Econ Duel! Is Education Signaling or Skill Building?

Marginal Revolution University has a brand new feature, Econ Duel! Our first Econ Duel features Tyler and me debating the question, Is education more about signaling or skill building?


Signaling mainly, but one can still learn useful skills through education (reading, writing and math were pretty important), there's a huge gap between teaching only useful skills and what we have today.

Should everyone who says "signaling mainly" declare their major? I'd say half of any technical degree (and this includes more than STEM) is showing that you can, but the other half is learning skills you'll be using in week one. Basically the signalling vs skills argument is an interesting one, especially as a shocker for students, but you realize less is there as time goes on.

Field matters! I still find some things I learned in medical school to have daily utility. Not surprisingly my friends with degrees in engineering and agriculture often say the same.

School matters! You can get a perfectly good education at a Cal State college, but people want to go to UC. Today's question is not either/or. It's one way for some people, another way for others, and for most it's both at a ratio which varies individually and is difficult or impossible to quantify.

Those useful skills would have been learned on the job anyway. What's not discussed is the opportunity cost of higher education.

High school doesn't have much opportunity cost (children and teenagers aren't very productive) and what you learn in HS is indeed very useful (productivity enhancing).

College, not so much. What you learn is useless or would be learned anyway on the job, and the opportunity cost is huge (college students are adults, and very productive). But you do have to go to college to figure out what you want to do and show that you can do it (signaling). And quite often College Education isn't an investment in future productivity, it is instead a consumption good. It is even sold as such ("follow your passion", "do what you like", etc.).

I thought Tabarrok had a stronger case. It almost felt like Cowen played along to make a good video, but didn't really believe his own arguments.

I think Tabarrok has the stronger case also, even though TC is obviously our hero. I don't think the issue here is what technical skills you pick up at University - I have a STEM degree, but the good programmers have taught themselves to program well before college, and spend much of their class time messing about with abstract theory or functional programming bollocks.

There are theoretical concepts you learn in school which you're not really going to ever learn on the job, because the employer isn't going to waste time trying to get you up to speed if you don't know them. He's just going to assign you to a less technical job.

I'm a STEM grad, and although I don't use a lot of things on a daily basis, I at least have a firm grasp of the concept when I encounter them so I can get the gist of what's going on and perhaps dive in if needed. A person without the STEM background would be lost, and have to have someone painstakingly explain it to them just to begin participating.


How much of that conceptual knowledge did you learn in college compared to what you learned in your job? You have years of experience in your job. If someone fresh out of school were asked to do what you do today, they be lost too, right? How is that different? True, certain things that have been learned in college don't have to be learned again, so that may look like a gain. But does it take 4-5 years? Does it justify $100k? Does it justify 4 years of unearned income and unproduced output?

We probably could cram all the math and physics, or whatever concepts are needed later, in the first 2 years of college (like they do in France), and forget about "critical thinking" and other magical concepts. If you haven't acquired some critical thinking by the time you're 18, college probably won't help you.

My engineering company hires less than BA's for entry engineering positions from time to time. Their wages are cheaper, roughly 2/3rds the straight out of school engineer. Usually in a contract or provisional position. Most of them don't last long. They tend to be either, not as smart, or smart but lacking a solid work ethic.

Engineering graduates are, on average, a higher quality than the non-BA hires. We aren't hiring for specific knowledge, but we do need some one that has the brains and work ethic. Is that pure signalling? Or do you need something like College to transform the smart kid into the smart kid with a good work ethic. Or does College just sort out the kids with the good work ethic from those who don't have it? And even if that's all it does, is that still worth it?

Employers choose in-major graduates over harder-major candidates in most cases. That is, they choose not to train over a higher signal.

Signaling mainly.
B.S. Phys. Chemistry
PhD Theor. Physics
Trading equities in a hedge fund

All that education and you never learned that personal anecdotes are not data. Signalling indeed.

How on Earth are you getting that from that post?

Are you kidding? The guy listed his CV and said 'see it's signalling'.

Not a single thing that is listed on my transcript has ever been remotely useful. But I wouldn't have been able to pull nearly as many figures without that diploma.

Wall Street does use more signalling. They might be ahead of the curve, or they might be less skills based than a civil engineering firm.

No! Wall street just like folks with impressive credentials to signal to suckers that they know what they are doing. You can train a chimp to be an above average hedge fund trader.

This is somewhat true of i-bankers, though I think they still prefer econ majors. I think this may be because one of the most important criterions for hiring i-bankers is, do you really want to be an i-banker? Since it's a tough job and they don't want you to burn out in 3 months. Majoring in something other than econ may be interpreted as a signal that, no, I'm not sure I want to be an i-banker.

But there isn't much practical knowledge that you can get from school in this area, other than a basic fluency with financial statements. In fact, it may be perceived as a negative if you think you know things. Better that you're high IQ and simultaneously know you don't know anything.

On that note, I'm pretty sure Pshrnk doesn't know that he doesn't know anything on this topic.

I did a Ms.Sc, in Theoretical Physics as well. Statistics based on previous years showed that about a quarter of the students would end up in finance. So, no, not just an anecdote.

And this is a clear cut case of signaling. Pretty much nothing you learn in Quantum Field Theory can be used in finance. But knowing QFT signals your level of intelligence, dedication, conscientiousness, and ability to understand formal and abstract ideas and systems.

I agree college is more signalling than learning for most fields. But signalling (and sorting/weeding out) has definite value.

And that's the value to employers. College provides value to students in other ways, it's basically like adulthood with training wheels. For most it's the first time you live away from home, and have to learn how to pay bills and clean your clothes. It's also where kids go from high school-type dating to learning how to navigate adult relationships....being away from your parents is why so many lose their virginity in college for example. It's a soft glide path into that stuff vs being thrown into the 'real world' at age 18.

So before the "it's all signalling" crowd declares that college is a waste of money (because there are other ways to sort people) remember that has much more purpose than signalling and learning.

So your extremely quantitative education was no help at all in your extremely quantitative job?

When you learn to solve Maxwell's equations you aren't ONLY learning to solve Maxwell's equations.

If it's all signalling, why aren't all hard majors hired equally by Google? Why are CS majors hired more than math majors, and overwhelmingly more than med school graduates?

To academics it may feel like their education has value.
But the vast majority of folks do not remember much they learnt in school or college.

I would argue that social manipulation is the most important skill and that is what should be taught first and foremost in school.
Here is what you should learn in school:
1) Selling and seduction , Most people can get ahead in life on just this skill.
2) Defence against the dark arts- learning to manipulate, control and defend against manipulation by other people.
3) Getting along with family and friends. Close relationships need to be treated differently.

"1) Selling and seduction, Most people can get ahead in life on just this skill."
Get ahead of what? This basically assume a zero sum game.

Agree with #2

As to #3: In a world of multiple nuclear armed nations that attitude is an existential threat.

I hope my dentist does remember what she learned in school ;)

Sounds like a combination of zero sum attitudes with some heavy clannishness to boot. Probably not very good at the aggregated level if everyone is thinking this way.

However, I think everyone should learn enough about manipulation to defend themselves against it - presumably this would not only make manipulation as a strategy highly unprofitable, it would also serve as a rapid signal of people who just want to fleece you and who should be rapidly excluded from networks.

Here's my question,

if its just a signal, is it actually any good as a signal?

depending on what sort of college you're talking about...

if I handed you 120K, and four years of runway, could you come up with a better signal?

those sort of resources could fund a lot of projects, which might create a signal that distinguishes you from your competition far more than a typical college degree

if I handed you 120K, and four years of runway, could you come up with a better signal?

Replace half of the four years with mandatory internships. Students are required to rotate through government, nonprofits and business to get a sense of similarities and differences. At the end of each internship they have to write a paper or pass a test or whatever is the best way to measure.

Companies either get the interns for free or pay them below-market (either can be subsidized by tuition). In return they must provide detailed, honest feedback to both student and the institution.

The "signal" could be LinkedIn recommendations (or lack thereof).

As a first step, we should incentivize more universities to experiment with the 'co-op' model that is available at Northeastern, Drexel, and some Engineering schools. Although perhaps on a large scale this may case labor market distortions?

Isn't Peter Thiel offering something like this?

"if its just a signal, is it actually any good as a signal?"

This is important: in the Humanities, it is becoming a weaker signal. Once you get away from the elite schools, few of the things that TC claimed are learned in college--critical thinking, creativity--are actually learned there. There has even been a decline in emphasis on such basic skills as careful reading, clear writing, and rigorous argumentation, And with grade inflation, a college degree has even become a dubious proxy for IQ. As a result, the two important factors are the status of the school you attended and the major you pursued. Harvard only accepts high-IQ students, so it sends a stronger signal than Podunk U. Chemistry sends a clearer signal than women's studies.

One other thing. I tend to believe the signalling model in the Humanities and the human capital model in the sciences. An engineer or a chemist has to have learned concrete things about science in order to design a bridge or produce a new polymer. People outside STEM fields--even high IQ people with advanced degrees--don't know how to do science. Knowing how to do science is a form of human capital.

The sheepskin effect, the fact that such a large part of the education premium seems attributable to the last semester of college when one finally obtains the degree, seems the most difficult to explain under the human capital model. Why would most of the acculturation, education, learning, etc. be packed into that final semester? If anything, one might expect the most human capital to be acquired early in the program since learning curves are often steep at the beginning and level off near the end. For most undergrads and graduate students, it seems like there is a bigger change in the student, i.e., one gains more acculturation, between the start and end of the first year than between the start and end of the final year. Yet, the sheepskin effect seems to place the most value on that final year, consistent with the signalling effect. (Being able to finish a job is a strong signal.)

There may be significant human capital formation through education. Signalling, however, is what probably generates the education premium in the marketplace. The human capital gained could be much like natural ability: even if it is present, a signalling mechanism is needed to demonstrate its presence.

Certainly the sheepskin test captures a signalling effect. But it is also such an unusual condition that employers probably miss it. They are filtering on degree as an easy signal of accomplishment, yes. And all other resumes go in another pile, without anyone checking if the applicant is one semester short, or five.

So I think we have to back up. The employers might think the degree pile is the skills pile.

Finishing the degree SIGNALS conscientiousness and perservance!

"sheepskin effect"

And the manifest "Cap & Gown" (Academic Dress) signaling effect.

Astonishing that this ridiculous medieval clerical garb and theatrical posturing survived into the 21st Century ... and is still extremely popular across the education landscape.

Blatant signaling emphasizing a core lack of substance.

Or, y'know, traditions which make us feel good about our lives as we grow and achieve milestones. Equally astonishing that we put on tuxes and white dresses and exchange expensive rings when we get married.

Yes that's right. Graduation ceremonies and costuming are very similar to other social rituals like weddings and various religious rites.
They openly 'signal a social rite of passage' within the culture.
That just reinforces the view of diploma-value as having a strong component of social signalling.

Objectively, one would think the ostensible acquisition of a specific body of knowledge in college in the 21st Century does not require a formal costumed ceremony to note that fact.

Such school rituals (of which there are many) have been studied by socio-anthropologists, supporting the view that enforcing these rituals is a means to promote the desired overall social identity, particularly in a complex society.
The college graduation ritual signals the societal transition of a privileged minority into a special cultural niche of society with its own outlook. This effect is more pronounced at elite institutions.
Signalling indeed.

Putting aside your obvious disdain for signalling, whats your problem with ceremony and ritual?

After finishing my third year in a four year program, i started sending out lots of resumes (it helped that I took a year off to work in between - and actually, the company that hired me misunderstood that I had not completed the degree yet, and this almost led to me losing the job several months in, regardless of the very high quality output). I routinely tried to argue that many universities offered degrees for 3 year programs, and therefore i should be considered as qualified.

The very few number of responses I got from employers requiring ANY BA were basically to say that 3 years in a 4 year program was worth nothing, even though 3 years in a 3 year program satisfied their requirements. And none of them, not even one, seemed to illustrate any awareness of the absurdity of the situation. Anyways, I went back to finish my degree.

Someone leaving a four year course after three years is signalling something. Why wouldn't someone go back and finish the last year? I would assume an expulsion for cause. These days that might mean sexual assault. Although violent personal disagreements with the teaching staff are certainly more common. As is mental breakdown.

All in all, not really the sort of person most people would like to hire.

If a CV does not come with a really good reason why you have to leave the program early, there is no way anyone should hire you. Most people I know will not hire someone if there is a single missing year in their CV just on the off chance they were doing time. Leaving without a degree is a much bigger flag.

Why should you stay in school if you can get a job? Do you attack Einstein for not going to uni?

Sorry man, but that's a truly pathetic attempt to smear me.

I took the year off because, by extraordinary coincidence, 60% of the courses I wanted to take that year were cancelled due to profs taking the year off. But there are a million legitimate reasons to take a year off, including "I just felt like doing sometimes else", and also "none of your God damned business".

I am pretty sure Einstein spent years attending two universities in Zurich. But if someone is Einstein, they can do what they want. Most of us are not.

This wasn't meant to be personal. It is just a fact of life about hiring. Sure, if you want to take a year off, take a year off. But you can't expect people not to assume you have done so for a bad reason. People should go back and finish their degrees or have a very plausible explanation. Companies were not being unreasonable. They were protecting themselves.

I had not considered that, especially at the time. On reflection, it sounds like it may be a larger part of the story than I might have surmised.

I've noticed you speak very confidently of things you know little about.

Einstein not only went to University, he went all the way to getting (the early 1900s equivalent of) a PhD.

He only deviated slightly from today's standard academic career by not getting an academic job directly afterwards.

Like, isn't "I feel ready to enter the job market" not a pretty legitimate sounding reason to leave school early, if you can?

It is signaling a lack of common sense. saying "none of your god damn business" to a prospective employer is not a brilliant response.

There are a few reasonable reasons: financial reasons (although this signals a lack of scholarships), amazing opportunity (although the that had better be on the resume), and, uhm, emotional instability?? No wait that isn't a good one either.

The sensible approach to the "none of your God damned business" line of thinking is something like "I had personal matters to attend to that year", and the employer, if they are respectful and don't press the matter, are free to wonder whether this might have been attending to a dying family member, helping to raise a newborn child, or time off to deal with mental health. Anyways, my response was driven by the assumption that SMFS was tarring me as he often does, whereas it seems he was actually offering a legitimate explanation for why this might seem suspect. I assume he will be tempted to make much hay over me jumping to a conclusion, without considering that he routinely does this himself.

Anyways, I mostly think this reflects an old fashioned mindset. It is very common for people to take time away from school for all manner of legitimate reasons these days. It is increasingly uncommon to place negative judgment on those who stray from "the path".

Since no information may indeed be interpreted in the worst possible way (as you correctly imply), on my professional website I have a section which provides a detailed explanation of the business and career endeavours which explain the holes in the CV. Tried to start a some businesses (2/3 of which failed, perhaps roughly on par with Trump, but I did not lose billions in investor money and did not benefit from connections), studied languages overseas, and, as a translator, geared much travelling towards developing extensive on-the-ground knowledge of the fields I specialize in as a translator.

Indeed, as SMFS suggested, you'd better have some damn good answers to explain those gaps, or high quality employers will easily want to steer clear. I tend to consider "I wanted to take a year off and learn about the world" as a pretty rock solid explanation, and see selection as a two way street - i would not want to work for an employer that does not see this is a sound explanation.

Sorry, SMFS, that was an overreaction. Indeed, some people might think in the way you suggest. I'm much more a "given the benefit of the doubt" kind of guy than an "assume the worst" kind of guy unless you have a track record on the individual to drive your thinking (hence, I assumed the worst in this case).

I think the more obvious answer is that employers are so generally enthralled by the piece of paper that they are unable to evaluate 3 years in a four year program as the educational equivalent to 3 years in a 3 year program.

Courses in your major -- provided you pursue a career in the field -- clearly do have a real human capital component. Electrical engineers really don't forget their EE courses. But even there (and perhaps especially there), the signaling component is strong -- yes, employers do care that newbie engineers know things, but they probably care at least as much that they've demonstrated an ability to master difficult material, since those new engineers have a great deal more to learn before becoming truly productive. Other courses, outside the major, are pretty much pure, generic signalling -- can the candidate read and write effectively? Get along with others? Discipline himself do good work even when not particularly enthralled with the material?

I guess that is kind of a depressing conclusion for a professor who teaches courses to non-majors (yes, these students will forget everything I taught them. I'm just giving these students an opportunity to get a score on a mental obstacle course). But is there a better way? After all, a 'costly signal' has to be costly, doesn't it?

It seems to me that a "four year degree," just because it is a fixed value, has to be wrong.

Astonishing too that every single field of study has a signaling/skills mix that is magically resolved in four years.

"Astonishing too that every single field of study has a signaling/skills mix that is magically resolved in four years."

Extremely good observation.

“Astonishing too that every single field of study has a signaling/skills mix that is magically resolved in four years.”

Well, but they don't. For 'vocational' fields, a shorter technical course is sufficient. For others, a bachelor's is actually not enough signalling -- another year or two of graduate school is required.

"Astonishing too that every single field of study has a signaling/skills mix that is magically resolved in four years. "

There not. A lot of engineering degrees are effectively 5 year degrees now. Generally the schools get away with calling it a 4 year degree by upping the class requirements every semester so that you can finish it in 4 years, but you'll have to work at it harder than an engineering student from 40 years ago.

I looked at the Auburn historical class hour load in the 90's. At that point it was 16-18 hours per quarter. The same major (EE) had a 13-15 hour class load. They kept it to the 4 year mark by loading up the classes.

The students respond by, a) not having part time work, b) taking classes during the summer and/or c) graduating after the 4 spring.

They're not....

And it should be: The same major (EE) had a 13-15 hour class load in the 1950's.

Of course, there has been a significant amount of technological development in the interim, which explains a lot of the additional work, but Universities are being dishonest when they market it as a 4 year degree.

I did my first year in engineering in 1999 (Waterloo). We had 40 class hours per week, and they told us with a very straight face that we were expected to put in 2 hours of our own time for each hour of class. The 4 year program was explicitly divided into 5 years - after the first two terms, it alternated between 4 months coop and 4 months class time. I left when I found out that the elective courses were actually choices between two similar engineering courses - i wanted to study some language too. Discovering that all the really interesting sounding courses were basically a cover for more calculus, more calculus and more calculus didn't help either.

The sample curriculums you see assume that the student doesn't have any AP credit. A lot of them do.

That doesn't undermine the point--it just shifts the timeframe forward instead of backward in expanding it from 4 to 5 years

A bunch of dumb answers. Yes, we fix other degrees at fixed time lengths, in addition to the four year span. Still, the four year span stands as a benchmark achievement.

And the other spans are curiously fixed as well. It is AA and BA because 2 and 4 are natural choices?

"Courses in your major — provided you pursue a career in the field — clearly do have a real human capital component. Electrical engineers really don’t forget their EE courses. - See more at:"

EE here. The things I learned in college that I use even infrequently were obtained by sophomore year. The rest (the majority, actually) is just signaling. Some EE utilize what was learned after that. Maybe even most do; I don't know. But there were alternative majors I could have obtained which would be more suited to what I do now (still EE, just different) but employers don't recognize those majors as "what they need".

Even as a member of a STEM field I'm of the opinion it's primarily signaling. Can Candidate X handle all that a four-year STEM degree throw at them? Yes? First hurtle cleared. The rest they can learn as they go along.

Courses in your major — provided you pursue a career in the field — clearly do have a real human capital component. Electrical engineers really don’t forget their EE courses.

Yes but if you could get equally intelligent and diligent people could they learn on the job what they need to know while getting paid to do work?

Are there two more likable professors than Cowen and Tabarrok. My view of education is greatly influenced by what I do: I'm a lawyer. The surprise to the young law student is that law school isn't about memorizing lots of laws, it's about "thinking like a lawyer". To me, that means learning how to solve problems. That may sound banal, but for the law student the most difficult challenge is identifying the problem (or legal issue); if you can't identify the problem, how can you solve it. Indeed, once the problem is identified, solving it is relatively easy. Of course, problem solving is what everyone does, lawyers and everyone else, in life as well as at work. Education enhances problem solving skills. Obviously, an engineer's problem solving skills are based, in large part, on technical skills, but only partly so. Ask any engineer. Tabarrok is right in that some people are born with better problem solving skills than others; indeed, some have no problem solving skills, and no amount of education will change that. Cowen is right in that education teaches critical thinking skills and creativity; in other words, problem solving skills. In a world with rapidly changing technology, people must constantly enhance and update their knowledge. Without problem solving skills, the task is hopeless.

Rayward: "That may sound banal" (to lern to think like a lawyer/liar). It *is* banal. Has it ever occurred to you why the best students in law school are humanities majors, not engineers and scientists like me? I flunked out. Signaling and persistence, not logic, is what drives law school. You can learn pretty much all the law in a BARBRI bar exam outline, or in the Nutshell series, in 3 months rather than three years. But that won't land you a top notch job with the DC government unless you went to Harvard...or Howard. Signaling my man, signaling, get the message?

The idea that lawyers solve problems also fails to pass the giggle test. Sure some of them do. But many/most litigators knowingly create many problems as they compete for the side that hired them. (And to pad their hours)

The best law firms recruit and hire the best law students, which means those with the best critical thinking (i.e., problem solving) skills. Sure, law students can get by and even make good grades simply memorizing black letter law via study aids (hornbooks), but their inadequacy will be exposed very quickly once in the real world. When I work on a new project I am greatly relieved when my adversary is smarter than I am; there's nothing worse than spending hours wasting time floundering about on frivolous issues because my adversary doesn't know the difference between what's important and what's not. That's not to say that a very good lawyer doesn't sometimes use that tactic (floundering about on frivolous issues) when she knows that she has the upper hand and her client can better afford all those additional fees as the lawyers flounder about. I suppose it's that lawyer that you might identify as a "liar". It's true that litigation can sometimes be a contest of endurance. In transactions, fortunately not so much.

Yes transactions and litigation are very different. Doing poor work with transactions leads to litigation and will not keep your firms clients happy.

"Has it ever occurred to you why the best students in law school are humanities majors, not engineers and scientists like me?"

This is absolutely wrong. The best students in law school are universally engineers. If you can succeed in engineering and also are good at lawyer-type skills, you probably have quite a high IQ

In my time in law school (long ago), history and English majors had a distinct advantage the first year because of their communication skills, but not so in the second and third years when the analytical skills of engineering and math majors excelled.

@Cliff -what Rayward said. Humanity majors are terrors in law school. Not just the first year either, but every year. They sit in front of the class and take notes like their lives depend on it. They know the fallback of flunking out of law school is becoming an English teacher, whereas for me, it's a nice career in tech.

I think a good response to Cowen's point about signalling's effect diminishing over time is that good signalling allows them to procure opportunities that signal better in the future. Harvard grads get the prestigious internship, get the prestigious finance job, and that finance job signals for a new, better job. As long as none of the individual jobs are very long, there's reason to believe no one would notice, because getting even more accurate signals for competence (work-sample tests, for instance) are expensive, and appear to have lower marginal value the more highly credentialed the applicant for the job is.

+1 to the Hickman. Indeed, for lawyers, it's been said the reason Harvard law graduates "always win in court" (or more so than say Howard Law graduates) is that the Harvard law graduates are defense attorneys for high-paying corporations and clients, and these guys usually win (money talks in winning lawsuits, and generally the defense wins about two-thirds of the time). So it's not a steel-trap legal beagle mind that the Harvard law grad has, but the better paying clients to the lesser school lawyers (but not lesser schooled lawyers, if you know what I mean, as largely everybody learns the same law, and btw typically not enough to pass the state bar exam, which requires a 'prep course' post graduation, which I always found amusing--what's the point of expensive law school if you can't learn enough in three years to pass the bar exam, but instead have to take a prep course? The point: signaling).

The answer to Cowen is that the signal is good. Harvard grads really are better, they don't just signal that they are better. They are not better because of what they learned at Harvard, but they graduated from Harvard because they were better (smarter, more conscientious, etc.)

It is signaling, but that doesn't mean it is useless. The signal is accurate.

This is largely non-responsive, because he's looking at similar people who do and do not go to Harvard. Identical in every way before going, yet the Harvard grad has sustained improved earnings. If it was only an _accurate_ signal, then over time the non-Harvard student would gather enough signals to show the same level of excellence. But instead, the gap remains, decades later. I guess in the video they don't speak at length about the studies and models that are used to demonstrate this, but this is my understanding as well.

Great video. While there are successfull and less unsuccessfull countries that invest a lot in higher education, there are also countries that are successfull but don't push higher education much, Switzerland being the example I have in mind. Graduation rate is only about 30%.

Agree completely that international comparisons provide interesting examples and counterexamples. Putting Human Development Index rankings ( ) beside percent of population with college degree rankings ( ) one sees Canada has a much hgher percent of college graduates (54%) than the US (44%) yet trails the US in HDI ranking. Australia and Norway on the other hand, with only 42% each, stand at 1 and 2 atop the HDI rankings versus the US at #8. Australia seems to have found the best balance of public policy trade offs of any nation, really. The US would do well to just copy them.

I don't usually comment on the blog but I wanted to tell that I enjoyed this format very much and would enjoy if you brought different economists to talk about different topics (and perhaps longer debates).

Interesting and fun but ultimately we need to see the scientific evidence testing the models -it didn't really get a mention here.

My scorecard:

* IQ Test. I think Alex easily won this. College is about more than IQ, it's hard. As an underachiever who skipped about half his classes and rarely studied and barely squeaked by, I can relate to this. I'm not good at external motivation and it's been a challenge for me in the workforce too. (See also: it's 9:00 and I'm talking politics online). My question though: what about small liberal arts colleges that provide smaller classes and more structure? Perhaps they are essentially ways that the rich can cheat the system and give the conscientiousness signal?

* Social context. I think Tyler is correct here. Ironically, present day colleges are partly failing here because they provide students too much autonomy and intellectual freedom. But they do teach the openness skills needed when you're working with people of different races or religions.

* Cancelling class. I think Alex is right but for the wrong reason. It's time preference. Few if any people naturally favor their future selves. It's about those who will begrudgingly and with great difficulty favor their future self vs. those who can't.

* Natural experiment. I think I'll have to pass on this. I thought the idea that earnings went up in Sweden after the law was changed to require more education is a strong point. But perhaps it means that graduates give a stronger signal?

Nature/Nurture; Signaling/Skill building: Really guys? It depends on the school, the field, family background, IQ etc.

I'm going to take some graduate courses in Sweden to find a young girlfriend like Ray Lopez!

No country gives everyone private jets. Pshrnk simply has to be in the top 0.01% and he'll have no problem in Sweden.

+1! Lol. I am an unreliable narrator however. Not for my expat sexploits, all true, nor my 1% status from my parents, nor anything else, but that I flunked out of law school. I did not, I just say that for effect, sorry. In fact I became, inter alia, a litigator and retired before age 50. BTW in the beauty pageants I notice Miss Sweden always has a dark tan these days (signaling she's into hip-hop / urban culture) and anyways the contestants that win are never from Sweden (too bad, as I think blondes are kind of cute, though I never had success with them) but brown girls like Miss Venezuela, Miss Angola, Miss Philippines (twice so far! Megan Young and Pia Wurtzbach, both mixed race, my theory again, the former born in Alexandria, VA of all places). I think the judges are doing that to show diversity, beauty has no color, etc. This year I liked Miss Australia (blonde) and Miss Japan (mixed) in the Miss Universe pageants (does Trump run that? Googling ....yep. That guy is lol funny).

Interesting. As I near my retirement, I've found myself becoming a data scientist, my job turning almost entirely to the skills I learned in graduate school both as a student and a research assistant. Now, do I remember all the details after nearly 40 years? No, nor would all of them matter given the radical changes in technology. New technologies I can learn.

What I remember, what I'm grateful for being taught, is now to frame a problem, how to pick the techniques to address it and the over all process for solving it. Most of the time I simply do what I think I should, only to discover my "muscle memory" has taken me down the path I find outlined in research papers, books or class presentations/notes I discover while looking for ways to improve what was done. I can't believe I'm that different from most people other than the timing of my particular circumstances.

Signaling got me my first job all right, but now and for most of my career, it's been what I learned in graduate school that helped me keep it.

it's a mixing of all three of the benefits Alex talks about at the start... but I agree with Tyler most strongly, the biggest economic benefit of university is picking up unspoken social clues on how to act and behave... and the biggest personal advantage is being given time to engage with reading material and professors or other students who have a deep knowledge of a subject area... in other words, it's simply good to think and talk...

Education through the 4th grade. signaling there after.

BTW my grand parents went to school for only one year, but they ran a Barber shop and seemed educated. They read and learned from radio.
Of course they did like FDR so maybe they could have used a little more schooling in economics but they did not believe in many of these baseless myths that college grads today fall for like the anti-vaccination folks.
One of their expressions was to say in Italian that someone was badly educated or showing bad education. I think the idea of education as expressed in that expression is what is important.
Now they did send their son to Brown University.

Consider fortunate children who attend nice day care schools versus unfortunate children who suffer neglect and abuse.

The day care is considered "education", but the benefit isn't skills or signaling, it's having a happy living environment and social circle.

Doesn't matter because these days education is free it is signals (degrees) that cost money.

"Learn creativity"? It's well documented that creativity in children is destroyed by formal education and is greatly diminished by 3rd grade where the student is conditioned to the approved answer.

This is one of the best explanations of the value of college I've seen. However, it is not evident over all students anymore:

<The idea is, of course, that men are successful because they have gone to college. No idea was ever more absurd. No man is successful because he has managed to pass a certain number of courses and has received a sheepskin which tells the world in Latin, that neither the world nor the graduate can read, that he has successfully completed the work required. If the man is successful, it is because he has the qualities for success in him; the college "education" has merely, speaking in terms' of horticulture, forced those qualities and given him certain intellectual tools with which to work-tools which he could have got without going to college, but not nearly so quickly. So far as anything practical is concerned, a college is simply an intellectual hothouse. For four years the mind of the undergraduate is put "under glass," and a very warm and constant sunshine is poured down upon it. The result is, of course, that his mind blooms earlier than it would in the much cooler intellectual atmosphere of the business world.

"A man learns more about business in the first six months after his graduation than he does in his whole four years of college. But-and here is the "practical" result of his college work-he learns far more in those six months than if he had not gone to college. He has been trained to learn, and that, to all intents and purposes, is all the training he has received. To say that he has been trained to think is to say essentially that he has been trained to learn, but remember that it is impossible to teach a man to think. The power to think must be inherently his. All that the teacher can do is help him learn to order his thoughts-such as they are. "

Marks, Percy, "Under Glass", Scribner's Magazine Vol 73, 1923, p 47

BTW the debate should be about schooling. I am a big believer that school is mostly signaling but education is very valuable. School is not the only place to get education. Education is everywhere.

Usage: Education, properly a drawing forth, implies not so
much the communication of knowledge as the discipline
of the intellect, the establishment of the principles,
and the regulation of the heart. Instruction is that
part of education which furnishes the mind with
knowledge. Teaching is the same, being simply more
familiar. It is also applied to practice; as, teaching
to speak a language; teaching a dog to do tricks.
Training is a department of education in which the
chief element is exercise or practice for the purpose
of imparting facility in any physical or mental
operation. Breeding commonly relates to the manners
and outward conduct.
[1913 Webster]

And I believe it what Churchill who said that he never let his schooling interfere with his education.

If all some people get is the ability to react properly to the following instruction, then it's probably worth it: "Here's a problem. Go research many plausible answers, evaluate them, and come back with the best three."

Of course, many people who have not studied in university think "research" is something like inventing the next improvement in an engine or developing an entirely new product. They do not understand that, most of the time, in a complex world full of insane amounts of information, most "research" is just wading through mounds of often contradictory informational outputs and trying to figure out which informational outputs are most useful to apply/modify to the specific context at hand.

If all you remember from uni is a few of the correct keywords at the right time, you might save months or years of research just by being able to rapidly hone in on the most applicable or good information (and in so doing, avoid a lot of low grade trial and error that might otherwise occur).

"In Africa they invested a lot in education and ..." they haven't done that well.

a) the education is of very low quality compared to other regions

b) educational advancement is not very high. Very few people continue on to higher education (often not even high school graduation).

c) it's a competitive world. If Africa were as educated as it is today, but 100 years ago, it could compete with a lot of other economies. However, as much as they have advanced, others have also advanced, and so they still struggle to compete. If this is not obvious, consider the analogy for capital - someone in Africa today could bring out loom technologies which underlaid the explosive industrial development of Japan - however, those technologies are so completely outcompeted these days, that even, say, adopting the 90 year old technology instead of the 100 year old technology would be similarly useless in a competitive world.

So, what is Africa to do? Giving up on education seems dumb, but economic returns are not as obvious at their present level of development compared to the payoff when the very same levels of skill and technology were at the cutting edge of a competitive world (100 years ago). It is not a good position to be in, and legitimizes potentially troublesome views that high levels of protectionism are required. While I'm sympathetic to the infant industries sort of argument, this could only possibly work if the world were to not retaliate and allow for lopsided trade deals where African products are tariff free while Africa continues to charge high tariffs (indeed, this is the present reality, and internal barriers to trade are often pointed to as one of the biggest impediments, after poor infrastructure and perhaps corruption, in Africa).

it’s a competitive world. If Africa were as educated as it is today, but 100 years ago, it could compete with a lot of other economies.

Economies do not really compete like that. If all the countries of the world but the one that you live in produce more you will still live better.

Africa benefits from $100 smartphones, but cannot compete in textiles or much of agroindustry, traditional routes into industrialization.

"educational advancement is not very high."

The point is that many poor undeveloped nations have successfully increased education levels of average citizens by every measure and have not seen coresponding benefits to basic standard of living.

"it’s a competitive world."

Standard of living is not zero-sum. Even without being a world leader in science and technology, a poor nation should be able to achieve basic improvements in standard of living.

"So, what is Africa to do?"

The residents of many African nations want to move en masse to Europe. Instead, invite Europe to Africa.

Cater to foreign investment, foreign outsourcing, and high skill immigration.

Perhaps somewhat unrelated, but on the matter of attracting high skilled immigration to Africa, this is difficult because elites and upper class see this as a threat to their livelihood.

As a translator, I compete head on with highly educated African elites. During my time on the continent, I found, surprisingly, that lower class people tended to response very very positively to the idea that foreign skills were being applied to their problematics (translation has long been a very highly respected field in Africa, in many traditions being equivalent to spokesman for the king/chief in foreign relations), but upper class people despised the fact that I was "stealing work from qualified Africans", no matter how much I protested that many clients would strongly prefer to contract an African for the job if they could find the same quality-price combination in a black man/woman.

In so very many ways, the protectionist and nepotistic impulses of African elites are a really big problem for the continent.

Also, while you can hire a house full of maids and servants at a very affordable price relative to income of highly qualified/experienced people, it is even MORE expensive to attract highly skilled immigration, perhaps not even really much related to racism or poorly founded security concerns, but for the fact that an awful lot of amenities that are taken for granted even by the Western middle class are simply for practical purposes unavailable in much of Africa. (most capital cities have a couple shopping malls which serve this upper crust, however). How much more would you have to be paid to live in a city where none of your favourite foods are available and where highly skilled contractors are virtually impossible to find?

"They don't remember what they learned a year ago."

Maybe not for a exam formulation sort of question. But when the right context comes up, even if they cannot precisely define the concept, they are loosely aware of them, and are often able to appropriately internalize the lessons even if they cannot pick quite the right words to elucidate the origins of their thinking. For example, elasicitiy is mentioned. Presumably the average person who studies as far as second year economics cannot properly define elasticity one year later. But, when encountering the news that oil prices are rising and there is a debate about how it will affect consumer spending, the person will easily identify the correct argument that people will not cut gasoline spending by enough to counter the price rise, and this will pinch other areas of spending. Without being able to properly define elasticity, this reflexive ability to internalize the main idea could prove useful in many types of non-optimized (rationally irrational) cost management or decision making processes. And, that's just a single example.

I studied loads of soil sciences in undergrad. Honestly, I forget most of it. I couldn't tell you much at all without heavy use of reference materials. But, in skimming a report on effects of climate change or a particular farm management strategy on soil, within a few paragraphs or just a quick look at a couple figures or equations, I can easily identify it as credible or bunk.

I think quite a lot of knowledge is that way. You don't need perfect recall or even be able to really explain something to be able to readily apply the knowledge.

Nice format, nice debate. Hope you keep these coming. As Tyler pointed out, it would be the easiest thing for employers to use aptitude and intelligence testing but then along came Griggs v. Duke Power Co. Possibly the Supreme Court decision with the most dire negative unintended consequences for the populations it purported to defend. Leaves us with the cockamamie college coercion signalling as substitute for testing. Now it is merely a public choice problem: the higher education industry will fight to the death to preserve indentured college loan servitude.

As a Political Science major who went on to attend a highly-ranked law school (where the classes were mostly theoretical), I am very much in the signaling camp.

In retrospect, the most useful thing I learned in school was Algebra I. Every few weeks I'll need to solve a single variable equation to make sense of some data. I never use anything else I learned in school.

Skill Building or signaling? Neither!

How about dating privileges and membership into a social caste?

Can you imagine traditional college or high school replaced by some online skill and credentialing system?

Education is the dominant social, cultural institution of western civilization.

Love this post, btw.

Looking at the photos alone, they signal that the participants think that Tabarrock has the better argument: easy smile, relaxed posture. Cowen is frowning with arms in defensive position, folded across chest.

Agree. Alex won on points. Tyler was struggling.

Enjoyed the video -- do more of these!

+1, yes more short videos on easily digestible issues.

Presumably the signal has some non-random content, otherwise why would anyone use it?

So it's not ALL signalling. It's more like the primary purpose is skill building, but a lot of people who fail at learning skills still use it to signal the presence of skills. Employers are looking for skills, and they can be fooled into thinking skills are there by sending the appropriate educational signals. Kind of like a pufferfish, or a mockingbird. It's an evolved strategy to gain economic advantages by mimicking the appearance of a person with skills.

When are we going to have education experts debating the value of economics?

My answer is the most boring: A combination of all three. What weights? Signalling = 50%, prior abilities=%30, TC's magical learning of critical and creative thinking=%20.

Psychologist Albert Bandura pioneered and validated a concept know as "self-efficacy" - an individual's belief in their own ability to achieve goals. College grads develop a higher degree of self-efficacy than non-college grads and approach challenges with more perseverance and confidence and are more likely to succeed. So college grads "learn" how to succeed not by the content of a degree, but by learning how to set and attain goals.

On the other hand, having worked in a corporate environment for 30 years, I have seen the clubbishness that elite college grads afford each other (more opportunities, more connections). Signaling.

Should have been a rap battle.

+1, like the epic Keynes versus Hayek!

Good debate.

It seems both Tyler and Alex agree that there's a significant component beyond and outside the discipline itself, as Tyler points out "critical thinking, creativity, and social context." It seems there's subtle agreement that these things are at least as important. I had to agree when Alex pointed out that I probably don't remember much of what I learned in many of the classes I took in college. In work I certainly didn't require much discipline specific knowledge, and my employer didn't seem overly concerned about specific things learned in college.

I'm left wondering is there a more efficient method of instilling critical thinking, creativity, and social context? In hindsight, while there was skill building, much of what I did in college seemed more like a hurdle.

btw - studied econ w/ background in math and science. At my employer (insurance) my work included financial planning/forecasting, risk management, product management, competitive position and market analysis, and a short stint in marketing. Probably my most valuable tools walking in on day 1 were familiarity with spreadsheets/computers/modeling - I remember they asked me questions in the interview in this area and I discussed some econometrics models I'd worked on, and we also discussed some things I was modeling at the horse track.

Once I was in the door though, it was all about proving my ability to do good work in areas that were pretty specific to my employer and were not generally transferable. As I learned the business opportunities continually opened from there.

One of my eventual bosses told me years laterthat in the interview I told him words to effect of "No matter what you need, I'll get it done." I don't remember saying it, but that was the impression that stuck with him. He said that was as important as anything in my getting the job.

The undergraduate university programs are not training programs for specific jobs. They are exposure to some systems of knowledge, like C programming or accounting. They don't really train you for specific jobs or even careers. A CS degree does not train you to be a DBA or a network engineer. It gives you some exposure to relevant information, but that's it. Similarly, an MBA does not train you to be a strong mutual fund manager. It's more about exposure to basic concepts, the degree of relevance of which varies. I work in IT and only a very small fraction of what I learned in college is relevant to my job.

I think another debate should be about how efficient the university process is at either signaling or relevant skill/knowledge development.

Count me in as saying it completely matters what kind of major you graduated with.

Engineering and the hard sciences are almost all about learning. The only 'signal' the school provides is some guarantee that you actually learned the things you need to do the job.

In the humanities, it's very different. Say you get a degree in Sociology and take a junior management position or a position in government that requires a degree (any degree). There can be no doubt here that the value of the degree is almost purely signal - the degree tells them that you meet some minimum level of intelligence, that you are capable of deferring pleasure in favour of work, that you can handle a certain amount of pressure, and that you have enough drive and focus to complete a four year education.

The thing is, the push to get everyone into college and to make it as easy as possible to do so diminishes the value of that signal. If everyone goes to college, the signalling value of a general arts degree goes down. If all the hardship of college attendance is removed, the degree doesn't signal your ability to deal with that either.

And, if you got a degree in gender studies or some other grievance field, the signalling value of that degree is negative. I would rather hire someone without any college than someone with one of those degrees, because someone who spent four years learning to be aggrieved is likely to be trouble, and also likely to not have learned anything of value in four years.

The value of the degree signal has gone up as regulations have made it harder to evaluate job applicants and harder to fire bad employees. That makes hiring a much riskier proposition, and so the value of a degree increases. Make it easier to test applicants for IQ and be able to ask them personal questions that help you understand what kind of person they are, and you won't need to rely on a sheepskin to know the person. Make it easier to fire them, and you make it easier to hire them. This would benefit the poor, people who do not have powerful connections or who couldn't afford to go to a good skill, new immigrants, and other disadvantaged people.

Comments for this post are closed