College has been oversold

Here, drawn from my new e-book, Launching the Innovation Renaissance (published by TED)  is part of a section on college education. (See also the op-ed in IBD)

Educated people have higher wages and lower unemployment rates than the less educated so why are college students at Occupy Wall Street protests around the country demanding forgiveness for crushing student debt? The sluggish economy is tough on everyone but the students are also learning a hard lesson, going to college is not enough. You also have to study the right subjects. And American students are not studying the fields with the greatest economic potential.

Over the past 25 years the total number of students in college has increased by about 50 percent. But the number of students graduating with degrees in science, technology, engineering and math (the so-called STEM fields) has remained more or less constant. Moreover, many of today’s STEM graduates are foreign born and are taking their knowledge and skills back to their native countries.

Consider computer technology. In 2009 the U.S. graduated 37,994 students with bachelor’s degrees in computer and information science. This is not bad, but we graduated more students with computer science degrees 25 years ago! The story is the same in other technology fields such as chemical engineering, math and statistics. Few fields have changed as much in recent years as microbiology, but in 2009 we graduated just 2,480 students with bachelor’s degrees in microbiology — about the same number as 25 years ago. Who will solve the problem of antibiotic resistance?

If students aren’t studying science, technology, engineering and math, what are they studying?

In 2009 the U.S. graduated 89,140 students in the visual and performing arts, more than in computer science, math and chemical engineering combined and more than double the number of visual and performing arts graduates in 1985.

The chart at right shows the number of bachelor’s degrees in various fields today and 25 years ago. STEM fields are flat (declining for natives) while the visual and performing arts, psychology, and communication and journalism (!) are way up.

There is nothing wrong with the arts, psychology and journalism, but graduates in these fields have lower wages and are less likely to find work in their fields than graduates in science and math. Moreover, more than half of all humanities graduates end up in jobs that don’t require college degrees and these graduates don’t get a big college bonus.

Most importantly, graduates in the arts, psychology and journalism are less likely to create the kinds of innovations that drive economic growth. Economic growth is not a magic totem to which all else must bow, but it is one of the main reasons we subsidize higher education.

The potential wage gains for college graduates go to the graduates — that’s reason enough for students to pursue a college education. We add subsidies to the mix, however, because we believe that education has positive spillover benefits that flow to society. One of the biggest of these benefits is the increase in innovation that highly educated workers theoretically bring to the economy.

As a result, an argument can be made for subsidizing students in fields with potentially large spillovers, such as microbiology, chemical engineering, nuclear physics and computer science. There is little justification for subsidizing sociology, dance and English majors.

College has been oversold. It has been oversold to students who end up dropping out or graduating with degrees that don’t help them very much in the job market. It also has been oversold to the taxpayers, who foot the bill for these subsidies.



There's too much information in the blogosphere and the Khan Academy to explicitly require college for anything.

What we need is a solid way of signalling domain competence without paying for a degree.

Good luck, that won't be easy. The ability and incentives to cheat would be massive if the Khan Academy actually was able to confer a degree. I like Khan and have been using it to help brush up on my math skills, but it is being oversold.

I think a better course of action would be to significantly decrease and in some cases eliminate, subsidies for education in non-science related degrees.

They already remove financial aid packages for programs that are not deemed 'in-demand' by the state. At least, in my state (NJ) -- and I'm talking about a public institution.

Federal loans and grants still go to everything.

oh wow, you had me till ur last statement... how can you understand the concepts and then say "for education in non science related degrees" just end ur statement at "subsidies" and let the market decide which degrees are worth what

Yes. Please do that. -- Sincerely, China.

Yes, but as Tabarrok implied some degrees, help create innovations that can spur economic growth, most if not all of these degrees, are sciences. Under this line of reasoning, we would want to subsidize people earning science degrees, because the benefit of people earning science degree's, affects more than just the holder of the degree. It raises income's for other's as well.

It's been easier for me to get a job as a self-educated developer than it has been for any engineer I've known yet to get a job. Not to spit in the faces of STEM degree students, but I regularly excel past graduated students in the workplace as I spent my 'college years' working lower class jobs as management.

Honestly, it's harder to start(60+ hours a week reading and doing google research to learn *nix and how the web basically works, and that was just the first 4 months) but, I had no pressure of failure, no debt, and am generally as well paid(in my current case, better paid) than my degree holding co-workers.

I couldn't say the say for some specialized fields(pneumatic engineering? probably not open sourced) but for many things, college is a overpriced waste that you'll never recover from.

Computer programming is a classic example of a case where you would expect a big payoff, but it ain't necessarily so.

Demand so exceed the supply of competent programmers that a person who knows how to think can do quite well absent a formal college education.

Yes, you would expect a big payoff and many of my peers who have programmed successfully for decades were getting those big payoffs until the trend to hire more young foreigners with H1B visas reared its ugly head. Now many companies won't hire older more competent American programmers because they can't work them 60 or 80 hours a week in a sweat shop atmosphere like they can the foreigners who are beholden to the companies for their visas. It is modern day indentured servitude.

Not according to many of the companies that hire software programmers and engineers. They say that they don't have enough qualified programmers, but from my experience they really mean they don't have enough cheap young software programmers that they can work 60 hours a week. Many of these companies don't like experienced employees who push back when asked to work harder not smarter. I work with the largest financial, insurance and telecom customers implementing J2EE software to pay employees and partners and in many cases they have offshored their IT and cannot even get simple things done like set up accounts or databases with the proper permissions or download the correct files. In this case, then I guess if they couldn't get these things done before with expensive Americans it might be cheaper to be incompetent with off-shore resources, but I think a better solution would be to actually get competent people.

Well said! STEM subjects should be economically attractive, but salaries have been flat. Careers have limited duration before one is deemed to be obsolete, i.e. having too high an hourly rate. Students have recognized this and figured that it often isn't worth the extra work up front. They are rational economic actors.

If you haven't studied algorithms and data structures in particular (linked lists, big O notation, graph traversal, things like that), I highly recommend doing that in some fashion (i.e. not necessarily formally). There are also a variety of things you cover in a CS degree that you wouldn't need to build out web apps, but you would need to, say, build search engines for Google.

I'm also self-taught. While it's worked out for me financially, part of me still wants to go back to grad school and learn about how computers work at a more fundamental level.

Needless to say, my music education degree has not been very useful.

Hi D, for people like you who are already competent enough to be gainfully employed in the field, but curious enough to want a deeper level of knowledge, I always recommend the book "Code: The Hidden Language of Computer Hardware and Software" by Charles Petzold - - it will probably just pique your curiosity even further, but you'll have a good background for any further studies you pursue.

I got my BSEE degree 25 years ago from Purdue. I learned how to build computers and program in 1s and 0s. Unfortunately, I remember PASCAL as well as I remember my Spanish. If you do not use it, you loose it. Education is something you need to do for a lifetime and continually be reinforced.

What really sad is that my pure-math formal education often means that I have a more accurate intuition than those with a college degree. I'm still smarting over an interview a few years ago when the interviewer claimed that hash algorithms run in constant time.

I would love to hear a more detailed description of that interview.

Maybe the best strategy is to nod and say something like "Sure, linear hashing." Agreeable disagreeing can be a wonderful business skill.

I often wish I'd studied more math, though it isn't all that applicable to my work, which tends to revolve around database querying and business analysis. But the Stanford AI class forces me to unpack some concepts that I don't have a lot of familiarity with but find very beautiful albeit time-consuming, and who knows, someday I might get into some real CS algorithm work.

If you are using hashes to search for things, they generally do run in constant time. (You still need to deal with hash collisions, which for a sufficiently bad hash algorithm could dump you right back to something like an O(log n) search along a binary tree.)

If you are calculating the hash of a blob of data, that's obviously not constant time. But people usually care a lot more about the former than the latter. If the interviewer was explicitly asking how long it takes calculate the hash of an arbitrary object then they were wrong.

The interviewer was correct in the context he probably intended it. Lookup time in a hash table is constant (amortized O(1)). Further, if you have a fixed set of data (i.e. read-only), you can compute a perfect hash function that has a lookup time of strictly O(1). See gperf for an example of an implementation of this.

Perhaps you were thinking about computing the hash value of a data sequence, rather than the lookup time once a hash has been computed. Computing a hash value for a given sequence of data is linear (O(n)) with the size of the data being n. However, the time complexity of this is generally not interesting, since any lookup algorithm requires an O(n) iteration over the blob at some point.

I was hoping I wouldn't have to explain myself. The context was open-ended, so my intuition went to the truly unbounded. In which case any constant runtime algorithm will necessarily have collisions.

I was so befuddled by his claim I didn't bother arguing. There are a lot of people who, having been taught something is true, cannot comprehend the possibility that it is not.

There are a lot more jobs building web apps than search engines. Also Google does not hire engineers without CS degrees, for good or ill.

You just need to read a discrete math textbook, SICCP, and supplement it with some Wikipedia. If you need more, you can always go on Google scholar, and read the literature, and/or look how things are done in open-source software.

I agree, the availability of such resources is one of the most wonderful things about this day and age.

I can think of something you'd still need college for: Teachers.

Education isn't simply a question of harvesting information. You could read Wikipedia articles all day for a century and still emerge as intellectually immature, dishonest and narrow-minded as when you started. You will be unable to express yourself on paper or in conversation, unable to reason, unable to identify errors and - to quote Cardinal Newman - lack command of your own powers.

A good teacher is able to steer you towards those abilities. Watching a video generally ain't.

"A good teacher"

Haven't met one.

That really sad, but then you likely did not attend worthwhile educational institutions...

Many people do not have a choice in their educational institutions, courtesy of our public school system.

Sucks to be you.

I attended median institutions. Even the best teachers were simply demanding and I did all the work.

Okay, let's try it this way, what percentage of your teachers were superior to the standard of doing your own thing?

About 1/4 of my middle- and high-school teachers were very very good.

Many were middling.

A few were actually destructive of knowledge.

Too bad college teaching programs do not produce good teachers.

Some do, some don't. Perhaps most don't.

Surely it depends on the person being trained, too. I always have a number of would-be teachers in my classes. What I'd like to see are more of the curious, smart and dynamic kids go into teaching. Wouldn't we all...

The problem is that the education department is a sewer of academia. Your average education major scores lower on the verbal part of the GRE than the engineers and lower on the math part than English majors. Elementary ed in particular is a dumping ground for MRS seekers. The saying that "Those who can, do. Those who can't, teach." is sadly true more often than not.

On first read through, I thought you were arguing that we need college to train teachers, on second read through I think you're arguing that college provides the teachers.
I can think of two issues here - firstly, to get access towards a teacher doesn't necessarily require going to college full-time. Secondly, some subjects can be engaged with more directly - the reason that people sometimes learn to be very good programmers outside of university is that they spend vast amounts of time for years trying to get the computer to do something, which means debugging, which means you get really good at reasoning about and identifying errors in code.
And one can engage in debates on the internet. Obviously a fair few people do this and remain intellectually mature, dishonest and narrow-minded as when they started, but then the same is true of people who attend university.

Also, the better teachers I've had, did college make them that way? If it did, then why so few?

No, college presumably didn't make them that way. What the colleges CAN do, and I wish they would, is select better applicants. I can't tell you exactly what makes a good teacher. But I can damn well tell you what is present in good teachers: curiousity, some smarts, and enthusiasm.

My position has always been that even if you are lacking in a number of areas, whatever these may be, IF you are curious and dynamic, you can compensate for other shortcomings...

Unfortunately, teaching colleges have plenty of students who are lacking, especially re: enthusiasm/dynamism.

I would like college to provide the (good) teachers for its students. Naturally, I don't think this necessarily happens.

I take your point about not needing to be taught by a human all the time. But what about programming... Does self-teaching really equip you as well as a college course? You program in isolation, when real life will put you in an organisation. You learn bad habits. You spend a lot of time solving problems that have already been solved. You have no critic.

As for learning to debate on the internet... In reality, that usually involves attempting to obliterate the ideas of one's opponent with one's own ideas. I rarely see a Socrates simply asking questions with a view to getting his or her pupils to think instead of echoing other people's thoughts.

The Oxford tutorial is a useful present day example of this Socratic aspect of teaching. I don't think it's sufficient on its own, or affordable for all, or even that it always works. But it's an ideal.

> to explicitly require college for anything

ANYTHING? Really? Not civil engineering? Not medical physics?

How about certifications? You don't need to go to university to become a lawyer, you just need to pass your local bar exam. I think that would be the idea. I think the idea would be to replace degree with certified tests and certificates.

Except that a lot of certifications are by bodies of practicing professionals. Conflict of interest?

That's appropriate in some fields. In others, certification has simply become the "union rules" of the 21st century, a form of protectionism to keep otherwise qualified and well-educated people out of the field so demand outstrips supply. Certification is not a solution by itself.

In point of fact, only California, Vermont, Virginia and Washington allow persons who have not graduated from an approved law school to take the bar exam in those states. New York requires that applicants who are reading the law must have at least one year of law school study. Maine allows students with two years of law school to serve an apprenticeship in lieu of completing their third year. All of the remaining 44 states require graduation from an approved law school.

I sometimes wonder if increased college attendance isn't a cause of Tyler's TGS, rather than a cure for it.

I definitely believe that increase college attendance leads to the pursuit of useless degrees.

Somewhere Alex's analysis seems to be arguing that college seems to be migrating from investment to consumption. So college becomes a place to go to spend four fun years, rather than increase knowledge and skills per se. Such a view would make the subsidy (financial aid) argument much weaker.

The sciences are for nerds and losers.

the nerds and losers who are not fending off rapists and bums at Zuccotti Park =P

"Free love...well, free to me." That's the difference between price and cost.

College athlete, Masters Degree, Engineer. Giant middle finger right back at you from this nerd and loser.

... Unless you were kidding, and then that giant middle finger is directed at all of the useless degree holders that want me to pay for their college debt.

I get the joke, but if you were a college athlete I really doubt you were a loser.

Me, I was in marching band. ;)

I was a swimmer. There is an element of weirdness in that choice, and I can assure you I am quite the nerd. I used a a statistical analysis of my fantasy football league's results to try and steer the other members towards some rule changes ... all STEM people have an inner nerd, even if they try to hide it.

Not sure if your comment is meant to be taken at face value, or if you are subtly making a very good point. The problem *really is* that it is difficult for anyone to think long term - let alone adolescents - and studying STEM is a bad short-term strategy for any 18 year old male or female because, well ... "The sciences are for nerds and losers."

As a heterosexual female with an electrical engineering degree, I don't remember finding studying STEM being a bad short-term strategy at 18 years old. I had already sworn off dating arts students at the tender age of 16, I've never had cause to make a similar oath about nerds (nor break the one about arts students).

I don't remember it being a bad short term strategy either. I was always proud of being a nerd haha! I am, as you are, a heterosexual female with an electrical engineering degree, and though I don't quite agree with swearing off dating arts students, I do agree that I don't prefer to date them because their mentalities are so different to an engineer's and it would just never work out.

Indeed, nor do I get it - at my highschool, being the guy who actually understood all the math and physics made me fairly popular...

You haven't sworn off dating art students? Hmmm. Let me just say ... soooo, wassup? ;-)

Obviously it's not a bad strategy for females. They are pretty much guaranteed a mate in such male-dominated fields.

The odds are good, but the goods are odd.

"What we need is a solid way of signalling domain competence without paying for a degree."

From my very short time as an analyst at a VC, I noticed that our principals always passed up good opportunities because of lack of educational qualifications. Nobody got in the door without an MBA. These companies would get turned down, and now they are popping up with eye-dropping acquisition prices...funded by smaller and lesser known VCs. Y Combinator has been doing the same thing for 6 years now with fabulous success.

I think there are plenty of people out there that can detect domain competence without degrees. Personally, I think the difference is that they are small, and run by people who have never believed in having an HR department.

I've reached the conclusion that HR is evil. It's the single largest source of inefficiency in modern corporates.

I totally agree. The HR and Recruiting professions are probably directly to blame for a large fraction of the unemployment.

Formalizing the talent acquisition function is inevitable as employee growth increases (which almost always coincides with an increase in the size, scope, and revenue of a firm).

Once formalized, there is no practical alternative to staffing this function with people trained in personnel management. Thus is born the Human Resources department.

Only small firms can avoid the Catbert role. And how many firms deliberately decide to stay small? Precious few.

HR is a place where the otherwise incompetent and non-competitive end up. No wonder they don't execute their jobs competently.

I'm a person who studied music in school, but I now work as a software developer. It's something I've always been into, and I used to participate in open source back in school and that led me into a career writing software once my aspirations of getting into a good symphony fell flat. So from my own experience, I think open source can be a way of signaling domain competence without having a degree. At least for the software domain. :) Perhaps people who work in other fields have some ideas that are applicable there.

I started my own company partly as a reaction to not being able to get meetings with VCs for job opportunities. They kept asking if I had an MBA. Once I had a business plan they stopped caring.

In software, having a degree is a very poor signal for domain competence, hence the never-ending hand-wringing over the best way to interview software developers, and the relative willingness of many people (not all) in a position to recruit to overlook the requirement of a degree.

The safest bet on establishing domain competence seems to be (and always has been) evaluation by a group of existing domain experts. Good software developers can spot good software developers. In software, degrees simply act as a prerequisite hurdle, or filter step. No one thinks they signal expertise on their own.

A degree should change how you interview a subject. You'd ask all candidates a basic filter question, like "write a routine to invert a words in a paragraph." But the people with degrees should also have a deeper understanding of the theory, like complexity in time and space.

Even self-trained professionals that have disciplined themselves to cover breadth and depth in their self-training will possess the basic theoretic knowledge of computational constraints (and similar aspects of the discipline.)

A self-starter who absorbs all of Knuth's TAOCP will have a very strong foundation in the discipline--stronger than most graduates of degree programs.

What makes the learning of computer science different from other disciplines in its amenabilty to self-training is the widespread availability of both the personal computer and the needed languages and development environments in which to write and test software.

Were everyone able to possess their own Tokamak, self-trained nuclear physicists would be a more common sight.

HR departments are annoying, but unfortunately a necessity in a large company. On the other hand a well run small organization should avoid them for as long as possible. At least for setting guidelines on hiring, promotions and retention programs.

"Personally, I think the difference is that they are small, and run by people who have never believed in having an HR department."

HR is a useless load of crap.

Engineering should be made considerably easier (and it is in some ways). I pursued (and struggled in) a major that few could complete, got a job and when I left I was replaced by someone with a much easier degree. It was crazy. Here is my favorite anecdote. We had a class of 40 that did not receive a single A. When we complained to the professor his response was "well, the other section had one." The other section was, of course, taught by a different professor.

How can you say it should be made easier?

I bitched and complained during the whole process, but getting pummeled by challenging professors and impossible homework assignments was like a "rite of passage." Your story is typical, I've never heard engineering graduates talk about how easy their classes were.

Graduate school classes were actually easier for me. Once I had been working and was a little more mature than I was at 20, it was a lot easier even though the subject matter was more challenging.

Maybe the argument is that a little more application based knowledge should be included in the undergraduate curriculum. Most fresh out of college employees are useless engineers, no one solves real world problems by hand. There is a good bit of learning on the job, but what an B.S in engineering gives you is the foundation for being a problem solver, at least that's what I've found after about 10 years of actually working.

Realistically, what should be done is there should be a natural progression from 'general engineering' into a more specific engineering field. This can be done via the course progression/curriculum, but right now, most school curricula are designed to force the 17 year old high school student to select a specific major. What high school senior *knows* what Survey Engineering Technology is? The school should start you off in the generic engineering classes, and then promote you within based on your strengths.

I can say it should be easier because it is my belief that the orientation of most engineering majors is to weed-out rather than weed-in. Now, I think we need to weed in because we aren't going to stop having children and we would rather they have an objective education. They can do the other stuff as a hobby.

I agree with the general engineering. Too much 'siloing' and turf wars in engineering. Most of this stuff is all the same. The divisions are just for supplier job security. Our department head gets irritated if we take classes outside of our department because they don't get the reimbursement. Of course, they get mad at us rather than the suppliers. Our committee members often require us to take courses outside of our department (particularly if they reside outside of our department). Even arseholes respond to incentives. Within a very difficult problem, working on this is low-hanging fruit. I'm almost to the point where I think we should just have two majors, liberal arts and techncal. Everyone should get both degrees.

@ Andrew' - I went to a small liberal arts college that only offers B.A.s. As a result I know quite a few people with B.A.s in biology, chemistry, or computer science. Most of them were double majors in music, studio art, or a social science. For the most part, they've been more professionally successful because of the soft skills. One (a programmer) got opportunities and promotions because he was the only non-manager on his team who could communicate successfully.

the 1st two years are primarily general engineering courses ... all Engineers need to take chemistry, statics, thermo, materials, programming, calculus 1-4 and physics 1-2. That's a pretty brutal 1st two years before any real focus on a particular major. Those are the "weed out" classes. They suck. They weed out some due to subject matter and others due to workload. In general, a STEM major requires a lot of discipline. My complete lack of it as a 18 year old almost got me ... ugh I cringe at the thought of having a history major and living in my parents basement.

Realistically an undergraduate engineering degree needs to be a 5 year degree.

In terms of credit hours as a "full time" student, the engineering program I did would have been 5 years. That is, to complete it 4 required taking more credits per term than the average student.

I was told at one point that engineering used to often be a 5 year degree. The idea being that engineering was (quite some time back) one of the few undergraduate degrees that a person would get and immediately be ready for the job field. Other undergraduate degrees (I think the example given was physics) were merely stepping stones to graduate degrees required in the field for research or academic work. I think this was based on a time in which the only people to go to college would generally have been people interested in becoming academics or researchers. Since I wasn't alive at the time and haven't researched well enough, my story is at best anecdotal.

There are a lot of just-so stories in academia. I'd believe that they just kept ratcheting it up until they realized they had overdone it.

Three years after I started my degree at UCSD, they took something like 20 credit units (5 courses) off of the Computer Engineering requirement. So they've definitely been ratcheting back.

It's not really just that it's a heavy load - those first year engineering courses are REALLY boring and tedious, even if you compare say Engineering Calculus to Math Major Calculus or Algebra, the Engineering version is Calculus with anything potentially interesting ( the more abstract stuff) sucked right out of it.

In my university (1984), engineers took 36 courses to graduate, compared with 30 for students in the general college (which included pretty much everything else except Architecture), although our thesis was a "class" and for the others it was not, so a more accurate comparison would be 36 vs 32, or an extra half year squeezed in. Essentially most students took 4 courses a semester, with 3 their senior year while working on their thesis, while engineers took 5 a semester for their soph/junior years and 4 the rest of the time. The other thing was that this pretty much assumed you had enough AP credits to pass out of at least something - for ChemE the total course requirements added up to 37 courses if you actually had to take things like basic calculus, so you'd have been scrambling like made trying to catch up from day 1.

My father was a ChemE educated in the late 40s, and his program was a 4 year BA followed by a BS in engineering one year later.

It is the humanities and social science students at elite colleges that are in some ways hurt the most by the slowdown in finance and consulting jobs.

"hurt the most".

Are you a comedian?

can we preorder book?

"There’s too much information in the blogosphere and the Khan Academy to explicitly require college for anything." You need "college" for the lab work. Which is rather the point of the post.

Hi Alex seems a little anecdotal to me with the whole college students going to OWS. Do you have specific demographics there or speculating? Cause if you have the evidence on that I would love to see it.

Here is my (not 100% PC) theory:

The rise of the number of Arts etc. students is due to the increasing number of female university students.

Men will generally chose a subject with career prospects in mind.
A woman is more likely to chose a subject based on what is fun rather than on what pays well.

That's because a man with bad earning prospects makes bad marriage material.
For women, that's not so clear cut.

Yes, the data backs this up. Though women are attending college more its all fluff majors. This makes sense, women don't see themselves as providers. They do things because they want to. Men, especially the beta men that go into STEM, realize that their best hope to acquire a mate is to fill a provider role by entering a lucrative profession.

So any smart man is automatically a beta? That stupid worldview (alpha/beta dichotomy at work EVERYWHERE) is so lame and, well, beta.

People who do boring stuff, like STEM jobs for megacorps, do so because they have nothing to offer besides the income derived from their boring jobs.

Does not compute. You are categorizing all STEM jobs as boring?

As a person with a STEM job, I have to admit that this is actually the cold truth. I never was an alpha male and none of my colleagues were alpha. The only people I can think of as Alpha in the 6 labs I've worked in and the dozen or so I attended in university were Beta males who pretended to be Alpha (but everyone knew they weren't). For example, I remember how quickly some men I knew who acted Alpha among the rest of us Betas would turn into a Beta himself as soon as he left the lab and had to deal with society. I worked with one of them for 3 years and it was quite the transformation. He would go from a 'outspoken cool guy' (in his mind he thought he was cool) to a more modest (dare I say slightly cowardly) man outside of the lab. A mutual friend of ours once brought this up to him and he blamed the rest of society for being too stupid and simple to socialize with him. I found it particularly interesting that he chose to think of the rest of society against HIM rather than US when we were clearly a lab team who weren't "stupid" and understood what each other was doing. The story of Dr.Jekyll and Mr.Hyde is a fictional work but it draws its root from a very real area. We the Beta males long for certain Alpha needs that we can't meet. We go into STEM to make up for it tangentially.

Hmm, I'm not a man, but the impression I got from my fellow students in doing electrical engineering, and from professionals in that field, is that they were doing it because they found it interesting in and of itself. I think the field is just too demanding to be in without some intrinsic enjoyment of it. (Not that enjoyment is enough either, a lot of hard work is required).

If you're a good-looking type-A male, you can probably make a lot more money with a liberal arts degree than with a STEM degree (by getting a job in law, finance, or management). For all other men, the opposite is true.

The girls also induce more males to take up the sort of fluffy, sexy majors that impress the women.......

But then dump them for investment bankers when its time to find a meal ticket for a family.

This comment doesn't make much sense - investment bankers come from fluffy, sexy majors, rather than the STEM majors that this article is saying we need more of.

Investment bankers come from fluff majors only from top schools (Harvard, Yale, Princeton, maybe Stanford). Fluff majors at these schools are different from fluff majors every other place in the country. Mainly because the purpose of those schools isn't to confer knowledge, but to A) provide a screening mechanism for companies and B) teach certain social skills and values.

Investment bankers have communications and sociology degrees? Not sure about that. Lots of physics PhDs, etc. work for hedge funds and investment banks.

I was a quant. Physics PHD quants are a minority on WS. Most bulk investment banking work gets done by your average run of the mill top 3 grad.

The guys with Math/CS/Physics PhDs are Quants - they're different from Investment Bankers (who are basically high end sales people). The Investment Banking types tend to be hired based on personality and having the "right kind of background" (lots of extra-curriculars, graduated from the right schools, hangs out with the right kind of people, etc.).

Data? Because when I went to college, back in the 80s, no one --not parents, not high school guidance counselor, not college academic advisor--ever once said anything about "This is how much the average English major makes; this is how much the average chemical engineer makes", or anything else that might have been useful.
I once had a job application that asked "what's the most useful thing you learned in college?", and my answer was "give the Dean's secretary flowers on her birthday."

My recollection during college in the 80s is that it was generally agreed that engineers make more money out of school but plateau quickly. Of course, we didn't realize that solid technical experience in a business domain is a great stepping stone to management.

It's unfortunate that no adult responsible for teaching you about more than the academic import of a degree took the time to relate this basic "facts of life" type of knowledge to you as you were seeking guidance on higher education choices and their impact on a life's work.

Though I won't generalize from my experience (which is of the same vintage as yours), I can say that my high school's guidance counselor did offer each of us information on the standard salaries of positions associated with common undergraduate degrees.

Furthermore, information about the real-world aspects of earning power was readily available to me for the asking when I consulted adult relatives, friends, and neighbors. It was my neighbor's cardiology practice (and his advice concerning it) that provided me with the additional guidance to choose that field for my own.

The consistent annualized mid six-figure income that he related has been borne out in my experience, readily offsetting the one-time mid six-figure cost to acquire the education, training, degree, and certification.

I agree with Cliff's criticism and don't see evidence for davver and Rahul's theories.

My anecdotal experience in STEM and non-STEM degree programs is that the difference in the men is small, and if anything the engineers were slightly more masculine, not less. And, of course, the variance is high.

What I do see is a very common bash-the-successful meme.

You ought to look up the data then.

I'd appreciate a pointer to public data. My experience is in electrical engineering, finance, and consulting.

In consulting I have some proprietary survey data to support my point. The skew there is more dramatic, but as you would imagine, it's a field that attracts less dominant people. My finance experience is mostly at hedge funds and trading desks, which was much more comparable to engineering environments. But with ties. My finance view is purely anecdotal.

Finance is not STEM.

> Finance is not STEM.

That's part of my point. Engineering and finance are pretty similar.

Like how football and bocce ball are similair? Yeah, I remember the head of the bocce ball team banging the head cheerleader.

I hope this was obvious, but I'm not saying that the relative number of men is not much different in different fields.

Obviously engineers are mostly men, and sociologists are mostly women. What I'm saying is there's not much difference in the alphaness, however you want to define that, of engineer men versus men in other fields.

Well, there is no survey of "alphaness", but my own experience, the experience of most people I talk to, and data that can serve as a proxy for alpha all point to the fact that the commonly viewed "alpha" fields have more alphas then the "beta" fields.

You implied you had data. I appreciate the correction.

I think your impression is incorrect, for what that's worth.

What is an "alpha field" and what data serves as a proxy for "alpha"?

> What is an “alpha field” and what data serves as a proxy for “alpha”?

I dunno. It's a fuzzy term. I suspect davver might be confusing "money" with "alpha."

The consulting skew is towards the submissive, if that wasn't clear. Dominant personalities are rare, and to some extent selected against. I've seen data. Personally, I had to really dial back my aggressiveness to fit in, and I know that's also been true for other engineers and hard-core finance types making the transition.

Alpha traits:
Extroversion scores on psych profiles
Charisma and looks (6'2" 220lb is a good benchmark)
Earnings (especially apex earnings in the top 1% and above)
Ability to score high quality women (8s and up, bonus points if they are educated, come from good families, and behave well)

Some examples of alpha fields would be professional sports (stars), investment banking, fashion model, high status artist/performer, BIGLAW.

A beta field is anything where you are a Dilbert in a cube or a shitty office. Making less then <200k (more in NYC/DC/LA/SF) a year is a good sign, but I would give a break to someone in a fun high status job that only made 100k. If at any point in your interactions with colleagues someone says, "make this work," and you have no choice because if you say no your fired and can't pay the mortgage your in a beta job. If your an alpha people won't give you that attitude, and if they do you can quit because your rich already.

Do most of the engineers you know fall into the beta or alpha category? I'd say beta. However, most investment bankers I know fall into the alpha category. There are outliers to be sure but outliers don't matter.

Wait, what? That is a really surprising definition to me. "Alpha" means you have an income of over $200,000? It sounds like you are just picking out all possible good qualities and saying someone who is in the top .001% of all people is an alpha. Well okay, if you define it that way then yeah, most people who are really superstars are going to be in entertainment or sports or investment banking or something like that, but does that have any applicability to peoples' lives? By that definition the number of alphas is super-tiny. I mean narrow the population to making $200,000/yr, then narrow it further to 6'2" 220lbs, then narrow it to people banging 8/10s on a regular basis, I mean that's nobody but professional athletes.

Also not sure that BigLaw is an alpha field, most people in BigLaw slave away at a desk and have no life. I could be working at a big law firm and who knows, might eventually end up there for a little while, but I avoided it because it sucks.

I'm completely with Cliff on this. That's just a list of positive traits.

Biglaw is all about supplication. Ibanks and hedge funds all say "make this work" and fire if you don't. Heck, pro football teams do that.

I think your definition is unusual and incoherent.

Lots of STEM majors can change into lawyers without much trouble. I've worked with a fair number of lawyers and found the discussions tend to be similar, since the laywers looks at things analytically and sometimes care a lot about the precise meaning of words.

If I may offer a vaguely scientific definition of alpha in DISC assessment terms, it's someone who is high dominance and high influence.

Things like looks, physical size, and money may help with women, but they aren't alpha per se. At least, that's how I normally interpret the word when other people use it.

Yeah, BIGLAW is on the outer reaches of alpha, because it requires too many years to get rich enough to have fuck you money (whereas IBers can do that within a decade). But lawyers tend to be charismatic and have other alpha traits so I give them a pass.

IBers are rich, so they don't have to take shit for money. Some do, because they want even more money, but they don't have to in order to survive. If they quit they have 7 figures in the bank and never have to work again if they don't want to. If some Dilbert quites he's got 5 figures in the bank and needs to get another job just like the one he quit ASAP or he'll lose the house.

If you want a good (though fictional) example of Alpha try Donald Draper. He has a job that is fun (advertising is a soft, interesting, people person job where he gets to be creative. DD isn't doing calculus alone in an office or debugging a program). He has people below him to handle shit work. Nobody at work gives him shit, and if they did he could easily leave. He's rich. He is(was) married to an alpha female and scores high quality females on the regular. Could you imagine Donald Draper deciding to be some nobody Dilbert in a cube debugging code?

I'm an economics professor and meet most of Davver's criteria though I only earn $150k. 6' 3" (a bit more than 220lbs). My wife is a PhD and 10 years younger than me, pretty good looking (Asian). I have my own office with a nice view of the lake and mountains and can pretty much do what I want subject to general performance criteria. Even if I don't perform well it would take years to fire me. This is the top university in this English speaking country (I have a US PhD and taught there). I'm also extroverted. I'm in the top few % of economists in the country on RePEc. And I enjoy my status in the profession, journalists calling me for comment, all that kind of thing. But I don't think I'm an alpha male or that any of this has that much to do with it. Or maybe I am an alpha and never knew it.

"The rise of the number of Arts etc. students is due to the increasing number of female university students."

Nah, the increase in fluffy subjects (like "arts") is that vocational degree programs (JD & MD) require high GPAs, and the only way to play the game is to do easy subjects. You can't be a patent attorney unless you've got a degree in science or engineering (the patent office won't let you). But you *can* be an attorney with a degree in underwater basket weaving. Not everyone who wants to be a lawyer can get into law school, so there are a lot of folks who go for the easy GPA and fail to get into law school.

I agree with Alex Tabarrok that college is oversold. He gives good examples of the types of jobs humanities graduates end up in. I think also as a rule of thumb that if a degree course has the word ‘Studies’ in the title it is often suspect and not worthy of subsidy. But don’t come down like a ton of bricks on Sociology. The fact that sociology self destructed leftwards into a feeble touch and feel micro does not take away the fact that it is a foundational discipline equivalent to economic and political science. My personal favourite route is through Weber, to Schumpeter, to Parsons, to Luhmann. From that perspective Hayek, Keynes, Mises, or Popper look like side issues or outgrowths. Sociology is worth a private or public ‘subsidy’ i.e. emphasis if for no other reason that it potentially explains the social system in which economics and politics are subsystems. For example, there’s a lot of idle economist opportunistic (viz. Greece) chatter about Direct Democracy this week. Only sociology brings together the intersecting aspects of law, politics, administration, and economics which allow understanding of the systemic origins and impacts of different representational procedures. It is laughable that anyone would seek to explain The Economic Crisis without the help of foundational sociology. This is stuff students need.

I like the rule-of-thumb about the word "Studies". For fun, I investigated the list of Departments at Wisconsin and we have 36 programs / departments with the word "Studies" in them. Do we really need a department for "African Studies" and then another for "Afro-American Studies"?? Sigh. Finally one for "American Indian Studies".

Let's keep going. Asia is not to be left far behind, of course and hence we also have "Asian American Studies". And then there are "Centres" for South, SouthEast, and "East" Asian studies.

More stupidity:

Celtic Studies Program
Chicana/o and Latina/o Studies
Development Studies
European Studies
Gender and Women's Studies
Global Studies
Hebrew and Semitic Studies

I give up. And we've not even covered half the alphabetical stupidity yet. The list goes on.

I'm not saying these programs are bad; but the trend to carve out these micro-fiefdoms just for hogging power, professorships, grants, courses and students is annoying. After all it's rare to see Physics departments create "Optical Studies" and "Gravity Studies" and so forth so why not allow the Departments of History or Sociology to encompass some of these micro-departments and programs?

Very often, these are about cross-disciplinary studies, accounting for history AND sociology AND literature departments AND geography AND economics etc. etc.etc. So a better comparison would be with, say, Biophysics.

Usually 'studies' departments/majors are actually interdisciplinary programs, and their faculty have home departments in traditional departments. The growth of such subjects in part relates to a desire for interdisciplinary inquiry, and usually they cost the College very little to maintain. Real departments are more expensive.

Gee, it's got crowded here since this morning.

Rahul: Totally agree. Now that I’m no longer in academe I can laugh about it. A pet hate used to be Chicana/o & Latina/o. The content is broadly insane,
sometimes bordering criminal. A lot of the Studies, e.g. Queer Studies or Gender Studies, would be just fine if only their size stayed proportional to their social relevance. But they develop a traction, self-importance, and appetite for students and research funds out of all proportion to their usefulness.

Of course their is the logic of multidisciplinarity. I see no need to avoid the availability of some highgrade Southeast Asian Studies or Latin American Studies or Development Studies (the Studies I taught). Still, the fact remains that the word ‘Studies’ is a reliable flashing light that should warn students of the probability of trash content, bad job prospects, and social inutility.

More often than not the reasons for trash content in humanities and social science and for the civilizational self-destruct that has its epicenter in the university have a great deal to do with motivated hostility of intellectuals towards modern society and capitalism. Cannot fault the pioneering analysis of Weber and Schumpeter on that self-censoring topic of the motives of intellectuals.

By the way you left out the icing on the cake or all-encompassing category Culture Studies where to some extent it all began.

Won't be able to read the 300+ comments in on this post but I wonder whether anyone brought up another contra-innovation pattern in academe/universities apart from grade inflation and the incentives/rewards for useful degree choice.

I speak of course of TENURE and other structural monopolistic tendencies which go far far beyond the Schumpeterian conditions of acceptable temporary monopolistic incentives to innovation, and which have a deleterious effect not only on the quality but also the (ideological and/or conformist) nature and delivery of education/research content.

I heard author of following book speak on radio and agreed with everything she said:

I just can’t exit this blog without mentioning that each of my favourite sociologists Weber, Schumpeter, Parsons wrote stunning defenses of science and technology as vocations.

Podcast & transcript of Naomi Riley balanced rant against tenure etc here:

"We add subsidies to the mix, however, because we believe that education has positive spillover benefits that flow to society. One of the biggest of these benefits is the increase in innovation that highly educated workers theoretically bring to the economy."

"As a result, an argument can be made for subsidizing students in fields with potentially large spillovers, such as microbiology, chemical engineering, nuclear physics and computer science. There is little justification for subsidizing sociology, dance and English majors."


I'm concerned that subsidies to STEM will just result in higher tuition for STEM fields with no change in results.

Somewhat easy: STEM majors should receive some sort of preference for Pell grants and the like. Yes there may be eventual STEM degree inflation but not any time soon. But don't forget accounting or other "hard" business majors.

There is already an additional federal grant for being in STEM. I received the federal SMART grant every semester I was an undergraduate.

A list of the spillover effects to the general economy from increased numbers of hard business majors is difficult for me to derive at present.

In your judgment, what value would a subsidized increase in the number of accounting or finance majors bring to the whole of society? What advances in widely-used fields of knowledge (not narrow advances to the capabilities of individual firms) would result from having more of these professionals?

Alex, correct grammar would be "more important", not "more importantly".

Of course, if you had gone to journalism school, you might have known that.

And we would all be vastly richer for it. How much do you get paid to go around correcting grammar on blog posts?

Cheers to Master of None for wittiest response to Alex.
Jeers to Corey for humorlessness.

Hey man if wasn't priced out of the education market and forced to take a minimum wage job maybe I'd have a better attitude about the situation.

Oddly enough, the Chinese people I work with are not concerned with my grammar.


Screw it bad enough and they might actually think you as one of them!

It's funny when you start using words you know are wrong because you know it is what everyone understands.

Alex was right to say "Most importantly" because that is the natural flow of the spoken English language and is more comfortably understood by most readers. It is also okay to say "Most important" because that is the Latinized construct that grammar teachers love. Grammar is a good thing, but taken too far it leads to negative marginal returns. Didn't they teach you that in journalism school?

This is an excellent post.

We could increase the number of science and math majors in college if we were able to
find a way to encourage children to take an interest in math and science and to help them maintain that interest through high school.

This year I helped organize a science fair at our local elementary school. We had a fairly high participation rate - about 35% of the students. It seemed pretty clear that young kids are interested in science. This makes sense because science is basically an organized way of asking and answering questions about the world through experiment - something kids do naturally in an unorganized way.

Unfortunately this interest seems to dissipate as the kids get older until we end up with the relatively low numbers of math, science and engineering students that Alex noted.

Science fairs are a good way to turn kids off of science, though it does teach them the reality of science. First, lots of students will do an experiment that has been done before which defeats the purpose of the experiment. The graders generally don't reward orginality in any way, so kids know they can still win by doing that. Any kid who thinks of an orginal problem will be undermined by unglamorous results which is normal for science, but lousy for science fairs. Secondly, parent involvement is poison and the norm. Those stupid google and intel contests are all won by kids who have parents who are professors/professional scientists. Very discouraging for kids once they figure that out.

I'm a scientist and I would say most of my American born collegues who continue to work in science were inspired by Star Trek and science fiction novels. That sounds stupid, but its true.

In one of my science projects my father and I made a wind tunnel using cigarettes to create the smoke streams.

I didn't win.

One of things we decided to address the concerns you raise (which we shared) was to make the fair non-competitive - no judging, no prizes. Instead we choose to review each of the experiments and talk with the kids in detail about their question and their experiment and results and so forth - hopefully a constructive and positive experience for the kids, regardless of how sophisticated their experiment. Our goal is to make the fair a good experience for all the kids so hopefully they learn a bit about science and even more importantly so they come away with a positive feeling toward science so they're not turned off by it in later years.

It's also true that you're not expecting cutting edge research to come out of a grade school experiment. That said though, this, and the sophistication of the equipment and experimental set up is really the only thing that does differentiate a kid's experiment from a working scientist's experiment - the latter is working on unsolved problems. (Indeed a big part of graduate training is learning what is solved and what is not in your field and getting a sense for what might be solvable.) The basic intellectual framework is the same - you ask a question, make a guess, design an experiment to test the guess, collect data, analyze results, draw conclusions, ask new questions.

The correct way to approach science fairs is demonstrated by Mythbusters.

jdm is on to a much more important question, in some ways, than the original post. How is it that our high school graduates are so often unprepared and incapable of doing a good job in STEM courses (and, by the way and of equal importance, in any course requiring competent writing)?

The recent evidence that school mathematics performance is improving is good. But science in primary and secondary schools is a big hole: and that hole is there, in part, because teaching beginning science is labor-intensive as well as (relatively) equipment-intensive, and has thus gotten cut back by our decreased classroom spending across the board. Public education faces many problems, in truth: underfunding, over-bureaucratization, weak teacher training, legislative and cultural interference (teach Creationism!), and so forth. (For those obsessed with the Right's favorite bugaboo, I'd say that teachers' unions are a fairly minor contributor whose negative effects do tend to grow as the other factors get worse: when teachers unite to fight for a slice of a [relatively] shrinking pie, then more damage is caused than when they unite to sustain themselves in a flourishing system).

If our public schools were doing a better job, then the college premise of strong AND broad education -- with all college graduates numerically, scientifically, historically and aesthetically competent, together with deeper skills in one of those areas -- would be more realistic. Not every college grad needs to be a engineer, but they should all understand basic mathematical analysis and statistics. Not every college grad should be an English major, but all should know something about how to use their language, and how to interpret the way others use it.

The economy has to be non-bullshit, the college has to be non-bullshit, and the student has to be non-bullshit (and the public schools need to be non-bullshit). There are many weak links here.

There is a threshold in STEM, an objective margin. A bridge that everyone likes but falls down doesn't cut it.

Just promoting STEM at the college level isn't going to be terribly useful as high school students applying for college probably already have an idea of their intellectual strengths, and for most of them, their strengths are... 'anything but math'.

I had the good (bad?) fortune of having a frustrated physicist for a dad who didn't understand that preschoolers weren't capable of handling multiplication yet and a mom who never thought math or science was for boys only, so I was never ever afraid of math. Didn't end up in a STEM major, but picked up a math minor junior year in college on a whim. So many of my frankly intelligent classmates, on the other hand, didn't have anyone to help them build that foundation in math; they were convinced math was something only some people were 'born good at' (and if it was a girl, chances were that they figured 'boys were better at math' and it wasn't for them, pah!) and past 6th grade, it was pretty easy for kids to avoid harder math classes. If you end up in junior year with maybe only a little algebra and precalc, there's not much hope for you to take anything math-intensive in college.

Honestly, what's the data on Head Start? I've always wondered if that might be a better use of resources.

Pretty abysmal, unfortunately. I don't have a link to the latest results but I read them and basically they said no long-term effect.

I really disliked science fairs as a kid. Science classes emphasized rote learning and memorizing mnemonics. I hate to say it (because it's contradictory to what science is) but emphasizing the exactitude of the scientific method is an easy way to get young kids out of science. I did the stupid "does the color of light affect plant growth?!" project so many times because you could just fabricate results and the cheating was pretty much impossible to disprove.

Despite this I love reading wiki articles on quantum physics and abstract algebra...weird.

At least for elementary and maybe middle school students, perhaps it's more important to enrich their curiosity, rather than drown them in boring methodology.

This is because high school teachers are usually people who couldn't cut it in STEM.

Lack of teachers and esp. lack of GOOD teachers teaching these subjects contributes to the low participation rates among older kids.

"College has been oversold."

This premise does not follow from the argument. Clearly, there are fields underfilled by American students. Is there any actual evidence that arts/humanities/social science degrees are actually "oversold", which implies encouragement from outside, or do students simply self-select and then reap the consequences?

I don't know, but a college should do the work of matching between the labor and the labor market. I think you are probably right in that they don't do their job at all.

Why should colleges "do the work of matching between the labor and the [jobs?] market"?

I prefer adults make their own choices.

college students count as adults?

Indeed they do. The vast majority are old enough to enter into contracts, get married, and serve their country.

To infantilize them is to become part of the problem.

Some people, and I know this is a tiny minority, go to school for edification and not as a one step in a career path.

Think of the poor kids of the 1%!

Lying on beach chairs on private islands can only fill so much time.

Nothing can technically be "oversold" or "undersold", as long as their are willing customers. Colleges and universities sell educations not jobs.

According to a rational and efficient market you mean, right? Are students who go into a field because they need to be taught about it the best model of that? Isn't that more like a model of information asymmetry?

No they don't. They sell accreditation not education. The education one must provide one'self.

Premises don't follow from arguments; conclusions follow from premises and that is an argument. /pedantry.

I disagree with your premise about fields being underfilled by American students. When businesses complain about "shortages" in some field, they are complaining about a shortage of people at a wage they are willing to pay. If the businesses were willing to ante up, they'd get all the employees they can hire.

As for the self-selection vs encouragement issue, I'm reminded of a study done by SWE (society of women engineers) back in the 90s trying to get a grip on why there were so few women in engineering and why they dropped out so much more than male students. Of women entering engineering from high school, around 90% had a family member or close family friend who was an engineer. For males, the number was around 60%. The implication was that smart male students were encouraged by guidance councillors to enter STEM fields, while females were not so encouraged.

>do students simply self-select and then reap the consequences?

That happens a lot. Popular fields end up with bursting classrooms. Fields with large publicized layoffs end up with huge drops in enrollment. Large scale layoffs in engineering in the 90s and 2000s caused many students (the sort of students who could do the math) that they were not likely to stay an engineer long enough to pay off their student loans. Large layoffs in the IT industry after the dotBomb resulted in large declines in computer science/engineering enrollments.

Making it easy to hire & fire folks is the root cause of the boom and bust cycles in what students select. Too many companies jump on bandwagons because some bizjournal made it fashionable to do so, and it takes years for the next crop of graduates to recover from defective business fads.

I don't understand this discussion at all. Alex just discounts prima facie the need for societies to cultivated educated and engaged citizens. If anything this need seems to grow more acute by the day as the processual pieces of our nomativity succumb to logics of incetivization or just stupidity. Isn't there a need to innovate and grow the pie and then at the same time, to disburse that growth in the interests of the values that make us human? I think so. The real question is, why do we need to pay 50,000/yr to do either? Not whether reading Nietszche is somehow justifiable by the same logic as building iPhone apps.

Does one need to go to college to be educated and engaged? Especially in the internet age?

Yes, absolutely. You cannot master a complex discipline without other real people to interact with. I mean, maybe you can, but I can't and I don't think its uncommon. Where do you get the subtle changes in tone that indicate that X question really matters? Or the dismissive tone that shows that Y may be overrated even if you have to quote him. I am not saying that writing cannot communicate X or Y, but rather that you need about 10,000 of those kinds of interactions with many many people to fully orient yourself in topics as rich and diverse as the ones I am familiar with.
Besides which, having friends who are also doing the same books, making fun of the same guys, and getting a beer afterwards is a big motivator to master various fields. You don't just need statements like "heidegger ontlogizes Kant's first critique" but you need someone to give crack jokes about it too.

> Where do you get the subtle changes in tone that indicate that X question really matters?

In case of any doubts, go back to the ancient Greeks. You mentioned Nietzsche, well, let's say someone has doubts about reading his works. What should she do? Go back to the Greeks, of course (and by this doing she'll probably be more "nietzsche-ean" than Nietzsche himself) . And I could go on and on. See? Problem solved, no need to pay hundreds of thousands of dollars for a worthless piece of paper.

This is a pretty asinine comment. Modern Philosophy is not reducible to Greek philosophy any more than modern politics is reducible to Athenian democracy. By engaging in a genetic fallacy, you just prove that you have not mastered the incredible advances in logic available to modern philosophers. Do you understand that the Greeks do not have a modal logic? Do you understand that nowhere in Aristotle is human freedom, subjectivity or agency broached in the same way that its explicated in the second critique? Do you understand that Descartes questions do not get off the ground for the Greeks? How about Wittgenstein-- any inkling of his language games in Heraclitus? Nope.

> This is a pretty asinine comment.
> Do you understand that nowhere in Aristotle is human freedom, subjectivity or agency broached in the same way that its explicated in the second critique

> How about Wittgenstein– any inkling of his language games in Heraclitus?

Is Wittgenstein a philosopher? I cannot tell. I mean, I actually tried to read his writings, but I couldn't tell what problem was he trying to solve, and I have a hunch he didn't know either. Or maybe you think "language games" is actually a problem worth solving, in which case good luck with that :)

> Modern Philosophy is not reducible to Greek philosophy any more than modern politics is reducible to Athenian democracy

Me again, just remembered: Maybe it's worth mentioning that in his later years Thomas Jefferson said something like: ""I have given up newspapers in exchange for Tacitus and Thucydides, for Newton and Euclid; and I find myself much the happier" , or how Thomas Hobbes (the guy that actually invented modern political science, together with Machiavelli) translated Thucydides into English. But yeah, maybe this is just a coincidence, and "modern democracy" was given to us modern-people by the Gods from the skies.

"Is Wittgenstein a philosopher? I cannot tell. "

Best comment all night.

What you're describing is a consumption good, though, not an investment good. Alex's post analyzes education as an investment, you're analyzing it as a consumption benefit.

Naturally, if you personally value the consumption-value of higher education, and are willing to pay for it, then it is not "over-sold" to you. But if we're talking about ROI, then Alex is right.


Thanks for that reply. I would say that you can invest in the quality and depth of the moral fabric of your society and that these are worthwhile investments. A society with a range of philosophical, literary and social interests is more likely to be fulfilling to humans who live there than a comparable society bereft of those important goods. I would not want to live in a society without bridges, roads, or IT infrastructure either. But its cheap and demeaning to say that normativity plays no role in building a society with bridges, roads and IT. Notions of providentialism, governance and rationalism which make the environment hospitable to modernity are not shot out of a cannon. They are argued for by Hegel, Adam Smith, Hobbes, Rousseau and Kant. Cf. Hirschman "The Passions and the Interests" Rosen "On Voluntary Servitude" Funkenstein "Theology and the Scientific Imagination" Habermas "Public Sphere" --- oh and cf Hegel "Philosophy of Right" and Adam Smith "Theory of Moral Sentiments" (also new books like Mccloskey "Bourgeois Dignity")


Are people really improving the quality and depth of their moral character by majoring in communications and going out drinking 5 nights a week?

Chakira, you'd be right if Alex were suggesting that all-but-the-STEM-degrees are worthless. He didn't say that. He noted the fact that there are A LOT more people studying journalism than math.

I think your own comment demonstrates Alex's point perfectly: Society is better served when there are a range of skill sets available from which to draw. Alex's chart suggests that the balance is tipping too far to one side, especially considering the high financial costs and comparatively lower financial benefits of that side.

Certainly nothing cheap or demeaning in pointing that out, right?

I think you are spot on. My undergrad was largely for consumption (econ at SDSU, a ton of fun, and payed off by getting me to MR 7 years ago), but my MS's were for investment (local college with a flexible enough schedule to take care of my infant son while working on Computer Science). The higher education degree market is bifurcated and it is a disservice to lump them together.

I agree, but Wendy Brown has argued that "liberal discourse itself… continuously recolonizes political identity as political interest—a conversion that recasts politicized identity's substantive and often deconstructive cultural claims and critiques as generic claims of particularism endemic to universalist political culture."

@Nathaniel: What.Is.Read.Cannot.Be.Unread - my g-d, that is some heavy postmodern sh*t ... I wonder if it is actually distinguishable from the word salad that schizophrenic patients produce. If Wendy Brown was institutionalized, I don't believe they could tell her from a 'normal' psychotic.



Thanks for your astute reply. I didn't major in communications or go out much in College. I must confess that I am perhaps really unfamiliar with typical American college experiences. During my time in college I was lucky to have a great social life which added to my learning experiences. Many of my friends were interested in discussing intellectual problems from many different perspectives. I was also lucky to "go out" to the Met Opera and the Frick. It was a really special time where I learned a ton and was able to read and schmooze intellectually all day, Maybe thats not what most people do?

Yea, that's not what most people do. Nothing wrong with it, but realize that you're in a tiny, tiny minority and that dreams of making college produce more people for you to intellectually spar with is likely to produce disappointing results. I mean, have you seen the kinds of things the average college grad believes to be true? After I graduated, it was a true revelation to realize how unconcerned the rest of the world was with Marx and Hegel compared with serving human needs in exchange for money or producing products people want to buy.


The problem is that for people like Alex or most posters here, if the "benefit" is not quantifiable(with money being the simplest and most visible metric), then it must not exist.

Its a common pathology for people who think like "economists", or like to pretend they do.

Get. Effing. Real.

We are talking about people going 25, 50, 100K in debt for this crap. They do not have the luxury of studying classical philosophy on borrowed money. They will need to show a positive ROI.

I think I spoke to the fact that it shouldn't cost 100k+ to do this. I agree that the cost of higher education is inexcusable. I disagree with gutting the humanities to get a higher ROI.

We are not 'gutting' the humanities. There will always be enough children of wealthy families with the luxury of becoming a professional historian, lit crit, etc.

We do not need a humanities department, for example, at Cattle U. in the middle of Iowa.

Also, does your knowledge of "the processual pieces of our normativity" tell you what role guaranteed loans play in education costs?

Anti Gnostic

Thanks for the replies. I certainly appreciate your nickname and your taking the time to respond. But I think you limn a very black and white (Manichean!) picture here which doesn't accurately represent my position. I appreciate the economic constraints of giving liberal arts educations and the advantages of the STEMs. I just think that there is a good case to be made for investing in the former as well as the latter. I think that if you continue to misrepresent my position as some kind of cultural elitist out of touch libtard nostrum you will at least rob yourself of the possibility of thinking through the problem in a less constrained fashion. I would love if you could reply to my actual position and not to something you made up! Also it would be great if these reply buttons worked properly (I guess for that we call the STEMs)

Drawing down the humanities to a sustainable level is not "gutting" them. Even the ancient Greeks supported only a limited number of philosophers, and many of those actually lived frugally.

Look at Japan - it is a very cultural nation, but it keeps its humanitarian class within tight limits.

The current USA has probably 4x to 5x more humanity majors than optimum. (And 10x more lawyers). I would guess that measurable net effect of another new young humanity major onto American GDP is actually negative.

Just curious where the author would place Econ in that black and white humanities vs hard sciences-based curricula divide.

Empirical review of the evidence would suggest that Econ be put in the same category as Religious Studies.

I don't get bogged down in that nonsense. Thinking about it now, I'd make a punnet square where liberal/theoretical and practical/applied are not different ends of a spectrum but different axes on a punnet square. Maximize both.

I don't know, theology students are usually good writers.

((/rant) Empirical review of the evidence tells me that a religious studies guy do not have any idea how to even evaluate the evidence, unlike an econ guy. The latter at least knows the difference between an autoregressive–moving-average model and a standard deviation.

Econ isn't just macro and a lot of econs don't have a high opinion of macro. At least in Denmark, we don't even have a formal statistics education - we call those guys economists, because an economist taking courses in statistics and quantitative methods tend to know more about that stuff than even the mathematicians do.

I'm really tired of people lumping econ and macro. (//rant)

I've been thinking, and there's a deeper sense in which economics (or at least macroeconomics, and frankly a good deal of microeconomics) is like religious science: that is, it's about, and also often consists of a moral enterprise. Mathematical modeling is of course a powerful tool, and can be applied to all kinds of questions; but equally, the world of the qualitative and the non-linear and the recursive has powerful tools that can be applied to all questions. Human behavior with regard to things that humans value -- that is, I take it, the subject of economics -- requires the ability to assess both quantitative data and use quantitative methods -- including autroregressive-moving-average modes and standard deviations -- but also to apply, where appropriate, non-quantitative analysis, something that Religious Studies is actually quite good at. Having listened to a Religious Studies professor at my university do a seminar using a case from the Talmud, and considering the complex, recursive, and subjective assessments of value, prediction or inhibition of human behavior, and so forth, I'm pretty sure that all statistical or mathematical methods would fall far short of the quality of analysis he performed: they don't even have the vocabulary to ask the important questions.

Thaus, I must reply to your (very polite) rant: saying that econ doesn't include macro is in fact an admission of defeat. It is to say, "there is a category, a rather large and important category, of problems that my methods fail at. Therefore, we will exclude those problems from science. Problem solved!" That doesn't work. We need to address all the problems the world throws at us, not merely those amendable to autoregressive-moving-average models or standard deviation analysis.

The social sciences, I fear, have often suffered from variations on this problem: either exclude what we can't measure/quantify/calculate/include in our models from 'science', or claims that we have a proxy 'variable' that kinda-sorta might correlate with what it is we really want to measure, then quickly proceed as if it DID measure...which is rarely the case. We see this every time the question of the 'value' of the liberal arts comes up. Whatever that value is, it is not easily quantifiable, resists proxies, is often wildly and unpredictably non-linear, recursive, and chaotic, and nevertheless, clearly has a value. Just 'cause you can't measure it directly, don't mean it ain't there.

"saying that econ doesn’t include macro is in fact an admission of defeat."

I did not say that. Don't put words into my mouth if you want to debate me - I consider myself a generally polite bloke, but I can't stand that. Also just a general remark; to me a debate is not about winning or losing, it's about getting smarter, learning new stuff, getting closer to the truth - so I don't give a #¤ if it's 'an admission of defeat' or whatever.

Anyway, econ does include macro, just like string theory is part of physics. And just as some faulty string theory stuff does not have any bearing on the validity of the work of Newton, faulty macro models are irrelevant to the work of a lot of economists. If you want to criticise macro guys, get in line, but don't think for a second that many economists will consider your critique relevant to themselves and what _they_ consider 'their field'.

"Whatever that value is, it is not easily quantifiable, resists proxies, is often wildly and unpredictably non-linear, recursive, and chaotic, and nevertheless, clearly has a value. Just ’cause you can’t measure it directly, don’t mean it ain’t there."

Sure there's some unmeasurable value somewhere there. I'm one of those students of economics that also happen to read Austen, Homer, stuff like the history of the Roman empire, philosophy of science ect. in my spare time. Nowadays people seem to think that in order to educate yourself on that kind of stuff, you need to take out many years of your life and pay tens of thousands of dollars for it - whereas I think that's just stupid when you can just grab the books and read them in your spare time. (check out my blog at the link if you want to know more about 'where I'm coming from' when it comes to these things. You can just skim the list of categories to the right to get some idea.)

There's a huge leap from 'young people probably shouldn't spend tens of thousands of dollars and years of their life studying X' (because of Y,Z,...), to 'X is a waste of time'. There's some value added by studying X, even though X isn't mechanical engineering, sure. I'm quite certain this has probably been pointed out elsewhere in these comments, but if you think of education as a consumption good - then there's really 'no problem', aside perhaps from the fact that some people are running up bills they can't ever pay. It's only if you consider education an investment good that some of the ugly questions start to pop up.

Some of the questions that have been forming in my mind about all of this: How then, do we validate in the future, all those many fields that do not guarantee a paycheck? Surely we don't want to completely drop them because they do not pay, does that mean we should approach them in ways non-monetary? How might that affect status of teachers who teach the non-paying subjects? And having determined that, how then might the "paying" subjects be affected in terms of monetary value?

I think you have causality round the wrong way. A university doesn't validate fields, a field is taught by a university because a number of people already found it valid. It's not only university students who are interested in philosophy, literature, etc. Famous people who didn't attend university but who from their writings are interested in those fields and well informed about them, include Jane Austen, George Bernard Shaw, Benjamin Franklin, J.S. Mills, Frederick Douglas, Abraham Lincoln, Terry Pratchett, and Robert Burns.

People are willing to pay their own money to engage with philosophy, literature, the arts, etc, in such ways as buying books, buying DVDs, attending art galleries, buying artworks, buying works of arts and literature criticism. Terry Pratchett makes a lot of money out of his work. The status of teachers in these fields can be awarded by their students - think of how famous Socrates and Plato are.

Just curious - where do degrees in economics fit in? I have one of those, but every job I've ever landed after college has been thanks to the fact that I tacked-on a minor in mathematics at the last minute. Thank god for that. I'll probably go back for more higher education in math sometime soon.

As for the topic at hand, most people are now saying that college is over-sold. I don't know if that's true, per se, because you still can't get a good job without some kind of college education. I think what's true is that there is a higher education bubble. I'm not sure what things will look like when it pops, but the market is pretty good at adjusting for these things. Probably the market for intelligent, analytical people will never dry up. I would guess that the number of successful analytical people with journalism degrees is somewhat similar to the number of successful analytical people with engineering degrees, but that there is probably a higher concentration of such people studying the "STEM" disciplines.

The trades provide MANY great jobs, and don't really require the same kind of college education that you're speaking about.

In fact, most trades, if you're from a good school which works from industry, will pay far more than most.

True. Some of the best tradespeople I know are also very intelligent, analytical people, though. When the bubble resets after all this stimulus spending, it will be the most intelligent tradespeople who still make good money. The bad ones will all go back to school.

There are a few things going on here. Historically, law school was the way to a solid career path for humanities graduates, but lately this is collapsing. There was a bubble as too many people enrolled at 2nd and 3rd tier schools...this probably happened because graduates with these degrees had few other options.

Partly that is due to technology which is having an impact in law and in other skilled professions, especially middle managers and knowledge workers. This is something that will get worse as technology gets better and better. Entry level jobs are especially going to be hit because they are more routine and easier to automate.

Check out this tv show featuring GMU's own Robin Hanson (who I think underestimates what's likely to happen within the next 10 yrs or so):

Might it be that the students flooding into psychology et al just plain lack the abilities needed to be successful STEM students? That is, even if we were somehow able to steer more students into, say, microbiology, would we markedly increase the output of the field of microbiology?

This! A thousand times this. That is EXACTLY the point of the "soft" degrees, giving stupid people "access". No doubt there are PLENTY of smart sociology students, but the dumbest student in soc is much dumber than the dumbest guy in double e



This statement strikes me as too deterministic. While I don't believe for a second that anyone could succeed in a STEM program, you could probably increase the numbers that would by lets say 10 to 20% (just making a baseless numerical assumption per standard economic procedure) if you had effectively directed resources to those areas at an early age. If student come in unprepared to a CS or economics program such as by not having a strong math background, they will be more likely to fail regardless of if they have a 115 or 125 IQ....

I agree with the overall point though. More people are entering into college now who don't belong there and colleges have adapted by increasing the number of students in easy and unfortunately, useless degree programs.

What degrees would you have been wrong on? I'm not being snotty several backwaters in the past have turned out to be really important Virology prior to HIV, physics has gone through stages of being a backwater. Arabic on September 10 2001 can't have been seen as useful.

I think you are right about not sponsoring basket weaving too much. But how much allowance should be made for educational black swans that might turn out to be useful?

Physics? Really?

I'm a physicist and lots of academic physics is incredibly useless and a dead end. Physicists on the other hand tend to make themselves useful in other fields.

I am told that prior to quantum theory and relativity physics just seemed a backwater filling in some gaps Newton left. " I have known more people whose lives have been ruined by getting a Ph.D. in physics than by drugs" Jonathan I. Katz not related but funny

I remember my first week of college as an Engineering Physics major in the early 2000s...the ME professor teaching my Physics II class asked who was majoring in physics (there were 2 or 3 of us), told us we should all switch to ME and that all you could do with a physics degree was flip burgers and drive taxis.

Government subsidies need to be removed from education entirely, but if that causes people to fall out of their chairs, then just make student loans dischargeable in bankruptcy. I bet that would eliminate a third of the problems.

Dischargeable in bankruptcy? What are they going to do, repossess your brain?

Wouldn't making student loans be dischargeable in bankruptcy be a _new_ government subsidy?

Not really, because in bankruptcy somebody somewhere has to realize an actual loss on the loan. So that forces Sallie Mae to extend credit only to people with marketable degrees.

The humanities were historically for trust fund babies.

I like the concept, but you would have to eliminate the gov't loan-backing program as well, otherwise you're just forcing the taxpayers to foot the bill.

So, essentially, you are saying "Let's make students pay market rates for debt, and give lenders an incentive to care about the value of the degree they study." I think that is an EXCELLENT notion. You would immediately see rates differentiate between humanities and STEM.

Wow, words cannot even describe how awesome that would be.

"Oh, you want to study art history? And you think we're going to lend you $100,000 at 5% for that? Hahahahahahahaha -- get out."

This is a pretty good idea. It would definitely incentivize some degrees over others.

That's a possible outcome, but the one more likely is that schools would no longer charge $100,000 for a history degree. (Or any degree.)

It didn't cost $100,000 to get a history degree 40 years ago. Why does it today? At least with the STEMs they have the sorta excuse that labs and equipment cost money, even if I don't believe them.

Schools have had minimal incentive to compete on price and every incentive to compete on amenities, and the result is exactly what you would predict.

> That’s a possible outcome, but the one more likely is that schools would no longer charge $100,000 for a history
> degree. (Or any degree.)

That's not such a bad outcome, either...

It doesn't help reduce the opportunity cost of non-STEM degrees nearly as much as TallDave's prediction. Which is too bad. Getting a $100k degree in art history doesn't just blow $100k, it also burns four years of labor which, while presumably of lower value, isn't worthless.


Another wonderful benefit!

Though I would argue that flows from the willingness of lenders to pay for them.

If you want to make the students as well off as possible, make it illegal to lend money to them.

Schools charge as much as possible. This is why they look through your entire financial history before deciding on your "contribution." Giving students access to more money and the schools merely charge more.

(I'm probably over-stating a bit, but that felt so iconoclastic I just had to go with it.)

For EE and especially CS, the lab costs are pitiful, at least judging by our equipment. Our CS labs were full of several year old PCs that cost, what, maybe $500 per student?

I assume were that to happen, humanities degrees would fall drastically in cost since financing would dry up.

Seems fair: correlate the cost of education to future expected earnings.

Yes, this would fix most of the problem.

Did anyone notice that not only has the number of American graduates in things like computer science or engineering been flat, but the amount of actual manufacturing done in the U.S. requiring such graduates has not exactly been booming? Almost as if someone has reversed cause and effect.

As a bit of a hint even in the post itself -
'Moreover, many of today’s STEM graduates are foreign born and are taking their knowledge and skills back to their native countries.' Because that is where the jobs are, of course. Which wasn't particularly true in 1985. South Korea would be an excellent case in point, actually - looking at not only such things as steel output or automobile production, but at fairly leading edge manufacturing such as memory chips, flat screens, and phones. Or developing a high speed Internet infrastructure that makes the U.S. look as if it doesn't even have an idea of how primitive it is.

Seriously, it is as if Americans have an exceptional blind spot when it comes to looking at what they have spent the last generation doing - basically, living in and off the past.

The supply of graduates in engineering is artificially limited. They don't call them "weed out courses" (weed out to the humanities usually) for nothing. How much does supply create it's own demand? How much does a shortage of labor increase the cost causing outsourcing? I have no idea.

Appropriately difficult freshman courses don't limit the number of engineering graduates. They limit the number of engineering students who finally realize in their third year that they have to switch to something else. An engineering curriculum doesn't start out hard and then get easier.

It does sometimes, actually, probably for exactly the reason you state. It is often harder on an absolute scale, but easier relative to your expectations. It is like a logarithmic scale. I suspect it is precisely that they don't want people changing in their junior or senior year. The total filter is the same, but they front-end it into the "weed out" courses. I'm talking about making the entire filter easier. In fact, I don't even want easier, just but more graduated. Noone minors in engineering (at least I've never heard of such). They should.

Junior year courses are brutal.

I disagree. It takes about 2 years of math and science to *begin* to understand what you need to learn to be an engineer.

Surely you realize that manufacturing location has nothing to do with engineering and computer science jobs? Engineers do not work in factories.

They do. I was offered many "opportunities" to help the company off-shore and outsource by moving to the location. I also worked from here with many countries. What do you mean?

Hilarious - tell that to the German engineers who work in Germany for German manufacturing companies. Though you do have a kernel of truth - a number of German engineers working for German companies do work in those company factories in South Africa, Brazil, China, U.S., etc. But what they don't do is work for a company that doesn't manufacture. German engineers tend to be involved in an activity called 'manufacturing' - it is how they earn their pay, after all.

Sometimes I really wonder about many of the posters here - do you really think that the people running a chemical manufacturing facility (why yes, I do know a couple, actually, for both Dow and BASF) work anywhere but at that facility?

Working for a company that manufactures is 100% totally different from working AT a factory. Maybe SOME engineers work at factories (.1%).

Isn't U.S. manufacturing is at an all-time high and the highest in the world?

I decided not to reply to my own post - but sure, there is a difference between the shop floor and an office building attached to it. However, I am extremely confident that the majority of Mercedes engineers work in the same area as the factories, going through the factory gates (a factory complex, actually) to get to their offices. At least, that is my personal experience concerning Rastatt (A Klasse), Gaggenau (Unimogs/transmissions), Wörth (trucks - and the world's largest truck assembly facility), Mannheim (busses), Sindelfingen (cars), Böblingen (cars). The same also applies to both French and German chemical facilities on both sides the Rhine - again, just personal experience from people who ran the facilities, and a person who designs such facilities for BASF (he works in Ludwigshafen). As for the engineers who work at LuK - why yes, they also go through the factory gates.

German engineers, in my personal experience, work at the same place the factory is located. Maybe it is because Germans are so conservative when it comes to how they manufacture high quality products to compete on world markets.

Or maybe, it is just because living in this region of southern Germany, it is hard not to meet engineers socially or have them as neighbors, or be parents of students in the same class as my children, and so on. All of them seem to work at the factory complex site of their companies, and did the same when they work overseas.

I'll admit this sample is skewed to chemicals and automobiles - but then, both industries are fairly major global employers of engineers. As for Mittelstand companies - most of them wouldn't be known, but the Mittelstand, as a group, is much too cheap to ever pay to outsource engineer offices.

They work at lot's of other places. R & D labs and design offices immediately come to mind. Equipment vendors and contractors are others. Of course some engineers work at factories but those are not all.

If you understand what most engineers do, they design things and test them, not fabricate them.

He's probably confusing "Engineer" with Operator, Welder etc.

Hey these days the guy Dell sends to replace a power-supply calls himself a "Service Engineer"

Rahul - I'm talking about the sort of people that run a major Dow facility, for example. Trust me, he isn't an operator or a welder. Or a person responsible for designing BASF production facilities. Or someone responsible for getting a Mercedes production line running in South Africa. Obviously, they aren't the only engineers at the facility, either.

Manufacturing companies, and not just German ones, prefer to employ engineers for such work - though in the U.S., apparently, the manufacturing plant is just outsourced, while the engineers, well, what do U.S. engineers do when there isn't any facility requiring them?

What about process engineering? The hard part of creating a Dell computer is not deciding which components to use, it's creating a manufacturing process that's both efficient and produces quality product.

I don't know about everyone else but I work in a building with 300 engineers and our production facility is right across the street. We go over there so often some of the engineers are trying to put together cash for a golf cart to drive the 300 yards faster.

Cliff -
'they design things and test them'
Of course - where do you think Mercedes engineers design and test things? The A Klasse is built in Rastatt - where do you think a lot of the engineers involved in design and testing it work? Mercedes trucks are built in Wörth - where do you think a lot of the engineers involved in design and testing them work? This is just based on personal experience, of course, and not every single engineer involved in the Atego sits in Wörth (or Gaggenau). Maybe Daimler has a giant secret complex somewhere the engineers sit around in - but it is well hidden from all the engineers I know who actually work for Diamler at those facilities. Not to mention the Stuttgart based ones, for that matter.

This is a bizarre discussion - I'm simply detailing how things are in this part of southern Germany, a region which is heavily oriented to export based manufacturing, and people keep insisting that this isn't how engineers work. Well, maybe that is true - in a country which can't seem to compete in manufacturing on world markets.

As for U.S. manufacturing being the world's largest - how is that export surplus in manufactured goods looking?


Well I will update my priors as apparently more engineers work in very close proximity to factories than I was aware. However, I do think your understanding is indeed skewed by your location in Germany, which does much of its manufacturing in the country, and therefore I imagine it would be logical to have all company facilities located close by. However for most engineers (not industrial or process engineers, okay) you can do design and testing without being right next to where the goods are actually made. Which is the point, that the location of the factory does not determine the number of engineering jobs in the country.

As far as the "export surplus in manufactured goods" i.e. capital account deficit, surely you realize it is irrelevant to the question of manufacturing output? Why would you ignore production for domestic consumption? Why not look directly at the number that is actually relevant, total manufacturing output, where the U.S. dwarfs Germany?

The more time I spent in the factory, the better the engineer I was, and the lower the probability of promotion.

I did look at World Bank data ( ) and considering that the U.S. population dwarfs Germany's, manufacturing output is quite equivalent (exchange rate plays a bit into this - all measures were in dollars).

What Germans don't really do is manufacture trash - and I mean that quite literally, actually. The laws governing packaging are quite strict, and generally require the seller of a product to pay for the disposal of packaging (DSD - Duale System Deutschland). I'm not going to attempt to compress 15 years of German regulatory approaches to trash, except to note that even as the German manufacture of packaging has declined quite dramatically. Interestingly, a giant retailer like Aldi finds this approach just fine, since it also reduces Aldi's costs throughout the entire supply chain - less material cost, less storage space cost, less transport cost, less shelving cost. And Aldi pretty much ate Walmart's lunch in Germany - Walmart looks like a bloated whale compared to Aldi's relentless ruthlessness in keeping costs down everywhere possible.

What has picked up is the global market for machinery that handles recycling - Germans may hate trash, but they love making money selling machinery to deal with it, especially when it provides a chance to recapture raw materials (which is the latest set of laws, actually - Germany would rather recycle than pay for something that will just end up getting thrown away).

American companies manufacture unbelievable amounts of packaging - which is just another way to actually refer to trash. This is not the case in Germany, where a significant amount of engineering talent was focussed in the mid-90s to meeting the standards requiring a reduction in packaging. Along with designing cars to be 100% recyclable, or dealing with the mandatory collection and recycling of all electronic products.

German engineers do a lot of work that apparently is unfamiliar to most Americans - though Americans do end up buying the products those engineers designed and refined.

What Germans generally don't design is trash - they don't really see any profit in generating it.

This is a complicated discussion, but American packaging is ridiculous. Along with how Americans prefer single use products, which is one of the most striking things Germans notice when they visit the U.S. - after all, some factory has to manufacture all the plastic forks and spoons and styrofoam containers and plates and plastic cups and lips and straws and the trash bags they end up in, day after day.

I am fairly confident that at least 5-10% of native American manufacturing is related to the production of trash - like the products related to eating noted above.

Cause Germans don't shoplift quite as much?

Why is it that people think the US could be Europe without actually being Europe?


Why don't you adjust the numbers for number of manufacturing workers, instead of population? Then you will see the U.S. crushing Germany by about 50%.

And regarding packaging, who gives a sh*t?

By the way in case you are trying to make some tangential environmental point, no we are in no danger of running out of landfill space any time soon, or of there being any environmental impact to "excessive packaging" whatsoever.

I'm going to quote from this article, in the hope that it provides a bit of insight into what Germany does -
( - 'Founded in 1945, Internationale Politik (IP) is Germany’s oldest foreign policy journal.')

'A large factor in Germany’s current prowess is that it has continued to emphasize manufacturing and exports over the financial industry. Chancellor Angela Merkel once was asked by then-British prime minister Tony Blair what the secret was of her country’s economic success. She famously replied, “Mr. Blair, we still make things.” In Germany, manufacturing still dominates because German capitalism did not succumb to the rampant financialization that swept the United States and the United Kingdom during the Reagan-Thatcher years that later precipitated the global economic collapse in 2008 (though some of Germany’s banks did get snared in the Wall Street web).

Germany has many Fortune 500 multinational companies that are global leaders in manufacturing in their field, featuring brand names like Volkswagen, Daimler, BMW, Siemens, ThyssenKrupp, and BASF. But what really sets Germany apart is its vibrant mittelstand, those small and medium-size enterprises (SMEs) that form the backbone of the economy. Some 99 percent of all German companies are SMEs (defined as firms with annual sales below 50 million euro and a payroll of less than 500 workers) and around two-thirds of all German workers are employed in these types of companies. That is about the same as the EU average, but higher than the United Kingdom, where SMEs employ about 60 percent of workers, and the United States, where they employ about half of American workers.
Germany’s SME sector is broad and vibrant, and its manufacturing SMEs are world-class. Nearly a third of Germany’s SME workers are employed in the manufacturing sector, but those workers account for a disproportionate share of Germany’s total exports (about 40 percent, compared to 31 percent of US exports from SMEs). A typical enterprise has been run by the same family for generations. Located often in small towns, they thrive on a very German brand of “family capitalism.” The manufacturing SMEs are noted for a high degree of specialization and the ability to dominate niches in the market. Typically they invest heavily in research and development and the creation of their own worldwide networks to deliver outstanding products and services to customers. They combine that with a relentless drive for quality, productivity, and a focus on improving operational performance. These companies are particularly well organized in their shop floor operations with state-of-the-art manufacturing practices.

A typical enterprise will specialize in making a single, high quality product that is crucially needed by other industrial enterprises, and be the best in the world at producing it. Many manufacturing SMEs have become world market leaders in their field, dominating the global market in an astonishing range of areas. Dorma makes doors and all things door-related, Tente specializes in heavy duty casters and wheels for industrial uses. Rational makes ovens for professional kitchens, and Würth is the leading industrial supplier of assembly and fastening materials worldwide. Other examples range from Koenig & Bauer (printing presses) and Utsch (license plates) to Hegra Linear, a leading manufacturer of mechanical motion technology.

Hegra Linear produces a little-known niche product: linear guiding systems and telescopic slides. These are high tech, low-friction shelves, slides, and drawers needed in a wide range of industrial applications, including factory automation, automotive assembly, airplane production, and more. “Let’s say you have a very large, heavy battery for a metro bus that you need to have easy access to for maintenance,” says company co-owner Lisa Theuer. “You need a way to smoothly slide that battery in and out. We can set that battery on one of our high tech, near frictionless drawers that make it easy to gain access. That’s what we do.” And they must do it well, because the company has grown rapidly in the past year, ramping up their number of employees by 46 percent to 22.

Increasingly, German companies like these are providing China, India, and other emerging economies with the high tech precision tools they need to become the mass production factories of the world. These companies are not high profile, but they have a huge impact. “America concentrates on the mass market and quantity, but Germany is king of niche markets,” says Professor Bernd Venohr of Berlin’s School of Economics.

Germany has fed its manufacturing sector with a steady supply of skilled workers, technicians, and engineers via a first rate system of vocational training and technical apprenticeship. Companies also work closely with regional technical colleges, often sponsoring programs to get the graduates they need quickly. The innovativeness of German businesses proved to be a driving force during the economic recovery, with Germany currently spending around 2.6 percent of its GDP on research and development, well above the EU average of 1.9 percent. The inventive spirit is also as strong as ever: in 2009 investors and companies from Germany registered some 11 percent of worldwide patents, third place in the world ranking. And the federal government has used economic stimulus packages to ease the burden on SMEs through a cut in taxes and improved depreciation allowances.

For all these reasons Germany also continues to be one of the leading nations in several promising new technology fields, including biotechnology, medical technology, nanotechnology, computer science, and electrical engineering. The German renewable energy sector (wind energy, photovoltaics, biomass) has been expanding into international markets, with manufacturers of wind energy plants enjoying a world market share of almost 28 percent. The electronics industry and information and communications technology (ICT) continue to be two of the biggest branches of German industry.'

That is what a manufacturing nation looks like, in my American eyes - a country like China, producing inferior products for a mass market determined almost exclusively by price (though China, like Japan before it, does have its ambitions), doesn't really fit the bill any more than today's America does (which seems lacking in any ambition at all, while still proclaiming itself the leader). Anyone can start and stop the machine that produces something, and then call the machine's output 'manufacturing.' But building the machine itself requires a level of expertise and talent that goes to the heart of what manufacturing truly is, which is not the same as flipping the machine's switches and counting what it makes.

As noted in the article, it takes an educational system actually able to produce the people able to make the machinery which a factory requires to manufacture something. Which comes back to the original point of the post. (And as a general sidenote to an ongoing discussion elsewhere in this thread - the German education system is much more heavily subsidized than America's, and the result is plain to see just what that means in terms of competing in industrial global markets.)

My last block of text, while searching for a comparison of workers involved in manufacturing -
(from )
'As befits a large, modern country, America's manufacturing sector remains very large and has been growing in absolute terms. In 2009, US manufacturing accounted for more than 18 percent of global manufacturing3 and its value was higher (when compared in nominal, exchange-rated terms) than the total GDP of all but seven of the world's economies (behind Brazil at $2 trillion and ahead of Canada at $1.6 trillion). The per capita value of manufacturing in 2009 was higher in the United States ($5,800) than in France ($3,900), Canada ($4,200), Italy ($5,100), and China ($1,500). When measured in constant monies, US manufacturing expanded by about 60 percent between 1990 and 2009, nearly matching the growth of overall GDP; it grew by 10 percent between 2000 and 2009, compared to a 15 percent increase in GDP.4

But these numbers can be deceptive. America's manufacturing sector has retreated faster and further in relative terms than that of any other large, affluent nation. US manufacturing as a percentage of GDP declined from 27 percent in 1950 to 23 percent in 1970 to 14 percent in 2000 to 11 percent in 2009. While manufacturing as a share of GDP has also declined in Germany and Japan, both countries have retained relatively larger manufacturing sectors at 17 and 21 percent, respectively. The contribution of manufacturing to per capita GDP is also higher in Germany ($6,900) and Japan ($8,300) than in the United States. The most shocking, but underemphasized, fact about global manufacturing is that Germany's share of global merchandise exports is actually higher than America's (9 percent vs. 8.5 percent in 2009), despite having an economy just one-quarter of the size.' (Germany's population is also basically one quarter the U.S.'s, and absorbing the DDR had roughly the same impact on West Germany as if the U.S. were to absorb Mexico, and assume all its debts and obligations at American levels - such as paying Mexican pensions at American rates, without the Mexicans having ever contributed to the American system at all.)

Here is the actual comparison of workers involved in manufacturing -
'The consequences, in terms of jobs, are plain to see. Today, unemployment in the United States is at almost 9 percent compared to around 7 percent in Germany and 5 percent in Japan. The loss of manufacturing jobs explains a hefty part of the difference. By the end of 2010 only 8.2 percent of American workers were employed in manufacturing, while about 19 percent of German workers and 18 percent of Japanese workers are employed in manufacturing.'

I see absolutely no evidence from your posts that Germany has a superior educational system or a superior manufacturing sector. Who cares if manufacturing has decreased as a % of GDP? That just means other sectors grew faster. How can you get any less meaningful?

Furthermore it is absurd to explain the difference in employment between the U.S. and Germany on the basis of manufacturing. In recent years the U.S. has gone from 5% to 9% unemployment and Germany has maintained a structural unemployment rate of around 7%, while manufacturing employment has hardly changed at all.

Engineers in america are mostly used to create models to value bonds or derivatives of financial securities of companies. If you don't want to do or move to a country with manufacturing then you are better off not getting into engineering.

You mean move to a country like the U.S. with by far the highest manufacturing output in the world?

Sure they do. Plenty of engineers work in factories.

To a first approximation, if you had 100 more software engineers show up in Silicon Valley, you would have 100 more software jobs within a month.

First, "college has been oversold" appears several times. Is it too much to admit who is doing the selling? Could it be...academics? Not all, of course. But in 1992, you had to have a college degree to be considered for a secretary's job at the university I attended for grad school. Not cool.

Second, if we somehow managed to limit the subsidies to STEM, the result would be that everyone would major in STEM for two years and then either switch majors or drop out. Until degree inflation kicks in, which would take about...two years. Then you have flooded the STEM classes with people who have no business being there, destroying both the real and perceived value of the education.

It is shocking to me that a supposedly libertarian-leaning blog would fail to mention the most obvious and direct solution: eliminate the college subsidy completely. The STEM majors that need to borrow in order to complete their degrees won't find it overly difficult to do so. They will also be more motivated to finish quickly and do well. The underwater basket weaving types, not so much. There is an entire raft of distortions that these subsidies are causing, the issue at hand is merely one on the list.

"It is shocking to me that a supposedly libertarian-leaning blog would fail to mention the most obvious and direct solution: eliminate the college subsidy completely."

+1. The critical word here is, "supposedly." Tyler is a well-known tool, but recall how much nut-hugging Alex does when it comes to, say, Tim Geithner.

Most people think distortions are a feature, not a bug. If you can show that their misguided actions don't even achieve their naive objectives that can yield a libertarian end-result.

That is in fact the whole point. The majority of the 'rational voters' think it is their moral duty to want more college subsidy.


The salaries of engineers will continue to rise until more people are drawn into the field. As an engineer, I am a fan.

You should leave out pure math and some of the "purest" sciences (say, string theory) out of the list; there is more or less no reason to believe that these things can create jobs. As someone who did a Ph.D. in pure math, one of my most harrowing experiences has been to see a majority of the peers finding that there are so few pure math (necessarily academic) jobs out there. So these people who were, prior to taking up graduate studies, filled with romantic dreams by liberal-types who sang to them in their impressionable years the stories of Gauss and Galois and Grothendieck, eventually resign to fate and end up at purely teaching-oriented non-research jobs. Sure, this "pays" in that it lets them survive, and having math Ph.D.'s teach math in colleges is probably better than having someone who doesn't understand rigor teach math, but in terms of job prospects it is a disaster.

So if you really care for economy and the underemployed you should really ask most prospective pure-science Ph.D. students to move into applied science/engineering, biology, chemistry etc.

*Ahem* Dude, buy yourself a computer, install Linux (yourself), install devtools, learn a programming language. If your degrees are worth anything, you should be able to sell yourself as a competent programmer within six months. In a couple of years, you had better be more than competent.

It's what I did.

Programming may seem mathematical to many programmers, but most of it is mundane paper shuffling to a mathematician. It is a good way to eat, though, for those who don't like teaching.

"Mundane paper shuffling", better known as "work", you know, the thing people get paid to do.

Dude, I am not whining for myself or anything. I am just saying that Ph.D. in math, for most people who actually go for it, is a colossal waste; people might have done well to have spent that time picking up more useful skills.

As a cousin of mine said, "Most companies when they claim they need a mathematician, really mean that they need a statistician." Incidentally, if you've a mind for statistics, there's always demand for actuaries.

Yeah, if you want to spend 10 years taking tests in your spare time.

Plenty of pure math jobs if you don't mind code-breaking for the NSA. =)

Or designing highly complex financial instruments, just off the top of my head.

I think your peers are unnecessarily limiting themselves by looking for pure maths jobs. How about trying to apply the pure maths?

Most of those things just don't have applications. If at all there are, only geniuses can find them. At any rate, it is too risky to try to build a career on.

Math is a necessary foundation for most engineering tasks, and as such, teachers of maths are needed to teach future engineers math.

Granted, they will only teach some calculus and algebra, not the arcane stuff from their PhD years, but at least they will understand what they teach to a significant depth.

Only in this case, there can be a real gap. There are fundamental differences between the mind of an engineer and the mind of a mathematician. I learned that really fast as a TA. Do we expect a good mathematician to be good at teaching addition to grade schoolers? Algebra to high school students? My wife can help you through differential equations. I take it from there.

Hmmm. That is a real question. I am a "pure mathematician" by training (algebra and number theory), but have spent such a long time in cryptography + programming that I am no longer representative of the mathematician breed.

I would say that both calculus and basic algebra need deep insight of the teacher, unless they turn into a pure equation tornado - the structure "under the surface" is important there.

But as I remember some of my schoolmates, I wouldn't bear them explain things even to their peers. So you have a very good point - many mathematical pro's are borderline Aspergers, and teaching non-mathematicians is a serious challenge for them.


A good question to ask is how much exposure are people getting to these fields outside of college? My guess is the answer is a lot.

Everyone my age I know of consumes lots of music, theatre, live performance, etc....most of it for free. You can argue about whether or not open source consumption dsitributes art/history/culture consumption in an optimal way, but I don't think we're lacking for enrichment in those areas.

1% of students gets 99% of degrees in science and tech. Occupy Science and Tech !

Yeah, put down the drum and occupy the business end of a graphing calculator.

You need to get the MR award for Humor.

I have no problem with the conclusions, but I dislike the presentation of the data; by picking just chemical engineering,the smallest of the major engineering fields, it seems like the chart maker wants to show a huge discrepancy between engineering and the liberal arts degrees conferred. It feels cherry picked, which is a shame, because the same data shows 97000 engineering degress in 1984-85 vs 84000 in 2010, making his main point just as well.

Before suggesting that there is not much value in the humanities or social science, I would suggest that Steve Jobs liked to think of himself as someone who combined his knowledge of the humanities with his knowledge of technology. Why did the iPod, iPhone, and iPad turn out to be so successful? A lot of it had to do with the designers of these devices using intuition in addition to the sort of logical thinking that makes one thrive in a field like computer science. Steve Jobs famously rejected asking users what they wanted; how could they know what they want from a product that has not even been invented yet?

Another issue has to do with the failure rate of people who go into the STEM majors. I am not sure if the issue is that these classes are not taught well or whether the subjects are inherently hard. (Computer science never seemed hard to me; it just fit the way I think anyway.)

I think it is very interesting. A lot of student decisions seem to be supply-driven (what do I want to be when I grow up?) rather than demand-driven (which skill set will be easiest to market?). I think there are some advantages to that. For one, someone probably isn't going to make very big contributions to a field they don't like. Maybe fields like computer science are just not for everyone, while some people really do like it...

Still, it does seem that more people in some other countries end up with more STEM majors. So, maybe the issue is also cultural. Maybe in America, we are too consumer-oriented. When people choose majors, they think about what they would like to do with their lives too much and too little about what others in society would like them to do with their lives and would be willing to pay them to do with their lives.

I think that this entire question goes way beyond issues of just economics to issues of cultural identity. A student with a consumer mentality is going to think about what THEY want to learn and who THEY want to be. But I think what Alex Tabarrok is suggesting is that they think of themselves as suppliers. The questions would be: What do OTHER PEOPLE wish me to do? Who do OTHER PEOPLE want me to be? How can I please them, so that I can make more money?

I think it is an interesting conflict. One of the downsides of capitalism, and probably any economic system really, is that people cannot be who they really want to be. But, the payoff in Tabarrok's world is that we remain relatively affluent and get to buy lots of consumer products. At least hopefully... What happens when the supply of computer scientists increases? Does it become simultaneously not so lucrative as it is flooded by countless new majors looking to make money? Do these people who are not really inclined to the field end up with the worst of both worlds? That is, unhappy both with what they are doing and with how much they are paid to do it?

Maybe not. It could be that the demand for what can be done with computers is unlimited. It just seems as though there is an infinite amount of beneficial work that could be done with them at this point anyway. One thing unique to computers is that as the ultimate information product, the limits of our demand seem to really be the limits of our imagination. In contrast, there is only so much food that can be grown, cars that can be manufactured, and physical stuff that can be produced before we are saturated with such goods. Due to the possibility of limitless innovation, will the laws of supply and demand be more kind to this huge new supply of computer science majors that Alex Tabarrok is suggesting be incentivized?

The two real times apple tried something that hadn't been invented yet, the products flopped. Apple's successes were with refinements of products that had existed for years.

Steve Jobs, who famously dropped out of college after one semester, might not be the best argument for the value of a humanities degree.

Snark aside, I think you make some good points about the student as supplier (or maybe investor). While the demand side is important (you're not going to have a very good life if you hate your job, even if it pays well), there may be too much focus on that side of the equation.

An actual liberal education is fine. I just wonder how much there is.

One that only put you $8,000 in debt instead of $80,000 would be a lot better.

"One of the downsides of capitalism ... is that people cannot be who they really want to be."

One of the upsides of capitalism is that my (and your) career choices have expanded beyond a, rice farmer, or b, hungry.

What happens when the supply of computer scientists increases? Does it become simultaneously not so lucrative as it is flooded by countless new majors looking to make money?

Society becomes far more productive, we all get richer.

There is competition, of course, but it's not a zero-sum economy.

Oooh, my coworkers and I have been talking a lot about this recently (we're all engineers of one stripe or another). One of our thoughts has been that there's a cultural stigma on the STEM degrees when you're in high school. The nerd label is definitely there and it's something that adolescents are especially susceptible to. Additionally, once you get to college, a STEM major is going to be working much harder than their liberal arts peers. I went out much less than some of my friends and had engineering classmates who dropped out and into the business school because when you're 20 years old it's much harder to see the long term value of the degree. The value for me has been a series of very, very interesting jobs that pay me well.

My second thought is more of an economic one about the contention that since liberal arts degrees pay less they should cost less. That would be fine if it cost the school less to educate an English major. Yes, a liberal arts major requires less "stuff" in the form of labs and equipment, but those degrees also tend to bring little to no money into the school as a whole. Hard sciences pay for themselves through research grants and endowments. What other degrees pay you to go to graduate school?

Good point about the social aspects. I gave up biochem in favor of econ when I was in college. Why work harder for lower grades in classes with virtually no attractive girls?

Yet for all the tail I got in college, I'm not sure I don't regret the decision.

Oy, there were girls! Just, uh, not very many (~12 out of 200+ in my graduating class).

I'll leave attractive alone and we'll call it subjective. :)

At UCSD, the situation has gotten much better in the last few years since I started. Not sure about the exact numbers but about 20% of my STE classes were female.

Also engineering guys have no game. No. Game.

This is a really interesting topic that I've had some thoughts about.

(1) One thing that shocked me going into college was the absolute lack of foresight when choosing a major. If you're picking at random, would you select the major that pays $80,000 out of college with almost guaranteed job placement, or $40,000 with months, if not years, of job searching? Doesn't seem like a close call to me.

(2) I hate to sound like a bigot, but in the "Intro to Programming" classes, you can sort out who "gets it" and who doesn't very quickly. It's blindingly obvious. And somehow, teaching someone "rational thought" is very, very difficult. I've accepted that there's some component of science and engineering that can't be taught. So, some of these kids drop out, and some muddle along, mostly through dishonesty. In the real world, god help you if you have the latter "working" on a project for you. In the best case you can contain them and avoid having them torpedo your project. In other cases, well, it's why some companies go out of business.

The question that stems from this is, what's the value of training the marginal engineering student? Are most of the "naturally" STEM students already doing STEM? Or are there plenty of students who have aptitude for this sort of thought, but aren't doing it for whatever reason? I kind of think the answer is the former, but only from anecdotal experience.

My intro to programming class sucked big floppy donkey d*ck if you had no previous exposure to coding. Not once did anyone go over the idea of flow charting or writing out your algorithm on paper before you sit down at your computer and the professor was so paranoid about the honor code that there was essentially no help from your peers. I did well because I could memorize the syntax easily, but didn't have a clue what I was doing. It wasn't until a math class a semester later (god bless you, Numerical Methods) that I actually learned how to code.

There's much to be said about good teachers. After freshman year I became very adept at shopping all the professors teaching a subject and would beg and steal my way into the section with the superior teacher. This made all the difference in the world and saved my sanity numerous times (though it was occasionally entertaining to see another section rise up in revolt against a poor professor).

Come to think of it, the first programming course I took in college was the same. Pretty awful, the professor focused on commenting code "correctly" and memorizing arcane syntax-parsing rules (like the right-to-left rule, useful if you're writing a compiler, but otherwise if you're using the rule you're probably doing it wrong). Instead of "intro to programming" I should have probably said "introductory S&E courses." In any case, my point is that, some people get engineering, others don't, and maybe that most of the people who "get" it are already doing engineering. Maybe, I don't know, how many marginal engineers are there?

The price schools charge have almost nothing to do with their costs, and almost everything to do with the available credit lines of the 18-year-olds (i.e., this is how much debt we think you can saddle you with for the rest of your life).

Moreover, many of today’s STEM graduates are foreign born and are taking their knowledge and skills back to their native countries.

The first part of the claim is true. The part about going to home countries sounds iffy. Anecdotally I don't believe it. Most foreign STEM students I met wanted to remain. (If they did or not was partly a function of how difficult the work-regulations were ).

Have the percentages of foreign graduates returning changed? Any hard statistics?

There were a lot of articles published in 2003/2004 timeframe after a lot of post-911 laws and regulations went into effect, making it much harder for foreign students to stay in the US: you had to return to your home country in order to apply for a visa to come back to the US (and any mistakes meant you were barred from returning to the US). The difficulty of getting a student visa in the first place went up and many were going to universities in other countries instead.

For the record, "micro-biology" is a rather old-fashioned term in a field where the discipline and sub-discipline names are extremely fluid. Does Alex mean 'molecular and cellular', as opposed to 'integrative and ecological' biology majors? I actually find his statistic for 'micro-biology' degrees implausible (only 2480 bioscience degrees? or only 2480 specializing in biophysics, molecular biology, cellular biology and developmental biology, as opposed to integrative or ecological?) and wonder if it's a terminology epiphenomenon. (I've tried to look at the NCES data, but it's not easy to find one's way over breakfast).

Here's a little data on 2009 Bachelor's degrees in bioscience from ONE rather small public university system, Montana's:

Biological Sciences: 47
Biotechnology: 11
Cell Biology and Neuroscience: 64
Microbiology: 17

That's 139 lab bioscience graduates from just Montana public universities, which is surely not 5% of all US college graduates! And then another 72 in applied biological science (which clearly does not belong to 'microbiology', but are equally clearly STEM fields). I suspect that counting "Micro-biology" graduates is an exercise in tracking a field that is being increasingly re-named at more and more universities, which makes it look like the field is shrinking.

Related fields requiring substantial biology:
[Animal Science: 26
[Environmental Science: 5
[Health and Human Performance: 8
[Horticulture: 9
[Land Resource Sciences: 6
[Natural Resources/Rangeland Ecology: 9
[Plant Science: 2
[Range Science: 3
[Soils: 1
[Watershed management: 3
[I'm assuming that "Health and Human Development" is teacher training, not science, here.

At some universities, your degree will simply be "Biology" whether you concentrated in microbiology or ecology. I don't know how the data Alex cites accounts for differing practices on how specific the degree title is. However, the point about STEM graduates not increasing while liberal arts are is valid and of concern.
I don't think microbiology is all that old-fashioned. Although it is not one of the newer disciplines, microbiology (understood as the study of microbes - bacteria, protozoans, viruses) is still a current discipline. However, Alex's concern about "Who will solve the problem of antibiotic resistance?" is typically driven by Ph.D.'s, not those with a B.S. It's generally agreed that there is a glut of Ph.D.'s, even in most sciences.

But we need helpers. The miracle of McDonald's, in my opinion, is that it allows people you wouldn't let fill your gas make you a hamburger.

While I see this point...the image of a homeless English major carrying a sign reading "Will read literature for food" is a bit too close to the truth for too many...I think Alex ignores both some anecdotal evidence to the contrary and some basic tenets of economic theory in his argument. First, there are the college drop-outs like Steve Jobs who have certainly contributed to the economy primarily through their design sense, not their technical expertise. In the wake of his death, Jobs has been praised not for his technical know-how, but his design sense and aesthetic. Similarly, Mark Zuckerberg is, by current standards, a lousy coder; however, his understanding of human psychology and his willingness to take risks has resulted in the the creation and growth of Facebook, a major economic engine.

Similarly, what were the majors of the Wall Streeters at Bear Stearns, Merrill Lynch, et al? I'm guessing they were STEM folks...particularly math...and their innovations drove this country's economy to its knees.

And, as a final point, suppose that Tabbarok's plea was answered by the masses and the number of STEM majors and humanities majors was reversed? Would we not, due to the law of supply and demand, be accosted by hungry STEM vagrants begging for loose change to use in an internet cafe?

This is to say that I see Tabbarok's argument as overly superficial. Certainly STEM majors are important parts of our economy and we could do a better job encouraging folks to grow into those fields. However, humanities majors also lend significant value to our lives...though perhaps in ways that economists struggle to measure...and we would be a poorer society without them, even if we were "economically" richer.

"Similarly, what were the majors of the Wall Streeters at Bear Stearns, Merrill Lynch, et al? I’m guessing they were STEM folks…particularly math…and their innovations drove this country’s economy to its knees."

You've just demonstrated that you know nothing about how these organizations actually function.

I'm a little in the dark...ha, ha...about what organizational functions I need to know about; could you be more specific?

However, to address what seems to be your point, Angelo Mozilo, David Loeb, and Eric Sieracki (Countrywide) all had B.S. degrees. Richard Fuld (Lehman Bros.), B.S. degree. Lars Norell (Merrill Lynch), B.S.. Harin De Silva (Merrill Lynch) B.S.. And, it is fairly well known that Goldman Sachs and their brethren recruit out of the math departments of the top universities in order to fill their ranks. I'm not a 99%er by any means (despite the fact that it seems clear to me that immoral actions by Goldman Sachs and other companies were at the heart of the financial collapse some four years ago); I just want it noted that not all innovations undertaken by STEM type folks, such as the creation of trenches of unstable MBS and CDO products and the subsequent deception practiced with the ratings agencies (how creative!) are positive.

I must add that this is only part of my point...I'd be happy to hear more specific comments about the other parts.

I don't have a citation, but recall an article a few years ago noting that GS and investment banks in particular chose the smartest _liberal arts_ majors for their recruitment at the BA level. (When they wanted mathematicians, they recruited PhDs in math or, notably, physics.) But for entry-level employees, they preferred those who had shown both intense work ethic and the capacity to produce outstanding work (aka a summa thesis) in literature or history or philosophy, fields where there are no fixed goalposts, and where the shape of the field is as much part of the discussion as anything else -- rather like life.

That's true to an extent, though as the examples I've shown prove, it is certainly not universally true. I guess more of my point is the nature of value and the importance of innovation. I would guess that there are very few true innovators in any field...which is why companies like Facebook and Google compete so hard for them. But there has to be a reason why Americans don't enter those fields...we are, after all, rational actors, right? So what is it, in the face of overwhelming evidence that a STEM degree confers economic stability and prestige upon a person, that causes folks to go in a different direction?

"However, humanities majors also lend significant value to our lives"

Some humanities majors do. But most of those (great writers) do not need their degrees to offer their contributions. The vast majority of humanities majors contribute to our lives in careers that had nothing to do with their major.

I guess if "significant value" lies only in the creation of great works of art then you are right. I think for many many people significant value lies in having a great kindergarten teacher, a trusted therapist, or a wise judge...these types of non-STEM folks add significant value to our communities both directly and indirectly and their work in their college majors prepared them to do so.

As a theatre phd student currently teaching urban undergraduates, my drastically unpaid labor in a humanities course is very frequently the first time anyone has ever taken the time to tell my students how to write properly, how to compose an argument, how to look at two sides of a problem, or even, in some cases, that they have brains, talent, and ability in their writing and argumentation.

The first time in their LIVES.

And their faces light up as they see a chance for a future, because they understand that there is finally someone who can open a door to them in a way they can translate into other fields.

This is stuff that they should get in high school, but humanities fields in general (whether high school are college) are ideal places for teaching skills that any thinking, functional member of society should possess.

* or college

No discussion of business majors is a conspicuous absence in this post. Painting humanities and social science majors as the enemy is wrong-headed when you realize that business schools are siphoning off more and more of their students already, and the research shows that the best way to learn as little as possible in college is to be a business major (while majors that require intensive reading and writing show significant gains). If you want to an area of universities that should receive zero subsidies, there you go.

Absolutely correct. The impression that many get is that too many college students are majoring in English or other humanities subjects rather than science and engineering. This dichotomy between sciences and humanities reflects an outdated view of the university. Today, most students major in non-traditional subjects -- neither humanities nor science and engineering. A full 21.7% of degrees awarded in the 2008-2009 school year went to business majors. Communications majors were 5.2% of the total. Education majors were 6.4% of the total. Nursing and other health fields were 7.5% of the total. Recreation majors were 2.0% of the total. Security and protective services majors were 2.6%. These six majors alone constitute 45.4% of total graduates.

By contrast, English majors were 3.5% of total graduates. The quintessential bete noir of those who argue for increasing science and engineering degree holds, area/ethnic studies, was a mere .5% of total. The reality is not that we have too many people majoring in the humanities. If anything, we have too many people majoring in things that were not even part of the university a generation ago.

But "business" majors are the sort of thing people like Tabarrok would approve of - since kids take them for the purpose of making money, which is what he approves of.

And yet the graduates of business programs are notoriously poorly educated and under-motivated on average.

Um, no. Business schools teach accounting, law, finance, operations management, even software application development. (Yes, marketing is a fairly worthless major, everyone else makes fun of them.)

By what measure are you claiming they learn "little?" Word-intensive humanities majors show significant gains in what? The answer certainly doesn't seem to be "skills valuable to employers."

"Students majoring in liberal arts fields see "significantly higher gains in critical thinking, complex reasoning, and writing skills over time than students in other fields of study." Students majoring in business, education, social work and communications showed the smallest gains. (The authors note that this could be more a reflection of more-demanding reading and writing assignments, on average, in the liberal arts courses than of the substance of the material.)"

Read more:

Also, on B-schools and how academically lax they are, read here:

Yes, but that study also says are small across all groups, and doesn't tell us whether liberal arts students started out further behind.

The second link is self-explanatory: especially in “soft” fields like management and marketing, which account for the majority of business majors." Marketing is usually a worthless major and everyone knows it. Other fields like accounting can be very difficult -- I went to the school with the highest CPA pass rate, and about 3/4 of students had dropped out before graduation. MIS is bascially Computer Science from a business angle and often has many of the same requirements.

*gains are small

What you fail to realize is that many are turned off from math and science (your so called STEM majors) because they view these jobs as having a poor future. You claim that there are many jobs in STEM fields, but the entire IT industry has moved overseas. Its no surprise that many students feel that a major in computer science would be useless. They will have trouble finding a job unless they want to move to india/china. Engineers also are unable to find high paying jobs, since there are very few of them. It makes more sense for students to major in something random and "fun" or something with buisness skills like commerce or economics, since then they can possibly get a job on wall street and make the big bucks.

Your argument that students are not following the money is flawed because it fails to take this into account.

* the entire IT industry has moved overseas*


Actually, visit any IT department and you'll find a large number of foreign imports.

Yglesias says humanities and liberal arts students learn more, with graphs to back up his argument.:
I don't know what a "CLA score" is, but it's possible STEM just attracts smarter people, so you can't assume if more dumb students went into STEM fields they'd be more productive.

Just anecdotal evidence, but I took some philosophy and history courses for fun in college. I found them to be much more rigerous than my business courses and to be just as challenging as my science courses.

uggghh. And I just spelled rigorous wrong

I was an aerospace engineering major and took philosophy and history classes and found them to be a complete joke. That said, I was not at a school that was highly ranked for humanities.

Endogenous. Some fields with few job slots are going to tighten their filter. STEM is hard because it is objectively difficult.

Same. My humanities classes were ridiculously simple-minded. In a few classes I wanted to stand up and yell at everyone for devaluing my degree with their stupidity.

But then I would have been a jerk.

To put it as delicately as possible, students attracted to math, science and engineering tend to be more analytical than your typical college freshman, more open to being guided by real world experience and data than by inner desires or dreams. Or to put it more bluntly, smarter. A long as the number of jobs available to STEM graduates continues to shrink as manufacturing companies and their research departments head offshore, the number of smart, analytical students willing to commit to those career fields will not increase. Creating a surplus of young science-savvy job applicants might be good economic news for the firms that remain behind, but not for their prospective employees.

Ah, the self serving ego stroking of the STEM crowd....hillarious. I would love to see most STEM majors attempt a decent paper.....

That said, having met many STEM majors (including the kid who ended up getting a Math PhD from MIT), yes, most are very smart. And the humanities kids and social science kids are also smart.

Its those communications and business management students that leave you shaking your head.

Nobody actually reads the papers in STEM. They look at the pictures.

This is a great CS paper where you only need to read the pictures:

Soooooo... I don't usually do this, but you seem to have spelling and grammar errors in your post derailing STEM's ability to write (hilarious, it's). That annoys me. Pots, kettles, etc.

We are getting warmer. Do we want an education system to sort out who is analytical, or do we want to train people to be a little more analytical.

We are back to the human capital, signaling, training issue.

You've met the wrong Humanities people....

Before we mint more STEM graduates, let's look at the reality of the job market for them. Between 2000 and 2008, the BLS reports the following changes in the numbers of people employed in the various engineering professions:

Aerospace Engineers: 43.20%
Agricultural Engineers: 12.50%
Biomedical Engineers: 122.22%
Chemical Engineers: -3.94%
Civil Engineers: 19.83%
Computer Hardware Engineers: 24.50%
Electrical Engineers: 4.69%
Environmental Engineers: 4.42%
Industrial Engineers: 8.48%
Materials Engineers: -26.06%
Mechanical Engineers: 7.69%
Mining and Geological Engineers: 9.23%
Nuclear Engineers: 20.71%
Petroleum Engineers: 143.33%

Total Persons Employed as Engineers: 9.22%

Unless these numbers are strongly supply constrained (and salary data do not suggest that engineering wages have increased drastically), I am left to question to basic premise that we have a shortage of engineers. Civil engineering, which is the second largest engineering profession, is driven largely by government spending and showed good job growth. The other two large engineering professions, mechanical and electrical engineering, were stagnant. Likewise, the increase in aerospace engineers, who had been devastated by defense cuts in the early '90s, is a product of unsustainable defense spending. Petroleum engineering, which had the largest increase in employment, is, like bio mechanical engineering, among the smallest specialties. The basic picture that this paints is that the large engineering professions that are government-supported have done reasonably well, whereas engineers in the private sector, with the exception of small specialties like petroleum engineering and biomedical engineering (arguably this is supported by unsustainable government spending as well), have suffered along with the rest of us. Creating more engineers will only make the problem worse unless we propose to increase government spending to put them to work.

Why should there be more people employed as engineers if there aren't any more people getting engineering degrees? Maybe we are just at replacement level here? Not to mention you have not made a comparison with any other sectors.

Total engineering B.S. degrees awarded in the 2008-2009 year was 69,133. I don't know whether this is replacement rate or higher given the 1.6 million total persons employed in engineering. I suspect it is more than enough to support current growth levels in the profession.

But many foreign graduates also come to the United States and work. Thus, the pool of engineers is much larger than the pool of U.S. B.S.E. graduates.

Another point worth mentioning is that we are graduating more engineers. In 2000-2001, we awarded 58,315 B.S.E.s. As I have already mentioned, in 2008-2009 we awarded 69,133 B.S.E.s. That is an 18.6% increase during a time when total employment only went up by 9.22%.

If I had time, I would run the salary data to see if their really was increasing demand for engineers.

Cliff, the household survey shows that total employment increased 7.1% between Jan. 2000 and Jan. 2008. Although the numbers are not directly comparable, the BLS numbers show an increase in engineers of 9.22% during approximately the same time. Thus, the number of engineers did increase at a faster rate than jobs generally, but the difference is relatively small. Both numbers reflect very poor job growth, especially relative to the population increases.

Thanks. Even if demand for engineers is not soaring, they do get paid quite a bit more than communications majors, logically it seems they are therefore in "higher demand". Perhaps if there were more STEM majors, STEM incomes would go down, but still people overall might be better off.

To me, this is exactly not the point. We are going to grant X degrees. More should be more practical, more rigorous, more objective, more mathematical.

Similar to what Bryan Caplan said, "get an economics degree and then call whatever you do economics," just not economics.

Andrew', I agree, but overall income/employment might not increase that much.

Chemical engineers, Biomedical, and Petroleum are largely the same people. It's actually pretty remarkable considering the outsourcing and economy.

You want to look at unfilled job postings, and the relative salaries and unemployment rates of engineers. Looking at the number of jobs doesn't tell you anything useful about whether there is a shortage of engineers.

This is the sort of logical analysis you learn in STEM classes.

Your argument is short-sighted and unfortunately all too common; a student's major does not inherently bind them to that career.

The value of an undergraduate degree is more than vocational, it is about teaching life long skills as young adults transition into adults. Skills like critical thinking, communication, self sufficiency, and research/study skills are just as important, if not more important, than the major they are studying.

Expecting young adults to choose their path in life is an absurd notion. Giving the tools to recognize and pursue their passions in the future is quixotically more practical in the long run. As someone who graduated with two liberal arts degrees, I had very little trouble transitioning from translating Aristotle's ancient Greek to self studying thousand page computer science tomes because I had learned the lessons of self sufficiency as an undergraduate.


I'm pretty sure you're the exception to the rule. Most liberal arts majors I know would not be able to teach themselves computer science. I also feel that you probably could have taught yourself computer science without having gotten two liberal arts degrees first, thereby confirming the author's proposition that college is "oversold".


First off, thank you for taking the time to reply.

That said, I could not disagree more!

The upshot of my argument is there is an overlooked value to a college education: the time to mature as an adult and learn valuable personal development skills. These are the skills that matter much more over the course of a lifetime than a specific domain knowledge learned in a major. There are always advanced degrees to pursue domain specific knowledge, or sometimes ways to teach yourself new skills on your own.

While it was my personal experience to develop these skills through the pursuit of Liberal Arts degrees, they can be found in any major. The argument about whether more students should study "Liberal Arts" or "Sciences" is a red herring. In the context of an undergraduate degree, it is less important what you majored in than it is what you matured in. I was not "pre-wired" to be a self sufficient studious person. In fact, coming into college I was quite the opposite. But through careful coaxing by advisors and the environment, I saw the value in these principles and have continued to develop them post graduation.

The truth is the world doesn't need more computer science graduates. It needs more hard working people passionate about what they do.

Thanks for adding some facts.

That was meant for KLO

Also, does anyone really think that a society of engineers would have as robust a culture industry as we have? How about as varied a consumer services sector? Its funny how China has lots of engineers and no one wants to buy their handbags, movies or design.

I happen to be in touch with many international engineering (former) students from India and Pakistan who pursued Masters degrees in Engineering in the US. Anecdotal for sure, but I am always totally surprised by the number that seem to walk into engineering jobs in the US. I am pretty sure that they have a negative premium because they are on various types of non-permanent resident visas. I have an affinity for these folks, but I always feel a tinge of concern for American engineers they might be displacing because (I think) they are willing to work for less.

Or maybe, it is also because they are willing to work for longer hours, complain less, accept less generous working conditions, and do the same or better job. (These folks always note that the Masters level work they are doing here was already covered in their undergraduate engineering degrees, so they find school here easy; not that I am proposing we do away with our broad based college education, but just another point to bear in mind).

Tough situation to figure out what is right and wrong.

I agree, RM is right here, that there are a large number of international people who have come to the US to do technology work. They are actually thinking of discontinuing the computer science program at my alumni because of poor enrollment. Meanwhile they have made Music a major. It's definitely lopsided, but it's a response to what the market is driving, and their market is what students want to do!

To add to the color of what's happening, there has been a shift in dynamic of parents and children in America. Once it was a virtue to pressure your children to get a good education so they can get a good job. Now the message to children is to express themselves and choose a passionate career so that they can be happy in life. I find nothing wrong with telling children that they can be whatever they want, but if your kid isn't a genius actor or musician, perhaps the parents should be guiding their kids in a different direction. After all, we can't all be starving artists...

I do know people who were overpraised for dubious talents in writing and acting in high school (to "raise their self-esteem") who were destroyed when they got to college and realized the truth...

I would note that students in the visual and performing arts may ultimately be contributing to one of America's major export industries: film. Naturally most of them aren't becoming Hollywood actors, but it presumably helps to have a large pool to winnow from.

The value of a college degree hasn't be oversold- just misrepresented. First, there was a time when hiring a college grad, pretty much regardless of his or her major, assured an employer of getting a bright, hard working employee, with great potential. Now, with too many kids going into debt to fund degrees in much less demanding programs, that is no longer the case. Today, a generic 4 year degree only insures that you're getting an employee with basic literacy and math skills you might have expected from a top high school graduate 20 years ago. Of course, the better Technical, Engineering and Pre-Med programs are still producing world-class graduates- but those folks are still in high demand.

First, the USA are discovering now the problem afflicting Europe since the '70s. With free colleges, humanities have been a huge problem to economic growth, with tons of graduate young people requesting a job as journalist, pundit, reviewer, movie maker, etc etc

The problem is: is it less distorting to do not consider the (huge) externalities, remove incentives and let the market works (solving exactly the wrong problem) or is it less distorting to add closed numbers to overcrowded faculties and let the regulation works (solving uncorrectly the proper problem)? I really don't know, probably a mix of both.

Sure, the German economy has been devastated by too many university students studying the humanities.

Well, OK, no, that is a ridiciulous statement - but it brought a grin to my face while I typed it.

Germany != Europe

Germany = Largest country in Europe (as we think of Europe) and typical of much of Europe (e.g. France, northern Italy, UK).

Hmm, I hear a lot of people extolling German exceptionalism these days. Certainly it seems to be in better shape than most European countries including the UK and France. Northern Italy is not a country! (yet)

That's the biggest hurdle, as far as I can tell: STEM is hard.

It used to be, if you were a bright go-getter, you'd go to college and get a STEM degree, and if you were closer to the median, you wouldn't go to college and get your experience on the job. Now, the median student is expected to go to college. What's happened is that the STEM student cohort hasn't changed--they were STEM before, they're STEM now. But now the median student is becoming a college student--but will be expected to have a median career. So, nothing has changed for the median student, except the requirement of student loans.

and lose 4- 6 years of work experience during their most important learning years.

Scary version - that 4-6 years is designed to intentionally keep people out of the workforce.

I'd definitely agree with this except that the overall population has grown while STEM degrees have remained stagnant.

most likely because most people go college just because it's the thing to do after high school, not because they actually want a college degree to do something with their lives. why do you suppose there's the Asian parents stereotype and why Asians are taking over the industry? We are allowed to choose our college degrees, but only to a certain extent. If our parents disagree, college is unfunded. In America, especially, there's the whole "do what you want in college". The sheer number of English Lit majors that I have met ... ALL unemployed! Did you not think about your employment opportunities when you started a course in a language you use every day that you could probably learn being home tutored instead of spending thousands at college to live off your parents??

a lot of people have made comments about how they're successful without a college degree, but how many is that vs the entire population? how many of the richest and most powerful people in the world in a STEM industry have college degrees? Most. You could count on one (maybe even two) hand(s) the ones who don't - Steve Jobs, Michael Dell, Bill Gates etc. Even then, they're borderline douchebags (Gates and Jobs at least) to get to where they are/were. The bottom line is still: You need a college degree for STEM unless you're a frickin genius with a business mind or just simply the "Knack" (cue Dilbert)

It's all part of the bubble economy. With all those 30 year old finance and consulting airheads making millions for doing simple and overrated work, why should students select something hard?

Let's face it, the name of your school matters much more than what you study. Also, studying something easy gets you higher grades and lands you interviews with respected companies. Then many of these kids may end up getting MBAs, later on, which is also an easy degree.

There are no longer financial rewards to difficult study, at least students don't feel that way and their schools certainly don't try attempt to convince them otherwise.

I'm not sure you're familiar with finance, consulting, or indeed the concept of employment generally.

“It’s all part of the bubble economy. With all those 30 year old finance and consulting airheads making millions for doing simple and overrated work, why should students select something hard?”

Just how many 30-yr old millionaires do you think there are in finance and consulting?

Young people in those fields are very-well paid, but they also work 80-100 hour weeks doing some very tedious work. It ain’t working in a rice paddy, but it also ain’t all that glamorous at the analyst and associate level. By the time people are making anywhere near millions, they’ve been paying their dues for quite a few years.

There may be a few young people making serious bank on trading desks or in asset management, but those people are some combination of extremely lucky or extremely good. They are not the norm.

Grade inflation is the unspoken problem. If the hums and SS students were graded as strictly as STEM kids (or even as strictly as Latin classes were in pre WW2 Ivy League) there would be no mismatch. Furthermore, if most jobs didn't punish STEM students with low grades by denying them interviews while interviewing high scoring grads of "xxxstudies" courses, there would also be no mismatch. But until the day comes when it's better to have a C in STEM subjects over an A from fluffy majors when applying for the median general interest job (not specialty jobs like programming), you will always have this "market for lemons" problem.

Yeah this is a big issue. STEM students tend to get WAY lower grades then Humanities. Amongst humanities majors it's considered ridiculous to have grades below 90% where as in many STEM programs having a midterm where the class average is a failing grade is not at all uncommon. However when you go for a job the employers generally treat all the graduates the same "Hey this history major has a 4.0 GPA but this guy with a physics degree only has 3.0 - this physics grad was definitely a slacker." Some employers take the difference in difficulty into account but not many.
You want a job in finance? WAY better to go into a jokey subject like economics (Sorry but at the undergrad level Econ is a huge joke) and smoke every course then try your hand at pure math or physics and do mediocre.

Less of an issue for employers I think -- you don't hire an A history major over a B physics major if the job involves physics.

OTOH this is a serious problem in graduate schools, esp. law school, where they use a formula that treats all majors as equal.

I'm talking about general jobs that don't necessarily require specific knowledge - the number of jobs that ACTUALLY require knowledge of high level physics, math, etc. is very small and only the best of the best are going to get those jobs (ie the B physics major isn't going to get the physics jobs any way). Most STEM majors have to go for jobs outside their field and since they probably have lower grades then Social Science, Business, or Humanities grads they're at a disadvantage.

Why would you take a job like that with a STEM degree? There are tons of STEM jobs -- employers have trouble filling them, because STEM is hard. B students are certainly employable. Unless your major was pure theory with no practical applications (like pure math or something) you're very likely going to find something related to your field of study.

Employers are NOT having trouble filling these positions. Listen basic economics states that if they were having trouble then pay should be skyrocketing but it's not really in those sectors. The reality is all these employers only want the best of the best and particularly with years of professional experience to boot. I'm sorry my major was CS and Math and I've looked for two solid years anywhere and everywhere. It's just not true that there's as big a demand as they say.

And why would you take a job in a non-STEM field? Well pay and job prospects are much better. The people I know from school who got jobs rather easily majored in Finance, Accounting or Economics and because they had very high marks (these courses are not difficult I took a few Econ electives and getting a 90%+ is no problem at all) and could therefore go and get hired at Deloitte or other consultancies or in banking/insurance. Yeah it's not exciting work but what is and it REALLY beats having no prospects - especially when you've been out for a few years and now you're considered out-of-date.

You seem to be talking about a totally different job market. Visit STEM workplaces and you'll find a large number of foreign workers -- who are well-paid. Many of them don't even speak English very well.

Accounting majors get hired by accounting firms because they know how to do accounting -- in my state you can't even take the CPA exam without an accounting degree, and your STEM degree would not help get that job you no matter what your grades are. Finance is a little different because they like the STEM guys to become quants -- but those can be brutally competitive jobs.

I was at my local university for a meeting recently. I saw a prospective student and his mother looking lost so I offered to walk them to their destination. We made small-talk along the way. The prospective student told me he wanted to major in theater. This is a $50k/year private university NOT known for it's theater program. I responded by saying, "Oh, my secretary got her theater degree here!" The mother understood the point of that comment.

But, at the same time, I have a number of friends with graduate degrees in STEM subjects (Ph.D's, MA's, etc. in engineering, math, etc.) who often complain that it's hard to find a decent paying job when so many companies are outsourcing such needs to other countries with cheaper labor. My "pure" math friend freely admits that despite his vast knowledge of theoretical math (he's truly brilliant), he doesn't really have any marketable skills, and that even for the rare job where pure mathematical brilliance is of use, some equally bright person on the other side of the world is willing to do the same thing for a fraction of the cost. Of course, anecdotes are not data. I'd wager they're in still in a better position than my secretary with a theater degree. Then again, of all my old college friends, by far, the most successful were the ones who pursued their passion and succeeded. In their case, that was music, and they've each one a grammy (and not together...each individually for separate things).

I'm not sure what my overall point is other that, in general, I agree, but the STEM distinction is probably too simplistic. I think the secret to great success involves a mixture of true talent, hard work, and quite a bit of luck. As an economist would tell you, specialize in where you have the comparative advantage. It may indeed be better to be the best photographer than a mediocre engineer. The problem is, many college kids confuse their passion for skill. There is something to be said for a "safer" choice, but there is also something to be said about taking on some risk.

One thing is only the E in STEM really gives most people any career prospects - the STM are just as much a crap shoot as a history degree.

I'm working on my M.F.A. in Interactive Media and my bachelor's was in Visual Arts. I have never had a hard time finding lucrative work, even during the recession. In the video game industry, engineers do get paid a bit more than artists and designers for entry level and mid-level work, but there is generally more demand for artists on a given team than for other specialties, particularly on big budget games. I see where you're going with this account, but the focus on STEM is reductive. There are lots of well paying jobs available for arts graduates, and I expect there will be even more in the future.

I agree. The problem isn't so much an orientation away from the STEM subjects, but an orientation away from commerce. There are so many kids in school or with random degrees who have vague plans of working in "non-profit," or "public interest," or "activism." And along with that is the mentality that someone else (i.e. the taxpayers) owe them the means to support themselves while living out their dreams. If you have some random passion in life, the onus is on you to figure out how to make an honest living doing it.

No there's not, there's tonnes of kids in school who want serious careers in engineering, science, IT, finance, etc. This idea that it's all kids with vague notions of wanting to be do-gooders is flat out nonsense. There are hardly ANY jobs for entry-level candidates whether in STEM fields or elsewhere. That's the real story.

No, the job market is tougher, but employers are still importing huge numbers of highly skilled foreigners for those STEM jobs -- and at good pay.

In what way is what you're saying exclusive of what I am saying? If you re-read my comment you'll see that "so many kids in school," is not an indictment of "all kids."

If you doubt the value of a liberal arts education, then most likely you haven't had one, or have had a thoroughly mediocre one. A true liberal arts education teaches you how to learn and how to think for yourself, and those fundamental skills apply to any profession, not just technical. My classmates are extraordinarily successful in highly technical fields, even though they majored in philosophy, anthropology, french literature, etc. Do not, in the words of Mark Twain, let your education interfere with your learning. And do not, under any circumstance, try to get a liberal arts education at a crappy university where students are just more chuck for the meat-grinder, and for whom a liberal arts major merely provides more opportunity to get drunk and laid. Do not blame liberal arts for what is wrong with American education:There is an epidemic of idiocy that would not be nearly as bad if people had learned how to think for themselves.

"It’s all part of the bubble economy. With all those 30 year old finance and consulting airheads making millions for doing simple and overrated work, why should students select something hard?"

Just how many 30-yr old millionaires do you think there are in finance and consulting?

Young people in those fields are very-well paid, but they also work 80-100 hour weeks doing some very tedious work. It ain't working in a rice paddy, but it also ain't all that glamorous at the analyst and associate level. By the time people are making anywhere near millions, they've been paying their dues for quite a few years.

There may be a few young people making serious bank on trading desks or in asset management, but those people are some combination of extremely lucky or extremely good. They are not the norm.

Fascinating discussion.

I'm curious to hear people's thoughts on military experience as an alternative to college. I'm a veteran, and self-taught developer/designer, and I've had a lot of success despite no degree. Personally I tend to agree with the others who mention less tangible benefits to college (critical thinking, communication, self-reliance) as being among the most important. I do aknowledge that my "research/study skills" are probably lacking, but with the recent glut of open-source academic material (Stanford, MIT, Oxford) available on the web I've found more than enough pointers to get me started.

It's not just about choice. I once taught several undergraduate econ courses (including econ 101 and several upper division course) at a UC campus. College students simply didn't have the math and analytical skills that's required for these courses. Their knowledge in math is probably only comparable to 9th graders in China and India. How can you expect them to do well in STEM majors? The problem is with the K-12 curriculum in the United States.

Yep, we get to middle/high school and our performance falls off the charts. That's because our adolescent-level public schools are basically a prison slash dating service staffed by unfireable drones.

+1 Hit the nail right on the head.

Its a shame that college experience is being measured solely in terms of wages. What about the value which we all place on beauty ? Civilizations are not just economies, they are more than that ...

When college just cost your 4 years of your life, it wasn't too crazy to study the history of romance novels. When it costs $90,000, it probably is crazy.

The easy credit to students has just bid up the price.

That's not what is being said. Wages are not being use to measure the "college experience." Wages are being used to measure the over-or-under-saturation of the market for a particular set of skills. If the wages of pharmacists are going up and the wages of curators are going down, it's a sign that there are not enough pharmacists and too many curators. It is not a measure of which society values more.

I'm not a statistician, but I know enough to know that the "science" requires massive amounts of interpretation to be useful, and it is the easiest thing in the world to manipulate results to prove any point you wish.

I would love to see the breakdown of the studies comparing college grad pay vs. non-college grad pay, particularly with respect to homeschoolers. There are plenty of intelligent, c...ompetent, "well-reared" and literate people who choose not to go to college. But in the "non-college" demographic, they are lumped together with the hoards of neglected, illiterate, recently-immigrated, non-English speakers that don't go to college because they have no foundation for real learning.

This grouping together, I believe, skewers the numbers considerably. And it gives a false picture of reality for young, intelligent people considering doing something besides indenturing themselves to a federal loan program for the rest of their lives when they're not even old enough to buy a beer.

In other words, if you come from a family that taught you the basics and maybe a little bit more, then your future may well be comparable - if not *highly* superior - to that of the average college grad- not just in terms of quantitative income, but in terms of qualitative conditions, such as self-reliance, self-direction, and freedom from debt.

Think about it. . .D

What is interesting about this post is that the headline "College Has Been Oversold" let's people respond with their feelings, and not their facts.

What if the headline had said: "Sausage is Oversold"

You would ask: What type of sausage?

There are more types of colleges and majors than there are sausages.

College is highly differentiated, yet the headline is: "College is Oversold"

What kind of Sausage?

Actually, I think that the headline is more or less correct.

Quite a lot of people tend to think about getting "A degree", with A meaning ANY.

If I were a surgeon, and operating on your brain, would you want me to be more or less correct.

Hehe, good reply, but headlines are not head surgeries :-) It is hard to be absolutely correct in <= 10 words.

Also, Marian, if you get ANY degree, unemployment and earnings statistics still show college graduates with both higher income and lower unemployment than those without.
It is Less correct to say that ANY college degree has no effect, .... more or less.

NO the STEM thing is a SCAM as well. Sure if you're in the top 10% of your class you'll do great but if you're just an average, or even somewhat above average student then you might as well have majored in Sociology and partied the whole way through university.

No, I am going to stick with Milton Friedman on this one. Subsidise the liberal arts because if society doesn't pay for them, we will end up with too little critical thinking about society. If Friedman endorses a subsidy, it must be doing something right.

These statements are very misleading. I don't think that we should subject ourselves to the idea that in order to have innovation in the infrastructure and economy in the United States is only through the development of an independent science and math centered workforce. By saying such things you undermine many basic principles in education, and the influence of all those other STEM careers, such as the fact that we do speak English here in the United States, that art and physical programs, as research has continuously shown, only serves as catalysts in the learning process in schools, and that perhaps media is the bridge to the intercommunication of the masses and the new method of communication worldwide.
Is the problem at root the fact that people are choosing the wrong degrees when they attend college? Are you suggesting that liberal arts programs should be cheaper? I don’t understand your bias. If anything, this article should make you realize that there is a problem that the subsidies are not enough for economic growth. What you are suggesting is that we should subject ourselves to provide a workforce that helps the giant companies. Companies obviously require to ends to the spectrum in the workforce: they require cheap labor for physical production (minimum wage) and college educated, STEM degree recipients. In that model, we would have even a larger disparity in the gap of classes in this country. Giving people two options, either go STEM or minimum wage. This is not only idealist on an extremist point of view, but it is also beyond irrational. Where do you leave teachers, news companies, government related jobs, social workers, all those careers that aid in the well being of society in the social aspect? History speaks for itself; they are required careers, and if it requires a college degree, the investment reflects in the prosperity of society. The problem is that today, there is little access to that workforce, because fundamental conservatism says those careers and jobs are not required (i.e. Planned Parenthood, NPR, PBS, Amtrak, etc.) devaluating their value as the years roll on. Since there is a high supply for these careers, economics tells us its price drops.
So where does the government and the occupy movement come into play? First off, they should never work for a top companies in the belief that if they are prosperous so will we. That is the occupy movement, defying the system of Reaganomics and trickledown economics, which history shows to be a failure. There should a system of accountability that holds businesses accountable for the mess that they put is in, and don’t put it on the back of the general public. We pay money in taxes into the government to provide us back with social services for our benefit, not to bail out companies, get us into a humongous deficit and to never blame it on the fact that the government is loaning money to college students who cannot get a job after college.
I wish that your argument was a little more elaborate, because that alone, is very misleading.

You must be in public education, nothing that incoherent could have come from the private sector.

Can you express yourself in less words, but with a structure? This is unreadable for me, and I am a voracious reader.

If you cannot, you're probably a teacher.

Just thinking it over, Alex sounds a lot like Thomas Friedman. We should heavily subsidize STEM's and shun the humanities like China. And like China, our economy will grow and we'll all be happy. I don't think it's that simple. Economies that still have "low hanging fruit" need more engineers than developed economies.

Also, STEM graduates perform great from a wage and employment perspective right after college versus business and humanity grads (espcially during recessions). But the long term picture is more ambigulous.

Yeah I've been performing so great - two years of unemployment forcing me to go back to school.

On average/ "in the aggregate"

I do not think that heavy subsidization of anything works. I would introduce programs for clever STEM students from poor backgrounds, though. Quite a lot of modern inventors and scientists came from modest households.

The entire "like China" is a nice strawman to beat, though.

The core of the problem is that having too many lawyers and too many chicano studies majors is a net drag on the society, because they are non-productive and tend to actually champion more regulation and more laws to make themselves some work.

Some amount of humanities is necessary, but the current situation in the USA is wildly off-balance.

"Most importantly, graduates in the arts, psychology and journalism are less likely to create the kinds of innovations that drive economic growth ... there is little justification for subsidizing sociology, dance and English majors."

Not true.

(And I apologize if someone has pointed this out already. 279 comments is a lot of comments.)

It's PRECISELY those graduates in the arts and psychology (and anthropology and sociology, though maybe not journalism or dance) who drive innovation and economic growth.

STEM majors may develop the code for iOS, but graphic designers and human factors teams -- many are psychology majors -- determine how it looks. Science fiction writers, many of whom were English majors, give inspiration to science fact. Creative writing majors help craft story lines for video games. Microsoft Research has at least one trained anthropologist on its research staff. Google hires designers to improve their products.

Innovation and economic growth requires an awful lot of these non-STEM skills. Rather than undervalue them, we should look for ways to educate non-STEM students in math, science and technology disciplines and vice-versa.

Many non-STEM students CAN'T DO THE MATH. In addition almost every STEM student takes humanities classes stop trying to further water down our curriculum to make us "More well rounded"

I wouldn't say it's watering down the curriculum, in all honesty they already pack so much into a lot of these STEM programs - is taking MORE engineering or math courses instead of humanities going to make a difference? They should get rid of the electives though because it makes the program longer then necessary but these schools want their $$$.
And I don't know if the non-STEM students can't do math, some people just don't want to do STEM programs. First year Engineering ESPECIALLY is ridiculously boring, I mean not just a heavy load - all the classes just suck, Calculus, Linear Algebra, Engineering Statics - these are all extremely monotonous and bland courses. It gets better later but the first year material is dull enough to turn even a lot of talented people away.

"graphic designers and human factors teams — many are psychology majors — determine how it looks."

As a software developer, I can say: a good graphic designer is a true treasure, but most credentialed graphic designers aren't good. If you school more of them, you only get more mediocre, unproductive graphic designers.

This is the common problem with all skills that need a strong innate talent. If you took the entire population into an actor school, how many excellent actors would you get from it? Well, some, but not that much, really.

A large part of the decline in traditional chemical engineering has been from poaching from biomedical or biochemical engineering. There are good arguments to be made that a lot of modern biotech has more to do with chemical engineering than with biology.

Thanks for speaking out on this topic, Prof. Tabarrok. This is one of the most ruinous long-term problems in our stagnating economy. It's especially encouraging to see these comments written by someone within the college industry.

Two observations as one who has done undergrad in the 90s and grad in the 2000s:

1. "STEM" major faculties (science, technology, engineering, math) DON'T WANT anymore undergrad students--or, if they could get away with it--any undergrad students. Cuz if you have students, you have to TEACH them. What a bother! They just want an endless supply of grad student slaves to do their research and fill out grant proposals. So, they on purpose generally have the worst teachers on the hardest material. It takes a lot of a student to decide to stick with hard material taught by people that openly hate them...that's why the biz degree or whatever always starts to look very nice after even a year...

2. Tech degrees are often woefully out of date. Mech. Engineers still study linkages and mechanical timers in classes. Comp. Sci will still teach Pascal or C on obsolete API's. Students know this...and often migrate to something non-tech as the major, but then keep up as a "super-hobbyist" on the side...often returning to tech. as a job field with their self-made skills. But, just think how many more could have the skill set of programming it was made into a "core" along with some other tech material...for some kind of new, relevant "Applied Tech" degree?

I said this earlier. The skill that you learn as a Mechanical Engineer is how to think critically and solve problems. Yes, it sucks while you are doing it, but you need to understand how a linkage works. What do you think is between your car and the road?

1. There's a good case to be made that student loans are not the only subsidy distorting the university system. Of course not every university is like that - quite a few pride themselves on teaching, but students looking for the school with the biggest name instead of the best learning experience will tend to overlook them.

2. There could just as easily be a generic Engineering degree with particular concentrations that differentiate after the second year. Engineering is a skill more so than a set of knowledge; university level students should be practicing the skill instead of learning a knowledge base that will inevitably be obsolete within the next decade no matter how current it is at the time of graduation. In that context it doesn't really matter if the students are using Pascal or C. Let the employers and trade schools handle the tools training.

There is nothing obsolete about C, it is still a language of choice for system-level things like writing drivers, AND it helps the students with "getting" how memory really works.

You DO NOT want a Java monkey coder who never saw a pointer or an allocation, and his way of thinking is "oh, memory, it simply IS somewhere there, and it is plentiful", anywhere near a production code.

Such unhappy person will inevitably produce a 500 MB - large hashtable for a modest task that actually required 50 kB of memory, choking your server dead, but, hey, they didn't waste their time with obsolete C, right?

I am no friend of Pascal, though. Too sheltered.

Quote "Who will solve the problem of antibiotic resistance?"
The problem is NOT lack of qualifed people; there are lots and lots of people, at least here inside Route 128 in Boston
The problem is that when pharma companies do an ROI study, how much they will get back for each dollar invested in a new drug, drugs for neurological disorders and cancer come out way, way ahead of new antibiotics.
so, no investment dollars for new antibiotics.
Part of the problem is how new drugs are used: if you have a new, super $$ drug, like say Avastin, with a good PR campaign, you can get billions of revenue.
But with a new antibiotic, doctors use it sparingly, because eventually bugs will develop resistance, so doctors save the new antibiotics and use them only for bugs resistant to established drugs. After many years - close to patent expiration - most bugs will be resistant to old drugs, so doctors will finally begin prescribing your new drug in volume
Hardly a good biz model.

Considering the outlays necessary for a functioning microbiology lab versus those of a functioning English class, could it be that humanities majors are subsidizing the "more valuable" students?

As someone who has pursued a graduate degree in Biosciences from an Ivy League school, I can tell you that the country has WAY too many scientists. Of all the graduate students and post docs I know, only a small fraction of them (maybe 1/3 or 1/4) can actually find jobs in science. The rest end up in Management or Finance, or they go back to school for an MBA. There's a reason that there's more Physics PhDs working in Finance than in Physics, and its not simply a matter of income (though that helps).

I can't speak for Engineering, as that it seems that the demand for their skills is higher.

You can take heart: Mike Scherer at Harvard did a study of who were the most effective managers of high tech companies (in terms of growth and profitability): it was an executive with a science background. (Actually, lawyers did pretty well too.) Surprisingly, CEOs without a science background but with a finance background didn't do as well.

I find the chart very revealing. Are there any numbers on ethnic diversity distributions between the high paying degrees and the less so? I know many students of Asian background have good academic performance and suffer somewhat when diversity preferences for admissions are in effect. But, even so, are the Asians outdoing the other groups in obtaining the degrees that lead to the high paying jobs?

Longest/best comment thread ever?

Comments are so important to this blog, you should really devote a good amount of effort to maximizing the commenting/comment reading experience. The new nesting is holding up pretty well, but how about a +1 system that orders the comment threads by rating? That would prevent early comments from necessarily getting all the attention, plus it would be a great check on dubious facts.

Oof. This is a tough one. I agree whole-heartedly that STEM education is crucial. I have a BA rather than a BS but almost all my classes were in STEM or STEM-related tracks (including a year of advanced abstract math.) And certainly when I worked in tech it was painfully obvious just how important computer science is and how difficult it is to find people with the combination of education and ability.

That said, even in software engineering and biotech the bulk of graduates don't actually earn as much as many of the "creative" types who... hire them.

The issue, at least in America, is that for better or worse there's a huge market for entertainment. It's absolutely the case that engineering, math, chemistry, electronics, and other hard science backgrounds are critical to the entertainment industry. But it's also the case that the money generated by Pixar, Warner Bros., Bungee, ABC, Disney, ESPN, and even nominally "hard science" companies such as Exxon, Apple, Monsanto, Toyota is only peripherally due to their STEM employees -- a fact that's reflected in the relative status and compensation of employees in those firms.

I remember taking a Perl programming class with a guy with masters degrees in electrical engineering and math. He was doing graphics-engine programming for a startup composed of... high-school dropout gamers. And getting what amounted to entry-level wages while the marketers and designers were getting what amounted to dot-com dollars.

And for the record my business coach (not as bad an idea as you'd think for anyone with or without a STEM-related degree) makes shockingly more per year now than when he was a genuine, no-kidding, photos-to-prove-it rocket scientist in aerospace.

Point being that like too many other fields in the current market economy the rewards systems for STEM are negatively skewed compared to their underlying value to the economy. Yeah, you can cut back on "liberal arts" courses in, say, anthropology (George Lucas) or calligraphy (Steve Jobs) or economics (Sam Walton) but you're not going to alter the visible desirability of jobs in those fields.

I don't like it. I wish it were otherwise. Not sure what to do about it. Except possibly to rate universities, colleges within universities, and professors not on admissions but placements. Except, oh wait, the popular entertainment industry (Business Week, etc.) seems to have exactly zero interest in ranking schools in those dimensions.


p.s. Natalie Portman, an entirely credible, published STEM graduate, wouldn't earn in a lifetime in STEM what she made in her average movie role in Hollywood.

Agreed, more or less. But I think there's a tendency for people with X interest to think that there should be more people like them in society. I think there should be more social scientists with computer skills, and vice versa. But will having more engineers really increase the rate of "progress"? As you point out, even companies like Apple are 80% humanities and social science, 20% engineering--the innovations were in design and usability. So I'm of two minds about the whole situation. STEM are great, and certainly having more STEM graduates wouldn't hurt. But I hate to see the "soft sciences" and humanities denigrated in a world where most of the challenge is learning to use what we already have. Entertainment--that is, making life enjoyable--is as vexing a problem as building a marginally better fuel extraction technique.

The soft sciences aren't really being denigrated (by me). You can't reliably expect to be a movie star (or a creative business owner who can hire STEM graduates). And Portman's performance in movies is probably not worse due to her science and math training. If you are creative, do you need creativity training or would you benefit more from STEM training? It is primarily about making sure you receive some intrinsic value from your education investment.

For every Natalie Portman success story, there are a thousand wannabe actors waiting tables to get by. As you note, students can be attracted to the tiny minority of success stories. But those are not the norm... and if sold to students as the norm, that's doing them a disservice.

You should go into your college education with your eyes wide open. And that means with a realistic assessment of the change in your career earnings prospects as a result of what you do with your four years. Yes, in working hard and teaching yourself programming, you can recover from getting an Art History degree. But you'll still be behind where you'd be if you had gotten the computer science degree to begin with.

Engineering is a pretty decent path to a secure and dependable income... if you're good at it and willing to work hard. The real issue brought up by this article, though, is that "everyone has to go to college" is a myth that serves only the colleges' bottom line.

Underlying assumption is that the only purpose of post-secondary education is to provide entry into job market. This aspect has definitely been oversold. Randall Collins explained the phenomenon of credential inflation nearly 30 years ago in Credential Society: An Historical Sociology of Education and Stratification. But education is also about developing one's fullest personal potential. About becoming an educated, cultured citizen. In capitalist society, it may be the job market that has been oversold: how did corporate American manage to get taxpayers to pay for the training of their workers? Public taxes spent to support the training corporations don't have to pay for, but realize profits from? Nice deal! Now students take on enormous debt to pay for an education so they can compete for opportunity to work for (relatively) low wages due to overproduction of workers paid for by the state and the victims (I mean students) themselves. Brilliant! But some of us want to actually learn stuff and know stuff and not just get a piece of paper that may or may not get us a job. It's thinking that univeristy and college is about job preparation that's wrong headed.

Is there any way to look at default rates by concentration?

I live in India. I wish someone can write about the useless degrees offered by higher education institutions in India at a huge cost to tax payers. Tabbarok and MR readers will find the scenario in India inane.
Even the teachers in departments with single digit or even no student enrollment get huge salaries on par with teachers of the medical school. Any suggestion to shut down non-marketable programmes are vehemently opposed by the very same teachers who advice their own kids to study only the more job-oriented programmes.
I once did an informal survey of 500 teachers of basic sciences and humanities to find out how many would like their children to study those subjects. Answer: 2. A friend found that in 5 universities she surveyed, not a single physics, chemistry, botany, biochemistry or math professor wanted their kids to study the subjects they teach. But they want the tax payers to pay fat salaries for teaching subjects they know are worthless in the job market and do not want their kids to touch with a barge pole.
The only reason there are some enrollments in the zero job potential subjects is due to a massive state subsidy for colleges and universities; in the state of Tamilnadu in south India , the government colleges do not even collect fees, except in so-called "self-financing courses". I found that invariably these are the courses for which there is demand in the market. If the students pay the full cost for all courses, many basic sciences and humanities departments in state funded institutions will vanish in 10 minutes and the money saved can be used more efficiently.

Yet the government of a poor country liberally throws scarce resources for funding courses few are willing to pay for but study just to say they are doing something rather than nothing. A government of a poor country spending on departments just so some students can tell their families and friends that they are busy doing something, however worthless that something is, is plain criminal. I wonder if such an insane situation exists in any other country

Even if we had more CS majors, would our system still make sense? From a certain perspective, yes--going to school is clearly worth the 100K+ for most graduates. But, if you apply Coase, the morality becomes a little more unclear. Just because the value of an education is worth, say, 200K, doesn't say anything about how it should be split up. Should 100K go to the schools, 40K to the lenders, and 60K to the students? The student has clearly gained from going to school, but it seems to me that the educational system could work just fine with significantly less student loan debt. And I am not even mentioning the occasional student who is unsuccessful, where the debt might have a significant deforming effect on his life. If we use financial terminology, a system with high student debt is a highly leveraged system. And now that returns on college have declined, suddenly leveraged is becoming "overleveraged" for a good number of students. Whether you blame students is immaterial to whether the system could be improved.

The problem with student loan debt is the rising cost of college, not the potential payoffs. You can argue all you want about which major students should go into--it's besides the point. Personally, I don't think it matters much--I knew someone who was interested in consulting from the get-go, majored in Southeast Asian area studies, and got a job at Bain. I was one of your hated Sociology majors, and ended up getting data analysis jobs that were preferable to those econ/math majors who got sucked into the dismal world of finance.

"Most importantly, graduates in the arts, psychology and journalism are less likely to create the kinds of innovations that drive economic growth. Economic growth is not a magic totem to which all else must bow, but it is one of the main reasons we subsidize higher education."

But I thought as libertarians we are not supposed to question the preferences of individuals in the market. The above quote implies that there are instances when preferences can be undesirable, and others know better than me what kind of preferences I should have.

The interviewer was correct in the context he probably intended it. Lookup time in a hash table is constant (amortized O(1)). Further, if you have a fixed set of data (i.e. read-only), you can compute a perfect hash function that has a lookup time of strictly O(1). See gperf for an example of an implementation of this. Perhaps you were thinking about computing the hash value of a data sequence, rather than the lookup time once a hash has been computed. Computing a hash value for a given sequence of data is linear (O(n)) with the size of the data being n. However, the time complexity of this is generally not interesting, since any lookup algorithm requires an O(n) iteration over the blob at some point. I was so befuddled by his claim I didn’t bother arguing. There are a lot of people who, having been taught something is true, cannot comprehend the possibility that it is not.

perhaps the students have realised the truth about, at least American, culture.
People who deal in or about other people and are at the top of the pile.
People who deal with "stuff" are peons.

As much as I hate to admit it, this post has hit the nail right on the head. Throughout high school, all students are pushed to go to college, since we all know that with a degree, your earnings will automatically increase. Right? Well, there may be some truth to that statement, but it is more misleading than it is helpful. I am by no means stating that college is useless and that one may get by without higher education. However, I will say that your salary ultimately depends on the job market. A college graduate will undoubtedly have an edge over a non-college graduate, but what is the point when there are no jobs to be had in the said field?
I was an avid student of music in high school; highly involved in both band and chorus. Both of my teachers urged me to major in music education, for they felt I had a gift for teaching others music. However, the practical side of me won out. I knew that a degree in the arts wouldn’t guarantee a job after graduation. I am a creature of habit, and find comfort in ultimate stability. Therefore I decided that I wanted to pursue a career as a paralegal. I am a Criminal Justice major in hopes that when I graduate, if I cannot find a job as a paralegal straight away, I will have opportunities in other positions.
I feel that college is not about doing what you love. Granted I love music, but music education just did not seem practical for me. Luckily for me, I love learning and reading about what goes on in our criminal justice/legal system, so Criminal Justice is another perfect match for me. I think many college students abandon this practicality and find themselves in a terrible financial situation; crushing college debts and no money flowing in to pay them off. As the article stated, there is nothing wrong with pursuing a career in the arts. But as far as job security goes, you’re better off studying mathematics and/or sciences.

Practice what you preach. If you had taken one math class since you were 10 you would know to label your axes.

The bigger issue is why more students don't want to study these things. I blame high school approaches to science and math. They kill creativity, kill the fun, kill any interest. The fields people want to study are chosen because those fields tend to feed their creative side. Science and math do this just as much - it just isn't taught that way.

At my school the engineering majors are extremely competitive and spots are limited. Is this true elsewhere? Are there enough open spots in other programs to support a hypothetical increase in STEM interest?

Can society support this? Government isn't investing in science, research, construction, anything that STEM workers do. When Bush had to cut SOMETHING, ANYTHING, to "finance the war" he chose science programs. And now we're planning to cut another $120 million from the NSF?

There are bigger issues than what we decide to major in. Make these majors more interesting and more appealing and they will fill up. You can't force a person to pick a profession - you make him want it.

You've missed the main concept here. The people who are upset are the ones who are sold on the idea that getting their MBA will get them a job in the MBA widget factory where they can assemble MBA widgets all day. Then they get pissed off because they don't get a job offer for tons of money that requires no thinking on their part. The arts majors on the other hand go in knowing they're going to have to put together a 21st century style job "outside their field" and thus frequently end up more successful.

Successful computer scientists are the ones who can synthesize language, design, and constant learning. Unsuccesful computer scientists are the ones who don't like computers but here there are high paid jobs.

"College has been oversold. It has been oversold to students who end up dropping out or graduating with degrees that don’t help them very much in the job market. It also has been oversold to the taxpayers, who foot the bill for these subsidies."

Non sequitur. Humanities degrees have been oversold, whereas STEM graduates are still in high demand.

Can't help noticing that the sectors with the highest graduation rates, arts, psychology and communication, are the hardest on the list to outsource. Maybe they're not what America needs, but a PR person and a psychologist is not likely to have their job moved to asia. Adverse seclection at work!

Signed, a well-compensated commercial artist with a visual arts degree.

My God, you're a moron. If you've ever studied a university budget, you would know that colleges of liberal arts heavily subsidize the more fixed-capital intensive STEM departments. A sociology major costs far less to "produce" than a computer science major.

A couple of other people have mentioned this. This is an important point. The humanities and social sciences, and definitely the law schools, are helping to fund the expensive STEM fields and medical schools.

First of all, I agree with Alex that there are social benefits to education. But, I also agree that our current system of loan subsidies subsidizes both the consumption good and the investment good aspects of our higher education system. In that vein, I'd like to throw an idea out into the ether that I haven't heard discussed: grade-contingent course-specific subsidies, independent of major (or full-time status).

Basically, if you can get a solid grade in calculus, statistics, an introductory CS/engineering/hard science course, or a (well-designed) principles of microeconomics course, then the government will pay your tuition for that course (preferably the average tuition across schools). If you can get a high grade, then the government will pay you a small sum (the grades would be scaled; see below). Once the course is over, you can go along and major in whatever you want, do whatever you want, etc. Perhaps the benefit could be made contingent on attending a seminar on the labor market for quantitative majors.

Outlays for a program like this wouldn't be tiny, but they'd still be manageable, since the criteria could be adjusted over time. The impact is fairly immediate, so it helps to address the issue of hyperbolic discounting, both in signing up for the course, and in completing it. The total benefit could easily be capped at one or two courses to prevent some students from simply taking every intro course at taxpayer expense. Create some guidelines for course material to ensure consistency, but make the program opt-in for institutions, so that schools who don't wish to teach a standardized curriculum don't have to participate. I suspect that most will, as parents will demand access to the subsidy.

To limit the ability of institutions to inflate grades, my suggestion would be to administer a non-binding standardized test of the material (perhaps modeled on the AP exams?). The results of this test would be used only to determine statistically the fraction of students at each school who meet the required threshold. However, the actual grades assigned by professors would determine who receives the subsidy. This mitigates the incentive for schools to dumb down their curriculum, but it does not actually take the job of assessment away from the professor in any way (you have to get academics to agree to this thing, after all).

This type of policy might increase the number of math/science majors only marginally, but I think that it would provide other benefits. A more numerically-literate populace is better prepared to deal with a wide range of modern issues. However, since the courses to be subsidized are only indirectly vocational, it would be less inefficient than the current types of government vocational training, which are often outdated and which focus primarily on credentialing. Finally, universities can't game this system very effectively to capture rents from the government subsidy, since introductory math/science courses are a small portion of their overall course offerings. Pass the subsidy along with a law to inhibit price discrimination in government-subsidized courses.

If designed well, I think that this would be a pretty reasonable policy. After all, even for most students that end up majoring in the humanities, I think that there's substantial signaling value from showing that they are capable of succeeding in a quantitative field. Every business major who learns just enough programming to write a basic VBA macro at work increases the PPF just that much, and a smaller number of students are induced to keep slogging through that weed-out material long enough to get past the learning curve and become quantitative majors.

There is nothing wrong with the arts, psychology and journalism, but graduates in these fields have lower wages and are less likely to find work in their fields than graduates in science and math. Moreover, more than half of all humanities graduates end up in jobs that don’t require college degrees and these graduates don’t get a big college bonus.

It should be pointed out that the data linked to at the end of this quote - which indeed shows a big disparity between those employed in degree-requiring fields and those not - is for 25-year-olds. To say this this describes the jobs graduates "end up in" is at best a serious warping of the relevant facts. There may well be data that gives a more realistic picture of the wage forecast for graduates in different majors, but it doesn't take a (no doubt highly paid) statistics or math major to see that this data is not that.

First of all, I agree with Alex that there are social benefits to education. But, I also agree that our current system of loan subsidies subsidizes both the consumption good and the investment good aspects of our higher education system. In that vein, I'd like to throw an idea out into the ether that I haven't heard discussed: grade-contingent course-specific subsidies, independent of major (or full-time status).

Basically, if you can get a solid grade in calculus, statistics, an introductory CS/engineering/hard science course, or a (well-designed) principles of microeconomics course, then the government will pay your tuition for that course (preferably the average tuition across schools). If you can get a high grade, then the government will pay you a small sum (the grades would be scaled; see below). Once the course is over, you can go along and major in whatever you want, do whatever you want, etc. Perhaps the benefit could be made contingent on attending a seminar on the labor market for quantitative majors.

Outlays for a program like this wouldn't be tiny, but they'd still be manageable, since the criteria could be adjusted over time. The impact is fairly immediate, so it helps to address the issue of hyperbolic discounting, both in signing up for the course, and in completing it. The total benefit could easily be capped at one or two courses to prevent some students from simply taking every intro course at taxpayer expense. Create some guidelines for course material to ensure consistency, but make the program opt-in for institutions, so that schools who don't wish to teach a standardized curriculum don't have to participate. I suspect that most will, as parents will demand access to the subsidy.

To limit the ability of institutions to inflate grades, my suggestion would be to administer a non-binding standardized test of the material (perhaps modeled on the AP exams?). The results of this test would be used only to determine statistically the fraction of students at each school who meet the required threshold. However, the actual grades assigned by professors would determine who receives the subsidy. This mitigates the incentive for schools to dumb down their curriculum, but it does not actually take the job of assessment away from the professor in any way (you have to get academics to agree to this thing, after all).

This type of policy might increase the number of math/science majors only marginally, but I think that it would provide other benefits. A more numerically-literate populace is better prepared to deal with a wide range of modern issues. However, since the courses to be subsidized are only indirectly vocational, it would be less inefficient than the current types of government vocational training, which are often outdated and which focus primarily on credentialing. Finally, universities can't game this system very effectively to capture rents from the government subsidy, since introductory math/science courses are a small portion of their overall course offerings. Pass the subsidy along with a law to inhibit price discrimination in government-subsidized courses.

If designed well, I think that this would be a pretty reasonable policy. After all, even for most students that end up majoring in the humanities, I think that there's substantial signaling value from showing that they are capable of succeeding in a quantitative field. Every business major who learns just enough programming to write a basic VBA macro at work increases the PPF just that much, and a smaller number of students are induced to keep slogging through that weed-out material long enough to get past the learning curve and become quantitative majors.

The past 25 years have gotten us where we are today; socially, medically, technologically, etc. so instead of worrying about subsidies for the fields that have traditionally kept the world moving forward, why don't we focus on what benefits we could be gaining by supporting the arts to meet the demands? Clearly I am not saying that we do not need continual advancement and support in the STEM fields and that the flatlining numbers aren't worrisome on a global scale, but, there are a growing number of artists and alternative fields that are crossing over into those of the STEM.....maybe it is time to shift our thinking about these traditional institutions and what they really mean for the success of a nation, a society, a world.....

A hundred years ago, we created an education system designed to mold individuals into cogs in the industrial machinery -- the current woeful state of education reflects that ideology. The new ideology, reflected in this pitiful blog post, is that education exists to make the US "economically competitive" through "technological innovation." This ideology is just as bad as the last one. Education is not job training, it is creating human beings who can think critically, express themselves clearly, and think creatively. Did you notice the last one?: "Think creatively"? That's sort of important to innovation, wouldn't you think? Also, I ask whether you can go 10 minutes in our culture without encountering the arts -- billboards along the side of the road, radio in your car, TV and movies on your TV set, dance in your videos. In other words, this is a growth field. All you corporate-minded drones need to keep your hands off of education and let people learn.

I have a physics phd. I graduated with a STEM undergrad degree, I got a STEM phd, and I work for an insurance company.

My boss has a history undergrad degree, my team have a english, communications, visual design and one CS degree. My story is not unique- why don't we graduate more science degrees? There aren't many science jobs.

If you are going to end up working for an insurance company anyway, why get a science degree? Why not work WAY less hard and get a communications degree?

As this is an economist's blog, perhaps some thought should be given to "market failure" in science education, especially in a country where the majority of higher education is in the private sector.

Possibly, as a national culture moves upward in Maslow's 'hierarchy of needs' in parallel with economic security (or complacency), peoples' decisions are less dictated by practical considerations and more by affective ones. Decisions are probably made differently in a conformist, developing country such as China. At the same time, more young people in the US are managing to access college education, so why are the numbers only going up in humanities?

My thoughts are purely intuitive but, considering that science developed out of human curiosity, empirical observation and experimentation (quite similar to the sources of creativity), it is surprising that it is often taught in such an aprioristic manner - based on dictating and cramming information, getting right answers and scoring high on tests.

In high school, I got decent B's or B+'s in science and math, but A's in humanistic subjects. I was reasonably intrigued by math and science and, given that I attended one of the hardest schools in my urban area, I was probably good enough to pursue higher studies in those fields. However, I often experienced my science/math teachers as aloof, mechanistic, a bit contemptuous towards all but the A students, rather lacking in affect, and in one case sexist.

Doesn't seem like the kind of stuff that will inspire the passions of the average teenage customer! As I was getting better grades in humanities, it seemed that I would be more "successful" if I specialized in those subjects in college.

Furthermore, alongside increasing evidence that scientific inquiry and economic decision making are not as dictated by 'rational' criteria as once claimed - that affective factors tend to creep in despite the best intentions - I think there is also a case for drawing more connections between science and the arts/humanities in education.

My first job after college was in a cultural foundation, working in a program that placed artists on short residencies in local schools. On a site visit to one elementary school, the dancer in residence told me that she had been approached by a classroom teacher who asked "what are you doing in your dance classes? The kids from my class who are attending your course are suddenly learning math faster than those who aren't." It turned out that the dancer had been teaching the kids about how steps or movements are arranged into repeating and varying patterns. It seemed that they had somehow translated this experience into their math lessons.

Knowing what scientists know today about the plasticity of the human brain, especially in children and adolescents, yet teaching subjects as if wholly distinct and unrelated to one another, probably does more to damage the way that people choose studies and a career than any manner of funding or subsidy for one field as opposed to another, as does the short-termist focus on grades as opposed to the wider motivations and implications of learning - not just making money but curing disease, solving problems, etc., all of which should positively impact economies.

This may seem like an overly humanistic and qualitative way of looking at things, but perhaps a large number of American youth are voting in a similar way with their feet.

I saw the title of this post pop-up in my feed reader and thought I'd check out the post. I figured I'd have something new to add to the conversation. Then I saw the 479 comments that came before mine. Well so much for having something new add! Regardless, thanks for taking the unpopular view and writing about it. That happens too infrequently these days.

I agree that college has been oversold to students. However, if we're concerned that too few students take up innovation-fostering STEM degrees, why pick out psychology, visual/performing arts, and communications/journalism as particularly problematic? They each account for between 5-6% of all undergrad majors, combined about 17%.

Business alone makes up 22% of undergraduate majors. This is twice as many students as the next-largest group of majors in social sciences and history. Shouldn't we be encouraging more business students, who may already have some math savvy and a desire to innovate, to switch to STEM degrees? The same could be said of the overwhelming number of undergraduates from elite schools (30 to 40% in the case of Harvard, Princeton, and Yale) who go into finance and consulting. While these fields attract very bright and motivated people, and probably do make a positive economic impact, they seem less like economic engines than like the mechanics who work on the engines.

Further, I don't agree that there is "no justification" for having college subsidies extend to sociology, dance, and English majors. Certainly there is no economic justification. But we subsidize many activities that, like these, have little chance of producing economic gains for anyone, but enrich public life in other ways. Consider public libraries, museums, religious institutions, and non-profits, which all get subsidies, directly or indirectly, from the tax payer.

Instead of subsidizing some majors more than others, perhaps we should make a greater effort to be upfront with students about their prospects for paying off loans and getting a job, given their choice of major. Giving special subsidies to STEM majors seems to me to get the whole exercise backwards; in that system, those who cost universities the most to educate and who will reap the greatest economic benefits from their education will also be paying the least.

a teacher pointed something out to me that might be relevant; math & science is right brain stuff, but left brain teachers & administrators diagnose right brain kids as ADHD & drug them into submission...

There is an implicit assumption that higher education = better jobs. It turns out not to be the case for lot of people including STEM graduates, alpha, beta etc.

The cost of higher education has to come down, for all fields. The current high costs are not sustainable in the long term. Whoever is paying for that education (loans, parents, government etc) can’t keep doing it with a bad ROI.

Check what Sebastian Thrun is attempting at Stanford with AI-Class.

I was one of those kids that went to college only because I was offered a (music) scholarship. After finding out how hard college was and dropping out, I got a job at a supermarket and later transferred to the local public university where I took courses in whatever interested me. I had no idea about a major anymore and just wanted to be done. I realized that having started off in the Arts, I had taken zero pre-req "STEM" courses that would allow me to apply for a useful major and so finished, 270 credits later (needed 180) with loads of debt (even though I worked full-time to pay my way) and my BA in Liberal Studies, Arts & Letters, and Spanish (plus some fluff minors).

I may be lucky to land a job as a bilingual receptionist.

I've hung out for the last three years in the company of Indian computer programmers and MSE students. It is more than obvious to me now that having a little more direction from parents and counselors (like my Indian friends) would have helped prevent me from going to college for all the wrong reasons. In retrospect, I would have best attended a vocational/trade school right out of high school or else community college -all cheaper options that can come with government funding/loans (helpful for some). While gaining useful job experience I might have had more of a chance to think about what I actually want to be "when I grow up." Low-paid is definitely my answer now.

Do you think any High School counselors or teachers ever recommended vocational schools? Heck no. It is currently preached that University is the only way to go since "those who hold BA/BS degrees are paid more (quote)." This statement is true for people that know exactly what career to pursue and the subsequent educational path to follow to reach that goal. For the rest of us, college was a waste of crucial years in which to mature in other ways.

If I could do things over I would -maybe pursue a STEM degree? Be an Optometrist? The pay is much better than secretarial work. I've noticed that my math test scores (SAT, ACT, GRE, etc) have always been much higher than my verbal scores even though the last math class I took was in 10th grade. I must of is learned nuthing importent in my linguistics/history lit courses. crap.

Please label your Y-axis. Every axis should have a label. I had to read the text to understand what the heck the graphic was. "Hm, maybe it's, that makes no sense."

Meh. Having a degree in STEM is not a guarantee of a job either. I graduated w/ mathematics BS several years ago and never got a related job.

Nor does a STEM major mean you will "create jobs" in an engineering field. People who start successful businesses usually (a) have worked in that line of business before, and (b) are good at sales. I don't think a class in Fourier analysis or finite element method helps much with either.

Some comments:

You forgot to include business majors.

What I also find interesting, and a general feature of the 'We need more STEM grads!' punditry, is that you didn't link to a chart of the soaring starting pay for STEM grads.

Or showing that 50-year old STEM grads have low unemployment (and outside your field doesn't count) and high salaries.

Odd, for an economist.

I have not seen a single data point in favor of a shortage of STEM grads in the US for the past couple of decades, with the exception of 1997-2000. Of course, I don't consider STEM grad employers whining about not being able to get who they want for chicken feed as data.

Im a US student most likely going to Mexico to study Industrial and systems engineering because of hte lower cost.

I realized that mexican schools just dont offer all the BS liberal arts majors, therefore mexican students end up becoming engineers.

One argument is, back then the professions with higher growths today were underpaid, while the conventional degrees paid well since the beginning. Not getting into the depths of it, just want to point out that your interpretation of data may not hold true.

Comments for this post are closed