates enter the market with some kind of publication.  The average candidate has served as primary instructor for one or two courses, plus as teaching assistant for more than three courses.

4. The five most frequently listed fields are labor economics, macro, IO, applied micro, and econometrics, each listed by 21-23% of the candidates.

5. 72% of the people on the market express a preference for jobs as assistant professor.

6. More than eighty percent of the job candidates “expected to place in the top half of the distribution for their graduate department.”

7. Although there is overoptimism, in terms of relative rank candidates have a decent idea of where they will end up.

8. Job candidates receive three offers on average (noting that only half of the candidates in the total pool responded, so there may be bias.  Three strikes me as a little high on average).

9. Number of publications predicts higher yield in terms of job offers, whereas gender, undergraduate school, having a Ph.d. from the U.S., and teaching experience have only weak predictive power.

10. As a candidate progresses through the process of interview, flyout, and the like, unobservable characteristics matter more and more for predicting outcomes.  This is consistent with the view that the process itself yields information, though whether that information is ultimately accurate as a predictor of success remains an open question.

11. Approximately 92% of candidates ended up with a job (!).

12. More than two-thirds of the candidates are “very” or “extremely” satisfied with their final results.

13. The average base salary for accepted jobs is $93,000.  The median base salary is $88,600.

14. The paper has many other results of interest.  As Bryan Caplan has previously observed, being an economist is a great life and a great career — do it!


I originally wanted to make a remark such as "So many young professional economists and so few new ideas."

Then I realised most new ideas are wrong.

Old ideas, too.

The average base salary for accepted jobs is $93,000. The median base salary is $88,600.

Those are high salaries for just recycling old, erroneous ideas.

The benefits of an economics Ph.D. are far more about the consumption and compensating differentials than the salary. People who can finish economics PhDs could also finish elite MBAs and make a lot more money applying the work of other people. Probably most people who get economics PhDs would even be making more money if they just worked right through from finishing their undergraduate degrees.

They'd also be competing against people who did in fact get 'elite MBAs' or worked out of undergraduate, so its a guaranteed 90k for life vs a potential salary. Nor is the salary stable, there is no tenure on Wall Street.

"Those are high salaries for just recycling old, erroneous ideas."

I'd wager that it's not more or less true than most fields, be it finance, sports management, manufacturing, or science Heck, the history of any subject is essentially the recycling of old erroneous ideas until that rare soul comes along with an innovation that's marginally better (but still not perfect), which then gets recycled until the next true innovator.

I'd say that one possible exception is religion, where innovation isn't as appreciated and the continuous recycling of old erroneous ideas is seen more as a virtue than a flaw.

Related concepts include tradition and institutional inertia. Both are widespread phenomena.

Actually we're paid well to ignore your stupid ideas.

Great prospects if you're willing to suck up to politically connected creeps like Brad DeLong. I've known too many economists who wish they had become heroin dealers and worked with a better class of people.

If you avoid the academic rat race, you can have a satisfactory and happy career. But then you are also stuck in mediocrity.

Stuck in mediocrity? Really? Some perhaps but I susepct they would also be working in mediocre departments -- a lot more of those than there are top departments even if they make close to the same pay.

But more importantly, for the most part economics is about explain what people do and the underlying institutions that exist to mediate those interactions. Those actually out in the economy creating those industries and relationships, as well as helping to form the underlying institutions via their interactions, are creating what economists then try to understand and explain (sometimes poorly and narrowly). That's hardly what I would call "mediocrity".

How many economists do you know? I mean real ones, not just just specters in your imagination.

It sounds like you hang out with the negative selection crowd.

worth what? what are you comparing to?

Anyone know why Tyler isn't discussing Gruber deceiving "stupid Americans" on Obamacare?

The surprise is the low base pay...having a college degree with computer science can easily bit that.

No it can't, not easily anyway I'm sure there are examples of SOME CS graduates starting at 90K but that's not even close, to close to the norm. Getting into Google is not even remotely easy.

erm.. really? I never got a CS degree but made more than that within a couple of years of learning programming.

Correction: I did not make more than that within the first couple of years, but made near that, and had friends who made more than that (with a CS degree). Also, that was during the boom, but I still hold that it is easier to make money as a programmer than as an economist -- and better still to have knowledge of both (financial products is where the most programming jobs are internationally).

People who can get PhDs in economics are presumably towards the right hand side of the CS abilities distribution. I'd guess they'd be in the top 20 percent. I caveat that I work with more econ PhDs than CS grads.

I meant to compare economics PhDs to people who enter the workforce with a CS undergrad degree, not CS PhDs, here.

I'd expect average econ PhDs to be comparable to pretty good CS undergrad graduates.

Even a half-decent cs undergrad degree would involve writing a compiler with some minimal optimizations and a mini os of some sort (Probably these days on raspberry). You think an econ phd without a cs background can manage that?

Or is it a case of the economist thinking they can actually do any job in the economy fairly easily?

Huh? No not at all. I just meant to compare on general brainpower and quantitative chops.

I expect people with PhDs in econ to be smarter than the average CS undergrad, but not by a huge margin. I expect people with PhDs in CS to be a great deal smarter than people with PhDs in econ, on average and at the extremes. But getting a PhD is a reasonably difficult thing to do even in a field like economics. You need pretty good math skills, even by technical field standards.

For what it's worth, though, most economists are actually employed as computer programmers with statistical skills. So the outcomes are not as different as you might think. Imagine having a little less engineering skill and a little more subject matter knowledge.

I was in that range for my first job out of undergrad back in 99. I was around 100k by my second year. Those numbers are frighteningly low for people who are presumably smart enough to be engineers. Not to mention the four or five years of foregone six-figure income.

I suppose they don't count bonuses, but academic economists typically don't get large bonuses, right?

"The surprise is the low base pay…having a college degree with computer science can easily bit that"

Probably, because it's academia and tends towards lower wages with higher benefits than the private sector.

Market Failure

A salary roughly equivalent to a Lieutenant Colonel with 20 years of service. Or a civilian manager of a multi million dollar enterprise. (Not an executive)

The officer has job security, but not job safety. The manager has job safety and no job security.

The officer in this case has a defined benefit pension but also has a high probability of not being selected for promotion and therefore forced to retire in a few years.

Interesting--the salaries are much much higher than in the humanities. A friend of mine who is an English professor at the associate level and who just published a book with Oxford makes $67k. His starting pay was $55k in 2006.

Major matters.

He's overpaid.

markets in generating elite-centric propaganda! Always a market for that!

Publications are a double-edged sword because they "reveal your type." If you are aiming high, no publication is better than a lousy publication (anything less than a very good field journal, or good general journal). But if you are aiming low, any legit publication should be helpful.

Most econ PhD's are not in it for the money, as evidenced by the strong preference for academic jobs. Jobs in for new PhD's in econ consulting firms start around $125K+, with non-trivial bonus potential.

The average of 3 offers seems low, based on my recent experience (on the hiring side of this market). The market for new PhD's is very strong right now.

Being an academic and consulting are not mutually exclusive. Academic consulting economists charge $650 to $750 an hour. Not bad for a days work.

In the long run, for the successful ones, sure. But it's a very long time before the typical newly PhD who stays in academia has any significant consulting income. The starting salary difference in consulting is only a small part, though, in the decision. Much bigger is the desire to do choose your research and to work in a less social fashion. Those suggest academia. If you're willing to do work that other people show revealed preference for (that explains the salary differential, by the way) and you hate working alone or with at most a couple of collaborators, consulting makes more sense.

Sounds like an economist who did not do time discounting for money, or is short on understanding short and long run payoffs, without including the benefit of tenure for one job but not the other.

Three offers sounds low to me as well (also on the hiring side of the market). The typical candidate I made an offer to last year had already rejected one offer and had one-to-three other offers in addition to ours. And money isn't a driving factor - every single candidate I met last year had a pretty high consulting offer and all of them eventually turned it down. (This feature of the market results in highly-entertaining phone calls from our HR staff, who cannot understand why anyone with such a high-paying job offer would even consider working for us.)

Since I hire in the government/semi-government research sector, I was most interested in their results regarding preferences; specifically they find that preferences over options largely dictates job outcomes. We lose a lot of candidates to university jobs, even colleges. Every job market cycle I make at least one offer to a candidate that turns me down for a low-ranked academic job (I completely understand being turned down for higher-ranked academic jobs), often against the advice of their own adviser. It's much easier to weed out job applicants on the basis of quality (too good, good, not good enough) than to try to suss out these preferences from the packet and interviews, but some years I hire no one at all, largely because I short-listed too many candidates that had strong academic preferences.

Not at all, Bill. For the vast majority of economists (though ignoring huge sample selection problems) consulting economists make more than academic economists (including their consulting operations) all along their career paths and, I suspect, all along the cumulative distribution functions. That said, there is clearly overlap between established academics and experienced consultants.

Jonathan, Admittedly sampling from a small sample size, my observation is that academic economists who also consult make far more than consulting economists who work for someone like Lexecon, for example. It;s also reflected in the billable rates. Someone has to do the dirty work in the background for the testifying academic economist. Consulting economists who work on their own, or who work for consulting firms, typically move around a lot, have less job security (why is it that consulting economics firms fail!), and are quite dependent on just a few clients or cases.

Your experience differs from mine.

Hire Lexecon lately?

Unless you've got a real top guy there are probably several people on the case team billing out above the expert. They are surely making more from consulting each year than the expert unless he's a real star, and vastly more over the course of a career. Most experts don't last all that long or dedicate that large a fraction of their time to it - it's a tough job.

This expert consulting thing is a nice late-career bonus for academic stars, but it's not a significant component of expected future compensation for graduating PhDs.

Is Anon at a university? Both Anon and DCEconomist say that three offers seems low, but DCEconomist is in a government/semi-government sector, not academia. I strongly suspect that candidates for jobs in the government and other non-academic sectors get more offers than candidates in the academic sector do.

Tenure track jobs are scarce and are usually viewed as the most desirable (even though their salaries are often lower than in other sectors). Tyler's estimate that most candidates get fewer than three offers may very well be valid in the academic market.

Those same candidates might not be applying for non-academic jobs at all, or might apply to just one or two as a backup plan, much as an academically strong high school senior might apply to the state's public university as a "safety school". (That was my strategy: I applied to one or two positions at Fed banks, and to the Urban Institute, and every other application was to a college or university.)


Numbers I witness in my department suggest a much more pessimistic picture.

Just thought I'd put it out there before everyone signs up for a PhD.

I'd be interested to see similar posts for other academic disciplines. For example, how about for new physics graduates? One difference is that the idea of getting a tenure track job right after graduate school is pretty rare in physics. The usual route is to do one and often two or more postdocs.

Economics is one of the few PhD job markets where the junior market 'clears' (i.e. almost all new PhDs get a full-time job on their first time out).

It's supply and demand, of course. Economics has pretty strong academic demand, partly due to B-schools, law schools, and public policy schools, and relatively low supply compared to humanities PhDs, particularly of students with undergraduate universities in the U.S. (the paper notes that only 1/3 of PhDs in economics were granted to someone who completed their undergraduate degree in the U.S. - many of the other 2/3rds find jobs outside the U.S.). There's also healthy non-academic demand from consultant groups, research non-profits (e.g. Rand, Urban, Mathematica), the Fed, the IMF, and government (e.g. Treasury, CBO, BLS, Census).

I suspect the main reason the market for scientific PhDs is much worse than economics is the demand-side. Government funding for scientific research is way down from the 1960s and 70s, student enrollment in scientific fields is down. Science departments are relatively expensive to maintain, while MBAs are cheap to produce. But I'm speculating somewhat here - I haven't seen any research outside my own field on the factors impacting the market for new PhDs.

It's strange that such an imbalance persists. Getting a good academic job in physics is really, really hard, even if you come from a very good program. The idea that you would even be considered right out of grad school for tenure track job is almost laughable. At the same time, I'm sure research in economics can be just as interesting (or uninteresting as the case may be) as research in one of the physical sciences. I also suspect that most people who go into the physical sciences with the idea of pursuing an academic career are more interested in the academic "lifestyle" (self directed research on topics of interest, a high degree of independence, interesting colleagues and students, etc) than they are in the exact topic they are working on - whether its polymers or poverty probably doesn't matter so much to their ultimate degree of happiness. So why hasn't there been a wholesale migration of would be physical scientists to economics to even out the imbalance? (I suspect one reason may be that as a young grad student, the reality of the job situation in the physical sciences doesn't hit home as hard as perhaps it should - young people just starting a program are probably unduly optimistic about their prospects. )

Possibly the most expensive whores on the planet, I would say.

Not so long as there are still lawyers.

The lawyers are the Johns.

Better than the least expensive morons, otherwise known as the Yancey Wards.

Ah, I have cut an economist to the bone, it appears.

I have several friends who got PhDs in physics from a top-5 or top-10 US dept and the job market stinks. They spent 4-5 years in postdocs. Then one got a good academic job (though not in physics), one got a glorified lecturer position (good University but no tenure track) and the other quit academia and got a regular job. Others similar but forget specifics.

Here's one on marketing PhD job candidates from 2011:

Business school PhDs typically have about 50% higher starting salaries than their econ counterparts.

Which is good for me, because that's what I'm doing. So Econ PhDs...stay away from business schools. Don't apply there! ;)

The interesting thing is that B-schools figured out a bit better than econ schools that limiting the supply of PhD candidates is a good way of getting higher salaries. Econ schools seem to produce far too many PhDs, which then leads to disproportionately lower salaries. B-schools intentionally produce few (except for the OB/OT/IO Psych people who can't seem to grasp supply and demand)

PS: For people who say this isn't about the money...of course it's about the money! I'm doing this because of the money! No question in my mind.

It's about the money, relative to the work-life balance offered by the position. People who turn down finance/consulting jobs don't do so because of the money. They do so because the amount of work is brutal.

The PhD in economics is like winning the lottery. Lots of tenure track jobs doing research. Intellectually, it's unbelievably stimulating. It's difficult so you spend your whole life doing something humbling but rewarding. I wake up every day, and have for years, deeply grateful for the life I got to have. At least professionally, I have the best job in the world, and everyone I know in economics feels approximately like that. I don't regret it for one minute.

Paper would have been more insightful if it regressed salary, offers on grade in real analysis.

And yet, they produce little

you wrote: " Approximately 92% of candidates ended up with a job (!)."

I added " retail."

" More than eighty percent of the job candidates “expected to place in the top half of the distribution for their graduate department.”"

Is there a law, like Pareto's law, which says 80 percent of participants in any endeavor think they're above average? To my surprise, I find wikipedia has it:

At least it's not 1996, the trough of the job market. I got my Ph.D. then. I had 2 interviews at the ASSA, one informational, a high score. A friend got his Ph.D. from a top 10 school and ended up in Israel for a lower salary and higher cost of living. I finally got a job in 1998, in demography. I got it because of my work experience and the employer was more desperate than I was. I'm still doing demography. The joke is that demographers are sociologists who can count. I don't know anything about sociology, but I do know how to count. This makes for lots of professional conflict.

Comments for this post are closed