Bryan Caplan  

Three Laws of Major Mismatch

Scott Winship on Inequality... Tax Hikes and Spending Cuts: F...
In Being John Malkovich, anti-hero Craig Schwartz whines, "Nobody's looking for a puppeteer in today's wintry economic climate."  The Great Recession has made a lot of college graduates feel like Craig Schwartz.  If you major in philosophy, history, or puppetry, you probably won't land a job as a philosopher, historian, or puppeteer.  If philosophy, history, and puppetry majors find a job, it's usually an unrelated field.  Labor economists call this "mismatch."

This week I've read all the leading academic research on mismatch.  The single most enlightening piece: John Robst's "Education and Job Match: The Relatedness of College Major and Work" (Economics of Education Review 2007). 

In self-reports, roughly 55% of college majors say their job and major are "closely related," 25% say "somewhat related," and 20% say "not related."  These self-reports probably understate the severity of mismatch; after all, it's embarrassing to admit that you couldn't find a job that uses your skills. 

In relative terms, though, self-reported mismatch seems meaningful.  Robst reports three robust laws of major mismatch.  The first two are precisely what you'd expect:

1. Vocational-sounding majors like engineering, computer science, health professions, and architecture have low mismatch; "Mickey Mouse" majors like English, foreign languages, philosophy, and liberal arts have high mismatch.

2. Mismatch leads to lower income.  People with "somewhat related" majors earn 2-3% less.  People with "not related" majors earn 10-12% less.

The last result, though, puts all the empirics in a new light:

3. The more vocational your major, the larger the negative effect of mismatch on earnings.  The least vocational majors suffer little or no financial harm from mismatch:
Individuals who majored in business management, engineering, the health professions, computer science, or law all face more than 20% wage penalties for working outside the field of study. The wage effects are insignificant in liberal arts, English, and are statistically significant but small in the social sciences and education.
Philosophy/religion/theology majors average 20% higher income if they don't use their major on the job!

For "Mickey Mouse" majors, then, mismatch is a red herring.  Mickey Mouse majors aren't imprudent because you're likely to suffer from mismatch.  They're imprudent because Mickey Mouse majors earn markedly lower incomes whether you're mismatched or not.

Comments and Sharing

COMMENTS (31 to date)
Andrew writes:

Been looking at identifying mismatch for a dissertation chapter, summary is that it's a way overblown issue. I think it's important to distinguish that there's a difference between mismatch and having skills that aren't valuable or have changed in value.

Chris Koresko writes:

Andrew: ...There's a difference between mismatch and having skills that aren't valuable or have changed in value.

Agreed. A simple way to interpret Robst's results would be that "vocational" majors are simply more valuable than "Mickey Mouse" majors. Then vocational majors have more incentive to find work in their field of study. A mismatch to useless skills costs nothing.

(To be clear, I'm speaking only of economic productivity here. I do not believe that philosophy or religion are useless -- quite the contrary, if they are done well.)

Silas Barta writes:

Along with what Andrew said, I've noticed it's pretty common for "Mickey Mouse" majors to complain that "they can't possibly know what will be in demand in 4 years", while majoring in something that would never be in demand in any case, and which would teach and practice useless skills (and meta-skills, for that matter).

I saw the unsustainability of the trend of getting useless degrees long ago, reasoning that nothing prevents employers from giving up on that whole selection method. Why didn't more people, and much later?

Ken P writes:

I had two degrees, Math and Philosophy. I ended up in the Biology field. I prepared for the interview by spending a weekend in the library, focussing on learning biological concepts most pertinent to the company. I went on to get my Master's degree in Biology. That worked out well for me. I've met many people since then who used a Masters to focus on the field they ended up in. It's sort of like the entrepreneurial concept of a pivot.

Bostonian writes:

I expect that Professor Caplan's book on education will discuss the extent to which the earnings premium of college graduates is due to their personal characteristics upon entering college. Another interesting question is how much of the engineering-over-humanities premium is due to the qualities of the students who major in those subjects.

Pat writes:

"The more vocational your major, the larger the
negative effect of mismatch on earnings."

If you have a vocational major and pursue a job in that field but end up in a mismatch job, there is a good chance there is something about you that is very unappealing to employers in your preferred field that has nothing to do with eduction.

Dave writes:

This argument comes from a paradigm of learning specific "just in time" skills instead of learning general "just in case" skills.

I run an Internal Audit function at our company, and have found over time that those candidates who specifically prepare for the job (generally through an accounting degree and, perhaps, a CPA accreditation), are far more prepared in the short-term to contribute on the job. However, in the long-term, the most effective internal auditors are those who are curious, well-read, outgoing, thoughtful people. (Often, this is exactly the opposite of accountants.) As a result, hiring someone outside their field can work over time, but I wonder how many employers have a long-term view instead of only concerning themselves with "butts in seats" in the current moment.

Finch writes:


The idea that folks in softer majors do better later in life because they have some sort of hidden ability is common, but it's a myth. The advantage of smart quants persists throughout their careers.

Counter-intuitively, this makes me think, as Bostonian suggests, that it's just a quality difference in the people who choose these subjects. (And in the case of engineering, make it through the more rigorous program.) Does the fact that you had to get a good grade in complex analysis really matter when you're 50 and haven't done it in 30 years? Or is it that the kind of person that can get a good grade in complex analysis is going to be a better employee at 50?

Sure, a chemical engineering degree is necessary to even approach the job on day one at Dow, but that effect must wear off, right? Could it be that path dependence is that strong?

bryan willman writes:

Many of the "stem" topics involve lots of reasoning by analogy, experiment of some sort, and so on.
There is a line of argument that creativity and intelligence are partly rooted in such thought processes.
Being widely read and curious can be a multiplier most anywhere.

But how much do these poperties of the upper echelons of stem, or i suppose most anything else, apply to the masses?

Tracy W writes:

If you have a vocational major and pursue a job in that field but end up in a mismatch job, there is a good chance there is something about you that is very unappealing to employers in your preferred field that has nothing to do with eduction.

Or that you found something very unappealing about the field, and prefer to take a lower wage in exchange for more utility.

Hazel Meade writes:

Well, we have this terrible student loan policy where we hand out money to absolutely everyone, regardless of whether they are majoring in puppetry or neuroscience.

If student loans were awarded in a private market it would be very, very, hard to get someone to lend you $50,000 to study puppetry. It would be considerably easier to get $50,000 to study electrical engineering, provided your grades were up to snuff.

But with the federal student loan program, the amount of money you get awarded is determined solely by financial need. It is pretty much an automated process, and there is literally no consideration of your prospective income after graduation.

So you end up with a substantial number of students who decide to take puppetry (or something to that effect) because it sounds like fun, eschew hard subjects like math and science, because they are more focused on getting laid than on making money several years after graduation. And then they end up with no marketable skills.

If we awarded loans more like a private market, where the focus is on recouping the "investment", we would be steering kids into careers where they can earn a living, instead of letting them study whatever random subject appeals to their 18 year old brain.

djf writes:

I don't know about puppetry, but history and philosophy majors have never been able to get jobs in those fields with just a bachelor's degree. You need a graduate degree to find work as a historian or philosopher (and, then, of course, you still might not be able to find a job). But I don't see that the situation for these kinds of majors has changed in any fundamental way.

I suppose maybe at one time, a person with a just a B.A. in history or philosophy could get a job teaching high school, but I don't think that's been the case for quite a while.

Mike W writes:

"Mickey Mouse" majors like English, foreign languages, philosophy, and liberal arts...

and economics(?)

biagio writes:

"These self-reports probably understate the severity of mismatch; after all, it's embarrassing to admit that you couldn't find a job that uses your skills."

There is also the case of people who could find a job but want to do something else. Who hasn't realized in the middle of his/her PhD that academic life will not suit them and yet powers through only to move to something else?

Also it is difficult to define what an unrelated field is. I have a PhD in Physics and, like quite a few of us, moved onto quantitative finance. For me the two activities are closely related: doing numerical research in one field is very similar to doing numerical research in the other. The environment, I admit, is different. Yet for some people there couldn't be two more different fields than Physics and Finance. What am I to say if asked the question about mismatch?

Jack P. writes:

I'm surprised you don't argue signalling here, Prof. Caplan. The problem with "Mickey Mouse" degrees isn't just job mismatch, but also that there is huge grade inflation, so earning a degree conveys a weaker signal than does a degree in STEM fields, where you do not get points for effort (ask my brother).

Anybody can get a degree in communications. It doesn't tell me, the employer, much about the applicant's skill, work ethic, etc. But if you get an engineering degree, you've had to sweat and persevere.

Mike W writes:

Would one of these be considered a "vocational" major and the other a "Mickey Mouse" major to pundits giving advice to college freshmen?

The BA in economics is designed for students with a strong interest in the liberal arts. It is appropriate for those who prefer a less quantitative degree program than the BS in economics and may be especially appropriate for students planning to attend law school or graduate programs in business or public administration.

The BS in economics is designed for students who desire a more technical program than the BA, one with a stronger emphasis on economic and quantitative analysis.

Would a BA in Economics be generally considered less "Mickey Mouse" to the employers of future business managers, public administrators or attorneys than a BA in Philosophy?

zgatt writes:

It doesn't seem difficult to make a strong argument that lower wages, Mickey Mouse degrees, and mismatch are all effects of imprudence, and the burden is on disproving conditional independence.

William Newman writes:

Finch, I don't know how important it is in practice, but I doubt the effect of ChemE training wears off completely even in decades of practical experience. The ChemEs will have the math that I think of as the operations research toolkit (not because studying operations research is the usual background for people who have it, but because in the real world it ends up being used by people of all backgrounds to solve problems that remind me of operations research). ChemEs will know calculus, linear algebra, Fourier analysis, and statistics. People with this toolkit can think about practical problems outside their field in ways they can easily express to each other, ways they can't easily express to people without the toolkit. (How effective was that sales campaign? Or if we build a manufacturing facility which is a few percent smaller than optimum, should we expect that to affect profits by a few percent, or tens of percent, or a few tenths of a percent?) Few people informally pick up enough calculus on the job even to understand how functions around a minimum tend to be quadratic. Even fewer people pick up calculus plus basic linear algebra plus Fourier analysis plus basic statistics on the fly, and without those it's hard to talk (and possibly not so easy even to think) about things like transients and uncertainty and variations on different time scales. A ChemE will be able to think and talk routinely about a price having a variance and a mean. From my limited anecdotal sample, non-mathematically trained people who have given financial advice professionally for decades often seem never to have figured out the concept.

dave smith writes:

Pat started a line of reasoning that might be a counter point which supports the signaling (Bryan's favorite) model of schooling.

But these results also might favor the human capital model of schooling. If you major in business, you earn more working in business because perhaps you learned more.

Eric writes:

@ Finch: Mar 28th 9:19 and @Dave: Mar 28th, 9:02 am

I agree with both of you, if possible :-). I am a quant guy who used to play college bowl (a wide-ranging knowledge based quiz game with few and mostly easy math or science questions). I would divide the intermural teams/players into three categories:

1. The absolute best. They could be any major (quant or non-quant) and were often law students. Finished in the top 10%.

2. The best to middling. Quants. Math and Science majors. These teams finished in the 50th to 90th percentiles.

3. Low. Nonquant majors who weren't among the absolute best. Rarely did a team of quants finish here.

My conclusions: Quants are, as a rule, more intelligent, better read and more well-rounded than other majors and much more than they are given credit for. On the other hand, the absolutely best come from anywhere (although I do believe that it is far easier to take a math geek and teach him history than it is to take a history geek and teach him calculus and statistics. This would show up if college bowl included college level mathematics and science questions in addition to the college level Literature, History and Art questions).

Mike W writes:

From William Newman's post:

...if we build a manufacturing facility which is a few percent smaller than optimum, should we expect that to affect profits by a few percent, or tens of percent, or a few tenths of a percent?

Once the "vocational" majors arrive at the range of possible results this question becomes a judgment call about the future and an issue of who can influence the decision. Is the *quant* more likely to carry the argument to a decision than is the Marketing manager who majored in Psychology or the General Manager with a degree in Political Science? Where is the most value the number crunching or in the politics?

Dave writes:

@Eric - I think it is possible to agree with both of us. I agree, generally, with @Finch.

The point is that there are two sides of the equation. There's the supply (the employee, with his education, experience, skills) and the demand (the employer, with his desired skills, desired education, and desired experience). When you only match a desired educational credential to a candidate's degree, it's hard to determine if that's relevant or not. Having a degree in economics signals that I understand economics to some minimal level. But someone holding a degree in English Literature could understand economics better. It's just not signaled on his CV. And that's to say nothing of each candidate's potential for further learning on the job.

Aside from professions where an express educational endorsement is required by regulation (e.g. teaching, nursing), employers would benefit from competing for the best minds from any field. (And this is, in fact, an argument for why perhaps even those fields should lift such barriers to entry.)

If employers behaved that way, then I speculate that mismatches would not be so vastly underpaid. However, the reality is that most employers rely heavily on such signaling and resumes that don't match desired education are promptly discarded.

English Professor writes:

Most liberal arts majors who are successful outside academia tend to believe that their college experience taught them to think critically, to argue intelligently, and to write clearly. As a result, many of these will say that their major is in fact "related" to their job. They might not be analyzing Shakespeare or discussing Kant for a living, but they're using the basic tools they acquired in school. Such people often do well in corporate structures.

Nevertheless, with rampant grade inflation in most liberal arts subjects nowadays, an employer can't count on a recent college graduate's ability to think, argue, or write up to the old standards. The "signal" conveyed by his degree is thus being distorted. So I expect the college premium to decline for these students. The reputation of the school that the person attended will become even more important over time as a new proxy for what the student might/should have learned.

Eric Rall writes:

I am curious about whether someone with a mismatched "vocational" degree makes more, less, or about the same than someone with a "Mickey Mouse" degree (mismatched or not).

Finch writes:

> I am curious about whether someone with a
> mismatched "vocational" degree makes more, less,
> or about the same than someone with a "Mickey
> Mouse" degree (mismatched or not).

You can get an indication from looking at the census document I cited upthread. Generally the vocational degree makes more. For example, someone with an Engineering degree will have synthetic lifetime earnings of $2.1M in "Office Support," while someone with an Arts degree will see only $1.6M.

Jack P. writes:

I could be wrong, but I think liberal arts used to be more promising a generation or more ago, when the graduates were being groomed for careers in banking, law, medicine, etc. Studying liberal arts was the upper class thing to do, but the career would be professional, involving further study (law, medicine, etc.).

Today we have pre-law, pre-med, business-finance undergrad degrees, so by studying liberal arts (except perhaps in top-5 universities) you are really saying you want to make a living as a writer, actor, etc.

Mark V Anderson writes:

I believe strongly in the signaling theory, partly based on my own behavior when I hired accounting clerks 20-30 years ago. I looked highly favorably on those with 1 or 2 year accounting degrees, just because to me it signaled interest in the job. It eliminated those who just applied for any job they saw in the paper. I didn't need any accounting knowledge. I needed some quantitative skills, but I gave a test to judge that.

I suspect that pretty much every employer that requires a four year degree is doing the same thing as I was, if not always as purposefully as I was. There are very few positions that one can't learn on the job. Taking classes was for the purpose of determining if you could figure out hard questions, not to teach you that stuff. So the education is just as valuable for the 50 year old as the 22 year old, because the purpose was just to prove you could think (although the 50 year old may have proved that by experience also).

Perhaps there are a few things that are difficult to learn on the job, as someone suggested for some calculus areas. But these are usually minor parts of the degree. I know that for my four year degree (Accounting), I could easily have learned everything I needed to know within a year. I bet that is true for all currently existing four year degrees.

Bostonian writes:

"I know that for my four year degree (Accounting), I could easily have learned everything I needed to know within a year. I bet that is true for all currently existing four year degrees."

That may be true of accounting, but I could not have learned the physics and associated math for my BA in one year.

Someone from the other side writes:

There are a couple of companies out there who will hire the top grads of whatever field they can get their hands on after putting them through their own wringer, of course. I am thinking the likes of the Big3 consultancies, tier 1 tech companies (Google above everyone else) and a couple of other players (G-S to some degree, at least)
Coincidentally or not, they are all extremely successful at what they do...

I also agree on the idea that is easier to teach history to a math whiz than vice versa (I more than once thought that advanced macro would be so easy to understand to anyone with a BS in math whereas all of us economist where struggling with it because of the math)

Disclaimer: Top3 consultant with a MA in Econ (as well as a top tier MBA).

Hunter Roberson writes:

When I read this it immediately made me think of my uncle. My uncle is a newspaper sports writer in Brevard and he is also a fly fishing guide on the weekends. From what I can remember, he has always been willing to help me with my English homework. English and writing stories are what he is best at. But, what I hate to see is how much money he makes for doing something that he loves. He has a hard time making enough money to provide for both his sons and his wife. I feel like people should make more money if they are doing what they truly love to do.
This also made me think about one of my friends from high school. He went to college for four years and he majored in History. He graduated last year and now he is working on a farm in Asheville. His major and his job now are totally unrelated. Now I am an accounting major and it scares me that I will have put all this work into accounting and when I graduate I might have a totally unrelated job. But, you never know until you get there.

Chris writes:

You must read


Comments for this entry have been closed
Return to top