From the comments, Boonton on education

The context is the discussion of why Mafia members with college degrees earn more:

Signalling doesn’t really work IMO. Who is he signalling too? Other criminals? Customers? Why do they care? It seems if this is what it is the economy is deeply inefficient. 40% of the population needs 4 years of college to ‘signal’? So if there was some way to pick up this signal without college huge profits would await.

I suspect there’s two aspects that make college valuable:

1. Narrative creation – humans work by creating and sharing fictional narratives. College is a lot of practice at that which is a skill that carries over into business of many types.

2. Burns off immaturity. I suspect a big portion of the benefit of schooling is babysitting. It keeps kids out of the way of adults (which our economy couldn’t function otherwise…imagine if *every* day was take your kid to work day). By keeping immaturity somewhat walled off until kids grow out of it, schools prevent them from damaging their lives.

2.1 This may be somewhat related but workplaces are very, very stable. If you are changing tires at 18 there’s enough tires in the world that you can still be doing it at 59. Perhaps by starting work at a younger age, it is a bit too easy to fall into stability. School forces you to someday break things up. No matter how good you are at school you’re going to have to leave that stability upon graduation which will land you somewhere else which you’ll have to figure out. That flexibility may be more valuable than premature stability.


'Narrative creation' is good. College is a bit of a Credentialist swindle and a few years of College does train you in bluffing and 'gaming the system'
Buring off immaturity is also good. It certainly explains the Indian higher education system which features a lot of organised violence in an otherwise peaceful country. I suppose one could add 'mimetic effects'- typically you get to observe higher class behaviour in College without having to emulate it right away- but a seed has been sown and a few years down the line you find yourself turning into a 'toff'.

Conventional signalling theory isn't relevant because there is no asymmetry of information but what about incomplete contract theory? Would it be better to link your fortunes to a gangster who has signalled lower impulsiveness through higher education, though also possibly higher sociopathy and dissimulation- the danger here is he may leave you in the lurch as he goes up-market or turns legit- or should you stick with the illiterate local thug? Presumably, other things being equal, lower impulsiveness increases 'residuary control rights' and the associated rent.

"It certainly explains the Indian higher education system which features a lot of organised violence in an otherwise peaceful country. "

Can you explain? I'm not sure what organized violence you're referring to.

Student Unions in Indian Universities developed a tradition of battling each other with the encouragement and under the patronage of the main political parties. In the late Sixties and early Seventies there was some pretence that such violence- which featured bomb making and acid attacks on rivals- might have something to do with a Maoist insurrection. Since then such violence has moderated and is fully integrated into criminalised local politics revolving around caste or regional cliques.
A generation of such gangsters became senior Professors, Deans, Vice Chancellors etc. and in some States it was considered unusual for a Head of Dept or other senior University official not to be a 'history sheeter'- i.e. have a long criminal record featuring kidnapping, rape, extortion, murder etc.
Sadly the younger generation is more effete and prefers to battle on Twitter and Facebook rather than by the time honoured method of bashing each other's skulls with hockey sticks.
Recently some promising young activists- PhD candidates and such like- were thrashed by middle aged lawyers at Delhi High Court. Truly a sad day for Indian Higher Education.

+!, informative

Is this true? If so, pretty crazy.

If not, an odd but enjoyable ramble.

Indian Student violence has spread to China-
What puzzled the Chinese is that Indian gangsters resident amongst them- part of a large counterfeiting industry- are polite and eschew violence. Indian Medical students however show a misguided zeal to conduct autopsies on each other with such implements as come readily to hand.

Well damn. My college experience was boring.

Pretty exaggerated. To be taken with more than a pinch of salt.

I agree, except American higher ed seems to nurture immaturity rather than "burn it off." Looking the other way at underage binge drinking, grade inflation, safe spaces, etc.

Underage binge drinking: keeping 18-22 year old males passed out drunk for 4 years seems preferable to a lot of other things they could be doing.
Grade Inflation: better they spend 20 hours a week passed out drunk and get a 3.0 than duel each other or rape women on busses and get a 1.0. We can rescale after the fact as needed.
Safe spaces: like offices, courtrooms, the senate? This seems a useful thing to internalize.

Signalling and herd mentality of employers cannot be discounted. We are now seeing a shift where the bachelor's is starting to be seen as not "good enough" and graduate degrees are required for any kind of meaningful advancement

Even today, the share of each age cohort acquiring post-baccalaureate degrees is around 13%, and quite a mess of these have little intrinsic value (e.g. MEd degrees and MSW degrees).

Could you clarify what you mean by "intrinsic value" here?

Obviously the piece of paper itself is meaningless, but that's true of literally all degrees. OTOH, the market values the degrees you mentioned; certain jobs are unavailable without them, so the degree clearly has value to the job-seeker.

You could mean something like, "While the experiences one must undertake to achieve these degrees have value, the formalized structure around them provides no additional benefit." This is plausible to me, but I don't see why you couldn't apply that statement to every degree or training program.

See Thomas Sowell on the utility of teacher-training programs.

Obviously, they vary in their content. Those which actually teach pedagogical strategies which have some empirical validity may be good programs. The question is finding that odd treasure amongst the trash. In New York, you need that MEd. to keep your job, and they do not care if the program is dreck (LeMoyne College, I'm looking at you).

Again, look at the curriculum of social work programs and ask yourself what the student is preparing to do. What do social workers do which cannot be done by teams of professionals with better operational measures of competence? (E.g. sheriff's deputies, counselors and clinical psychologists, public health nurses, managers schooled in public and philanthropic administration, and child-care workers).

On a much smaller scale, you have programs in library administration, which have pathologies of their own (aside from preparing people for an occupation which is evaporating).

It's all just a constant delay of adulthood. The traditional nuclear family is not a valid goal for young people. A 22 year old man can't offer the modern college-educated woman anything other than to be a roommate with benefits/netflix. A lot of sitting around playing xbox waiting to turn 30 when maybe a woman will be interested in playing house with you.

Problematic is that male career value is tied strongly to income, whereas female career value is centered around "doing whatever you're passionate about". The result is that men aren't actually that useful because they're not substantially out-earning women, and women experience a lot of anxiety when they fall short of the dream job they think they're supposed to have curing cancer (which is objectively hard).

The traditional nuclear family is not a valid goal for young people.

What are you talking about? It's a perfectly valid goal and most people eventually put it together.

The result is that men aren’t actually that useful because they’re not substantially out-earning women,

Men do out-earn women, and they're dominant in performance-oriented occupations not addled by 'diversity' ideology.

"What are you talking about? It’s a perfectly valid goal and most people eventually put it together."

What I mean is that people don't get together out of college and do the nuclear family routine. Yes, I agree they do it eventually, as the new (urban) American dream is to go to college, durdle till your mid 30's, and then reluctantly have kids.

"Men do out-earn women, and they’re dominant in performance-oriented occupations not addled by ‘diversity’ ideology."

Yes there are still some "alpha" professions, but the gap is getting smaller all the time. Most young men aren't doing anything substantially different from what young women do on their own.

Most young men aren’t doing anything substantially different from what young women do on their own.

There are huge disparities in certain wage-earning and salaried occupations (building trades, engineering, elementary school teaching).. There are some where it's now fairly close (law, high school teaching).

What I mean is that people don’t get together out of college and do the nuclear family routine. Yes, I agree they do it eventually, as the new (urban) American dream is to go to college, durdle till your mid 30’s, and then reluctantly have kids.

The median age at which a woman bears her first child is 26 for whites and 23 for blacks. The median age at first marriage is now about 27, There was a disconcerting increase at age of 1st marriage between 2002 and 2008. Not sure what to make of that.

(1) is basically the standard comment about "communication skills".

"Learning how to think"

"Writing skills"

etc etc etc

Besides the fact that college is an absolute blast, I think a key corollary to the "narrative" point is that college is an ideal method to build and then maintain a network, which is an incredibly useful piece of personal capital.

1. The admissions process filters your cohort so you end up spending 4 years with people of approximately equal intelligence, ambition and curiosity. Not only does this mean you're more likely to get along with those people during the 4 years in college, but your classmates are more likely to end up in similar professions on the way out. Of my high school classmates, maybe 5% are even conceivably business partners for me. Of my college classmates, it is probably more like 85%. This is not to say that my college classmates are "better" people, just that since I do financial litigation, I am more likely to do business with my college classmates who tend to be bankers, traders, lawyers, run tech cos, etc. The effect is not linear because each banker friend is more likely to have another banker friend--each of whom is a potential client, or otherwise business partner.

2. Shared experiences. At the broadest level, I can generally talk to my classmates about which college cafeteria had the worst food; the best bars and happy hours; the most obnoxious professors, etc. At the narrowest level, I went through the same fraternity pledge process and lived with the same 30 people for three of four years. For the former, it makes conversation easy; for the latter, it creates loyalty and trust.

3. Shared culture. I did not go to Williams, Stanford or Duke, and each of those schools is obviously very different, but I'm fairly sure that if I met a graduate of any of them for a breakfast meeting, that we would get along. We probably read many of the same newspapers; know many of the same books; have a small-l liberal worldview. Even if we have our own preferences in suit cut, we both know how to dress for a meeting.

These factors are an ideal network-generating machine, and four years feels like about the right amount of time to do it. Sure, there are other ways to build a network, but all seem pretty inferior. People keep mentioning military service, which obviously makes life-long friendships thanks to the shared experiences. Military service is probably one of the most extreme shared experiences you can have. But there is not much of a filter--none on the enlisted end, and only minimal for the officer corps. You may leave with great friendships, but those friends are less likely to be helpful in providing financially for your family once you're out. Working for four years may provide some filter, but the total network size would likely be much smaller, and because the focus of those years will be some form of blue or white collar labor that is specific to that particular job, you're less likely to come out with a broadly useful cultural base. IE, you'll know all about Wal-Mart, but not about General Motors--its far more useful to have the common culture that can help you navigate the managerial class at either institution.

A four-year network-building exercise! Guess what, you don't have to live with people for four years to build a network. You just have to go out and meet people.

I suspect the 'networking' argument is humbug for all but a small sliver of private research universities and fancy private colleges and humbug for most fields which require disciplined preparation.

You'd be wrong.

The network I built/joined at one of the state Maritime Academies has been critical to my success. It combines in an extremely useful way with the technical skills I learned there.

I was actually going to add a disclaimer to my initial comment noting that my experience is limited to finance/top-tier litigation on the profession side, and ivy/liberal arts colleges on the education side, but 1) it sounded a bit humble-braggish; and 2) I think (recognizing the limits of my view) that the value of building a network really is more universal. Having a robust network outside your own organization may not necessary, and there may be certain professions where it is more helpful than others, but I think in the vast majority of cases, having a broad and deep network provides a huge boost to an individual's odds of success.

I'll acknowledge that when I was younger (I'm not THAT old), I also thought that networks were nonsense, and that you succeed solely through disciplined excellence in the technical aspects of your career, whether that means doing the best legal research and writing, the best fundamental financial analysis, or designing the best new mousetrap. But having seen how the world works, I am convinced that while excellent work product may be critical to success, having the ability to reap the most personal benefit from that work product is often driven by your network. That may mean finding funding for production of your mousetrap; having someone to brainstorm creative ideas for identifying targets of financial analysis; or meeting and being memorable to potential legal clients.

I love Maxine Hong Kingston's line: "the sweat of hard work is not to be displayed. It is much more graceful to appear favored by the gods." To some extent, I think that having a good network is what lets a person appear favored by the gods. Not to say you don't need to work hard! Just that maximizing the value of hard work is also important.

Also, to be clear "networking," in the sense of artificial industry events, going to alumni meet-ups, etc., is not synonymous with building a network. I would agree that a lot of "networking" events are boondoggles.

I don't say networks are nonsense. I think it's much more of a concern for those in professional-managerial occupations than others. The question is how beneficial is college selection (rather than building relationships at work or in trade associations) in building networks.

I have no doubt that there are people who need no help building a network, but my thesis is that many people--probably nearly all people--do better with the benefit of college. And the question is not whether you need four years of college to create your network, but rather what is the alternative use of those years. Some people may be ready to jump into a career earlier than 21/22 years old, but many--again I would suggest most--are not. Heaven knows I would not hire 18-year-old me as an associate at my law firm. So if that assumption is right, what is the more valuable way to spend those 4 years?

Good comments BDK

Mean time in tertiary schooling for those holding baccalaureate degrees is 5 years. About 30% of your program consists of fulfilling distribution requirements and some of the remainder is just meandering around the course catalog. This is atop several years of slap-dash secondary schooling which functions quite well at the task of ... employing people to work in secondary schools. You have a great deal of misdirected effort at all levels.

There's no biological reason people have to wait to age 23 to get their working life off the ground. One of my great-great grandfathers began earning a living at age 12 and 12 was the normal school-leaving age as late as the 1890s. It may not be optimal in this culture and society, but that's different.

A conservative reform might abolish the current degree architecture. Instead, for academics and the arts, have a series of 1, 2, 3, and 4 (academic) year degree programs in which your course of study consists strictly of one subject and cross listed courses which act as prerequisites. A school which offers the 4 year option would have to offer the 3, 2, and 1 year option; one which offers the 3 year option would have to offer the 2 year option, &c. On the vocational side, have an array of degree and certificate programs of a length adapted to the character of the discipline, with the calendar year (48 credit) degree the mode. Medicine and peri-medical occupations would follow a standard pattern: a preparatory certificate of varying length (up to 2 academic years), followed by professional school of varying length (up to 4 calendar years), followed by clinical work. Other professional schools (law, architecture, clinical psychology, public policy, planning) might also require a preparatory certificate, but nowhere near 4 academic years of study to earn that certificate. A research doctorate would consist of a dissertation atop a 4 year (academic) degree or atop the terminal degree in the vocational discipline in question. A professional doctorate would be a classroom program exceeding 105 credits or thereabouts. For most people, a 1 year degree in an academic subject followed by 2 years in occupational school might work passably.

I agree with the benefits, but how many people are going to be professionals? Even for law, you could have been a clerk, then paralegal, then lawyer in the time you got a law degree. The vast majority of people do not need college.

While I don't discount signalling so easily, I think both of these points are valid.

With regard to point 2, I would even surmise that several occupations (not to mention advanced degrees) extend this process beyond the bachelor's degree. Management consulting is a huge one, various teaching programs are also popular, corporate management trainee programs, certain finance roles, etc. They all mimic familiar processes and environments for college grads and give them a sense of continuity and stability. These options are very attractive to 22-year-olds who aren't mature enough at that point to handle risk, work environments where most colleagues are significantly older than you, self-learning, self-management of mistakes, commitment to niche industry knowledge/experience, and just good old-fashioned divorce from "brand names" (whether Yale or Bain).

Also Boonton on signalling, "If it’s about signaling then why doesn’t at least a few people ignore the college degree and hire non-college grads of equal or greater skill for less money thereby achieving superior profits? It only takes a tiny portion to be able to see and take advantage of such an opening in our bias to initially achieve economic rents and very soon convert over the entire economy."

Real brilliance, folks.

"In contract theory, signaling (or signalling: see American and British English differences) is the idea that one party (termed the agent) credibly conveys some information about itself to another party (the principal)."

Because signaling (to work) has to actually signal *something real.* In this case, college completion signals (a) a more mature discount rate; and (b) some combination of sufficiently mature work habits and native brilliance to get through college without ending up in jail or dead. Non-college grads contain, disproportionately, those lacking these two. The signaling argument is only statistically superior, not ironclad.

I think it's partly signalling, partly irrational. Libertarians see an invisible hand which will ruthlessly eliminate inefficiency. I see a light breeze, pushing one in the direction of efficiency, but a company or industry can resist it. If a company could be making 10 million in profit but is making 5 million because of some inefficient practice, it's still making money, and will have many advantages, know-how and connections and the like, that a start-up competitor will not. The employers don't want to offer wages lower than they think is "fair" for a given job. To give a non-college graduate lower wages to do the same job as a college graduate feels like exploitation.

"No matter how good you are at school you’re going to have to leave that stability upon graduation ": unless you decide to become an academic.

Colleges make people infantile leftist and against decent working. Kids starting the work directly better socialize to older adults. Switzerland uses learning by working to several "academic" occupations too. If studying is required, that could be done while working.

Before quoting Switzerland as example take into account: a) cantons (states) fund 75% of the vocational training program, the rest it's the Swiss Federal government and some business b) For example, it takes 3 years to get the Federal Certificate of Competence in retail sales. That formation allows the young individual to be a cashier in a supermarket or manage a cell phone shop.

Imagine the following scenario: Texas funds 75% of the state vocational training program to train supermarket cashiers for 3 years and all the states agree on the validity of Texas issued certificate.

In Switzerland occupational licenses (certificates of competence) are valid in the whole country. Next, vocational training is funded by states, state taxes used in a "Federal license". Finally, you have to be comfortable knowing vocational training for cashier, florist or cook takes 3 years. So, it's not possible to complain people can't work even as a cook if they don't have the 3 year formation and certificate of competence.

So, is it only colleges making people infantile and leftist or is there some cultural barriers/defficiencies in the US that get in the way of Swiss-like vocational training? I'd like to see state level politicians agreeing on using state taxes to fund federal occupation training and licensing. Also, I'd like to see blue collars being happy with the idea of 3 years of training to do "low-skill" jobs.

"For example, it takes 3 years to get the Federal Certificate of Competence in retail sales"

Holy crap. I'd say this bars ever using Switzerland as a good example of anything again.

On 2.1, manual workers also face incertitude at work. It's easy to be a productive brick-layer or carpenter at 20 years old. At 40, any work or hobby related injury will make worker productivity go down. Some older manual workers become managers of younger people, while some others are in pain and become ZMP workers. So, we blue or white collars face incertitude, we all face it. The difference may be in the expectation of incertitude. White collars expect to reinvent themselves in another career while blue collars go on strike to avoid changes.

It's amazing that business has been able to get away with requiring a key component of their operation, personnel, to qualify itself at its own expense. Except for Goodwill and some other thrift stores they must pay for the acquisition of raw materials, the development costs of physical plant and real estate, etc. In the case of management, engineers, and labor they expect to hire capable people whose education costs them nothing. What a deal! If corporations were serious about success they'd be scouring the secondary school system for the brightest prospects, hire them as apprentices and train them in their own schools. But, that would cost them money.

Businesses do reimburse individuals for qualifying themselves at their own expense. It takes the form of higher wages.

"In the case of management, engineers, and labor they expect to hire capable people whose education costs them nothing. "

That's just laughably wrong. My company pays a significant premium for talented engineers with a degree.

For those of you ignorant of the process, there are "engineers" without any education past high school. Generally, they've been working at a plant for a decade in maintenance or facilities engineering and have picked up a lot of hands on experience and various training classes.

I put engineers in quotes, but I don't mean that derogatorily. They are engineers, they just don't have a credited degree.

and there are "engineers" who have just graduated and never seen a manufacturing floor, etc. who need a lot of experience to become useful. Product design is another field like this.

As an employer, the US military trains its members itself. Enlistees are legally required to fulfill their contracts. Their compensation is not negotiated. You would probably say "Being a soldier is different. They must be trained by the military". However, there are a few educational institutions with military programs, Citadel and Norwich, for instance. While graduates of these schools would be easily accepted into the military, they wouldn't be exempted from its training. Also many positions in the military don't involve firing rifles or lobbing grenades and can make an effective transition to a civilian life possible. The military also members to universities for further education at no expense to the student. There's no reason why large corporations or industries couldn't operate in a similar manner. The current paradigm is simply a matter of custom.

If corporations were serious about success they’d be scouring the secondary school system for the brightest prospects, hire them as apprentices and train them in their own schools. But, that would cost them money.

It is also a basic principle of Anglo-American law that you can't force a person to work for you, no matter what a contract says. They can leave a day after the apprenticeship is over.

I think antebellum apprenticeships typically incorporated contracts of indenture (and the apprentice was a juvenile and a member of the master's household).

The closest thing you get to that is non-compete clauses in contracts. I've heard of those for veterinarians and physicians, but I think they're fairly unusual outside of licensed occupations.

Incorrect. They're common in a range of industries @ the executive level, e.g. retail, tech, management consulting.

why good SAT scores is not enough for signalling ? more extreme example is East Asia goukou exam which score is very important in your life. wouldn't it simply cheaper to use SAT or goukou score if the need are signaling only ?

No because, it is relatively easy a "brilliant but lazy" type (the kind of people who is largely useless in a real job) to have a good SAT score. But to have systematically good grades during several years, this requires more hard work, I think.

The 'brilliant but lazy' are a tiny segment of the workforce and can be weeded out by employers through the evaluation process. The notion that you need 4 or 5 years and $125,000 of higher education services to protect employers from these types is chuckleheaded.

I disagree.

Lazy is probably the incorrect term. However, it's not uncommon to get a smart but non-directed persona. Someone that can finish a 4 year STEM degree is someone that can do, not only hard things that they like doing, but also hard things that they don't like doing. Jobs often have plenty of crap that is boring, tedious or just not fun. You need someone that can plow through the crap to get to the two marshmallows on the other side.

Well, I would agree, that you don't actually necessarily need the expensive higher education for many jobs, but you do need some evidence that a person will work hard even on unpleasant tasks. What other evidence is reliable?

I think military service can be a proxy. But that's about it for indicators that are consistent and widespread.

If you have a problem employee, you have to fire them if they're not amenable to an improvement program. It's difficult to believe the deadweight loss from the occasional bad hire (and you will have them, no matter what the credentials of the applicant are) exceeds the cost of higher education as we know it.

" It’s difficult to believe the deadweight loss from the occasional bad hire (and you will have them, no matter what the credentials of the applicant are) exceeds the cost of higher education as we know it."

Sure, if 100% of the cost of education was purely from that I would agree. But STEM majors actually learn useful information in school.

You can learn 'useful' information in all but a few occupational programs and in some academic programs as well (though my guess is that physics is not the best use of your time). The most 'useful' academic degrees are likely economics and chemistry.

"You can learn ‘useful’ information in all but a few occupational programs and in some academic programs as well (though my guess is that physics is not the best use of your time). The most ‘useful’ academic degrees are likely economics and chemistry."

That's a great opinion, but plenty of firms are willing to spend their money on a different idea. It's possible all of those business people are wrong and you have discovered the right answer. It's also possible that you are wrong.

It’s also possible that you are wrong.

Wrong about what?

Sorry, I wasn't specific.

Wrong about the value of a STEM degree. And to be fair, I may be putting words in your mouth.

Currently the average Public in-state fees for a college degree are roughly $10K and out of state is $24K. The cost for a Private school average $34K per year. Paying $40-50K plus living expenses for a STEM degree is, in my opinion, worth it. Paying $125K plus living expenses for a general degree (business, liberal arts, etc) is extravagant and not worth it.

I'm not commenting on the value of STEM degrees, but of academic degrees v. occupational degrees. Science is academic, while technology, engineering, and medical disciplines are occupational.

Math and science degrees are more valuable among academic degrees, but some are more so than others. Biochemistry is valuable (unless the market's saturated). Astronomy, not so much. In social research, the quant degrees (economics, geography, and the quantitative wing of sociology) are where the value is.

A "brilliant but lazy" person can pretty easily get great grades in college. The key is to go somewhere in which their level of intelligence is still in the far right tail, and perhaps to stick to liberal arts or social sciences. Most of the other students will be really lazy too, and a little bit of knowledge combined with their reasoning ability will put them at the top of the class.

A “brilliant but lazy” person can pretty easily get great grades in college.

Not the schools I attended.

Definitely at the school I attended. Admittedly my only source is the anecdote of "Turkey Vulture, ages 18-22."

"A “brilliant but lazy” person can pretty easily get great grades in college."

There were a whole cadre of liberal arts students playing in the quad as I would trudge off to the Engineering Library every afternoon after my last class. I always referred to them as "hackey sack" majors.

Because it would be against the law.

Also the consistency argument.

But mostly due to the law.

About the signalling theory, I think we could have two different versions:

a) an "hard" version - people go to college to "signal" that they are intelligent, hard working, etc, and employers hire these people because they think that people with college degrees are more intelligent, hard working, etc.

b) a "soft" version - college graduates are more productive because they are more intelligent, hard working, etc, but economic agents don't really think about that; employers simply notice that college graduates are more productive (and, because that ,they hire more and pay more to college graduates) and students simply notice that college graduates earn more and have more job opportunities (and because that, they try to have a college degree, if they have the ability to that).

I think the "soft" version is largely immune to criticism "Signalling doesn’t really work IMO. Who is he signalling too? Other criminals? Customers? Why do they care?".

Maybe, just maybe, people learn things in school?

In my opinion economists have gotten too infatuated with the idea that schooling is nothing but signalling because it's an unintuitive idea that's makes you go "aha!" and feel smart. But then they end up having to make up all sorts of stories about market failures to square it with reality. How about Occam's razor?

No, you don't.

1. You have an arms race between families.

2. Employment discrimination law has effectively outlawed less time-consuming ways of sorting the labor market.

3. Pedagogical achievement milestones are antique and conventional, and are not influenced in their content by employers.

4. HR staff play CYA (in part because of the threat of litigation).

Points #2 and #4 are U.S.-centric and could be put to the test by looking at countries that don't have American-style discrimination law, of which there are many (perhaps Japan, for instance). From the link below, it does seem the U.S. has a higher-than-average wage premium but it is by no means an outlier:

The US is up there with largely post-communist Eastern European nations. Outlier....ehhh....Luxembourg and Austria and the UK are up there, but Canada and Australia are both way below us.

Burns off immaturity. I suspect a big portion of the benefit of schooling is babysitting. It keeps kids out of the way of adults (which our economy couldn’t function otherwise…imagine if *every* day was take your kid to work day). By keeping immaturity somewhat walled off until kids grow out of it, schools prevent them from damaging their lives.

Since youths between the ages of 17 and 25 are perfectly capable of working, higher education cannot function in this way. The notion that the way to 'burn off immaturity' is to forgo f/t year round work in favor of college doesn't pass the laugh test.

I took him to mean that the key benefit of college is that you exit it 4-5 years older than you enter it. Which is actually a keen insight.

It's not that college makes you more mature than if you spent that time working. Not by a long shot. You just get to let time pass while in a fairly isolated and safe place. (Regardless of CNN reports that colleges are basically just rape factories.)

For non-STEM fields, it's 90% babysitting, folks.

It's not babysitting. The youths would be better off if there were a housemaster around with real authority. The problem is not that it's babysitting, it's that its time ill-used. Most people do not have it in them to be intellectual hobbyists. Look at Gerald Ford or George Bush, both of whom hold degrees from Yale. Ford, what is unusual for a lawyer, betrayed no avocational intellectuality. Neither did Bush. Neither had intellectual deficits. It's just that for them, your noggin you use at work, while free time is for sports. Even when you get a student with such interests, their time is frittered away by the structure of college curricula (something Allan Bloom called attention to). About 1/2 the manpower deployed in academe is devoted to the academic arts and sciences or the visual and performing arts. These are liberal arts and not meant to be discretely practical (though chemistry and economics can be). It's just de trop. As for occupational instruction, a certain share of it is devoted to programs in teacher-training, social work, and library administration which are dubious in essence. And, of course, a number of academic and artistic disciplines have been suffering from a decadent period: literary criticism and history, art criticism and history, studio art, American history, and non-quantitative sociology to name a few ('women's studies' is a pseudo-discipline).

I suspect for STEM fields it is also 'babysitting' too. In the past it was not unusual for someone to learn chemistry or physics by being an apprentice at a lab. In the field of math, for example, I took calculus, statistics and econometrics yet accounting and many finance jobs rarely get more complicated than addition/subtraction/division/multiplication. I've built many complicated Excel spreadsheets but maybe once, on a lark, did a regression in one and never once used anything from calculus in one. I also remember doing a semester of Russian, which I did very well at and enjoyed but foolishly never followed up because of some silly scheduling conflict which caused me to decide not to do the next level.

I think school and college have multiple ends, one of which is babysitting. At the college level the 'babysitting' might be thought of better as allowing someone to experiment wildly before settling down to a more traditional career. Even STEM classes would fall into that category.

So here's where the elements come together. On the one hand there's some limited ability to indulge in immaturity in a somewhat safe environment. On the other hand there's an ability to dabble in different places. Often the dabbling might result in nothing important in terms of the rest of my life (i.e. the one Russian class), other times it might open up something really great.

I suspect for STEM fields it is also ‘babysitting’ too. In the past it was not unusual for someone to learn chemistry or physics by being an apprentice at a lab.

These are both academic subjects. People may have learned technological disciplines derived from physics or chemistry in a lab, but you're not going to learn the thing itself that way, and it's doubtful many people have done so in a century or more (Albert Einstein, Nicola Tesla, and Albert Michelson all had tertiary schooling, at a time when a secondary diploma was unusual). Trade apprenticeships may be what's necessary in certain fields, but they're a long haul.

> So if there was some way to pick up this signal without college huge profits would await.
Some *legal* way.

Near-universal college education has diluted the product. By that I mean that college today isn't nearly as rigorous as it was long ago (to be clear, before I attended college). And not only that, but what's considered "smart" today isn't all that smart. By that I mean today's "smart" young men and women are those who can navigate in a world dominated by media and celebrity, those who can take a small idea and reach millions by virtue of media. And lest we forget, college is a business, a big business, dependent on lots of customers; and success in the college business, like success in any business, is achieved by selling what's offered to the customer.

It's not universal. As we speak, about 43% of each age cohort are acquiring a BA degree. (A generation ago it was about 1/2 that).

The dilution is a consequence of a higher than average IQ being generally necessary in order to perform college level work. But more people now attend college than have the IQ to perform at what used to be regarded as college level work; the only way for this to be possible is for the colleges for the masses to dumb down.

I am quite sympathetic to the argument that college teaches non-academic skills and provides valuable experiences, but I would object to the idea that a highly costly signal is suspect.

The opportunity cost of a signal is one of the strongest ways for a signal to demonstrate robustness. If a pundit signals to me that he is certain the stock market will go down by saying such on CNN, I afford the signal some relevance as he risks reputation. If he buys puts on the market, I am more inclined to trust his signalling. If he publicly liquidates his asset and rolls it all into puts I will take his signal most seriously. None of his arguments have changed, but the cost of him being wrong is extremely high so I will have no doubts about there being mismatch between his signal and his actual beliefs.

Likewise, national security is often about the cost of the signal. It is one thing to say that aggression will be resisted at all costs. It is another to draft every able bodied male into the reserves, lay explosives into the mountain passes, and produce mass quantities of small arms. Spending money on "pointless" defenses that will nonetheless be overrun is an extremely strong signal that invasion will not be economically beneficial. This seems to be the thought behind a lot of Estonian and Latvian security arrangements - strongly signal to the US and Russia their intentions in part by spending copious amounts of time and resources.

Even in healthcare we have found that many patients will intentionally seek out the most expensive surgeons and treatments because cost is usually a good proxy for true quality.

For education and job markets, I believe it will be similar. Saying that you have the ability to navigate a bureaucracy, to follow upper-middle class norms, and that you will be at least semi-reliable is cheap and most alternatives to education (like say skills testing) are also cheap. If credentialing is cheap, then faking the credential is also going to be cheap. Requiring four years, massive amounts of debt, and the arbitrary rule following of college is very high cost, and I suspect that is why employers love it. Faking it is just too hard and you can trust the signal that you will get workers of the college type.

If higher education is mostly signalling, it will almost certainly be the cost that is allows it to be signalling and not the credentialing that allows it to be costly.

How about "the set that earned degrees were smarter to begin with, and that's why they eventually do better"? Or "the set that earned degrees had wealthier upbringings, and they eventually earn more because they *expect* to eventually achieve a certain standard of living; the non-degree set settles for less because they weren't raised up to expect more."

Just theorizing.

This. BTW, does the study correct for the rank of the previous generation? Michael Corelone not only went to college. He was the son of the Don. That was a lot more important than his education. In the real world, I would guess that high ranking mafiosos, like strivers of every sort, want to send their kids to college. And if their kids become involved in organized crime afterwards, they have the advantage of their family's status and relationships.

“No matter how good you are at school you’re going to have to leave that stability upon graduation “:"

Wouldn't the same effect then apply to military service, which generally starts at 18 and runs from 3 to 5 years.

1. and 2.1 either have zero impact on the return to education, or negative (assuming they are actually benefits). 2. is somewhat plausible, but I don't think it ultimately matters much, if it matters.

Why would schooling be any more practice at narrative creation than going to work, joining the military, or living in your mom's basement for 4 years? In each situation, you still need to construct a narrative about what you're doing and why you're doing it.

If anything, schooling provides an extremely easy, lazy narrative because it's what people are expected to do by the larger culture. "Oh I'm in college." Then you are on a pre-approved Narrative Track and don't need to explain much further. You're respectable. In that view, the impact of schooling on the ability to construct a personal narrative should be negative - it is just a continuation of childhood, K-12 becoming K-16 or K-JD. You aren't creating your own narrative, you are working off one society created for you.

Essentially no one who gets a job changing tires or anything else at 18 is going to be doing it at 59. Only a small minority will be doing it at 20. My brother is currently doing it at 30 and is already sure he doesn't want to be doing it at 31.

Again, given how quickly people change jobs, particularly in low-skill service or manual labor jobs (the type someone will have at 18), being an employee rather than a student at 18 will mean several job changes by age 22, as opposed to just "being in school" during the same period. So the "working" track will produce much less stability, and much more ability to cope with changing circumstances, than the "schooling" route. Moreover, tying it back to 1., it will take a lot more work to create a personal narrative out of your multiple low-skill jobs between 18-22 than creating a narrative out of your schooling.

School could be somewhat of a walled garden to be stupid without causing harm to your future earnings potential. But the impact would be quite small, and would certainly not be an actual "return to education" as it could be equally realized by putting kids in other environments (say the military).

I think this explanation is based in part off an assumption that everyone in "school" is living in a dorm and totally cut off from the larger world in which they could be stupid. That might be the experience of most people who comment here, but that isn't the experience of most people who get an "education," and hence of the people for whom a wage premium from college education is measured. From everything I recall (and what seems reasonable), commuter students make up the great majority of the total number of people receiving an "education" in any given year. They commute to their local community college or state school. They often work part-time. They are "walled off" for maybe 15 hours a week, but they are also exposed to the wider world in which their youthful idiocy can cause them trouble.

And I suspect there will be a negative correlation between being the type of person who does something stupid between 18-22 if you aren't in the fully walled garden of a residential campus life, and the type of person who is in that environment from 18-22. In other words, the type of person who ends up going to school full-time, lives on campus, and doesn't work, will tend to be less likely to do something stupid that ruins their future from 18-22 than someone who commutes to school and works part-time.

#1 There's a lot of narrative creation, destruction, navigation in academic settings. You demonstrate mastery of various theories, argue for them, argue against them and so on while at the same time measuring how you are doing with the feedback you get from others. Mom's basement doesn't really offer that. The military might....but does it make sense to send a majority of people through the military in an age where total war is rare and the wars we do fight tend to be light on manpower and high on tech?

#2 I think can be expanded's a walled garden to try out stupid as well as a bit of a sampling buffet. Take a class in Russian, take a chemistry, try this or that. Out of a 100, maybe for 60 the 'dabbling' is harmless and of just minor benefit. For the remainder it may open up a totally different area to go into that would have never come up had you not gone.

In terms of doing 'stupid things', this is a relative term here. It may not be so much doing stupid things but lowering the risk and cost of taking various chances. Taking risks is a very intelligent thing to do if the negative payoffs are low and unlikely. it's only stupid if they are costly and likely. Many blue collar type jobs have a stereotype of a somewhat conservative type of ideal worker (meaning personally conservative, not necessarily ideologically so). It's possible while that's quite a rational approach to take in the blue collar type of world (keep your neck down son, put in your 20 years and collect the pension!) college offers greater gains by making the opposite approach more rational for a time.

I think in both cases you're arguing for what happens to a fairly small right-tail of people who go to college, at best.

Most people aren't doing a whole lot of mastering different theories, arguing for them or against them. They're memorizing some stuff, getting good enough of a grade, and moving on. They could get more of that narrative-creating or theorizing experience by working in some low-skill job with a diverse assortment of characters, and then going to a new slightly better job and explaining to the interviewer why they should get hired even though they were only at the previous job for 6 months. Or they could just argue a lot in internet comments sections.

The same for taking academic chances. Most people aren't sampling from a buffet and seeing if they like it, taking chances, regularly switching majors and planned careers because of their self-exploration and risk-taking. Most have an idea of what they're going to do, and they follow that path. And college is a terrible venue for a lot of that potential risk-taking anyway, as it gives you no real sense of how life will be if I major in X vs. Y. Bouncing around jobs in your late teens or early 20s can be a way to experiment too, but you end up with a much better sense of what you actually like or don't like (you may discover that you hate all forms of manual labor, or customer service, or working at a desk, or working outside).

I don't think either type of experimentation is especially risky. There may be a societal preference for bouncing around courses and majors in college than in jobs from 18-22, so maybe an employer will judge someone who had 6 jobs from 18-22 more harshly than someone who had 6 majors. But I'm not so sure about that.

I think for some really smart or really hard-working or extroverted or [other good trait] people, college can potentially function as what you're describing here, and potentially increase their subsequent wage premiums as a result. But I think it has little to no impact on the vast majority of students.

I would say the creation of 'shared narratives' is pretty broad. No one is saying they have to create a new school of philosophy from first principles. Simply writing reports, reading and arguing different ideas is an exercise in narrative creation. This gives you skills in being an applied generalist. Figure out the rules of where you land then quickly start applying them to tasks that develop, if you're really advanced maybe even get to the point where you can tweak the rules themselves. Again this may be why you often see people with degrees that seem to have no relation to the job they work in (i.e. the CEO with a degree in art history or philosophy).

Outside of college, time on the clock counts for a lot in terms of career advancement. If you put 3 years into becoming a plumber, then decide to ditch that to instead run a dog grooming business, you are running a big risk. Plumbing builds in pay quite a bit as you get years of experience under your belt. Dog grooming business probably runs along the same pattern. Both jobs at the very entry levels though do not pay well. Also someone who suddenly leaves a job they spent a few years at is sometimes seen as suspect. A stability centered mindset would say get in as young as you can, 'pay your dues' then advance and reap the benefits.

Could someone flap around for a few years trying out lots of jobs and then discovering one that really fits them exceptionally well? Sure but I don't think the system is really set up to encourage that for jobs that don't require college.

"I don’t think either type of experimentation is especially risky. There may be a societal preference for bouncing around courses and majors in college than in jobs from 18-22, so maybe an employer will judge someone who had 6 jobs from 18-22 more harshly than someone who had 6 majors. But I’m not so sure about that."

But you don't have to have 6 majors. Even a single major will give you multiple opportunities to sample electives and since college usually require you to take some liberal arts courses, some science, some math etc. you will probably be required to take at least a few courses outside your comfort area even if you are a person whose just dead set straight on a single major from year 1 to year 4.

"I think in both cases you’re arguing for what happens to a fairly small right-tail of people who go to college, at best."

That might be. Someone gets to college figuring they are going to study business but then discovers they love Russian language studies and ten years later you find them working as a translator or doing international business. I suspect if you exclude the top 5%-10% of wage earners, you'll find the college premium will shrink dramatically but not entirely as a portion of it is made up of people who found something really different an unexpected in college and that altered their life path dramatically.

Alright, I think I am convinced that this has an impact (hard to say how much, but there would be instances in which it is large - if someone decides to go into engineering instead of social work, say), as long as we are just talking about a comparatively small portion of the total college-educated population.

It sounds like a version of a "matching" story: for some portion of people, college provides an environment in which they actually do find a better match given their underlying skills, potential, and interests, and thus find a better use of (and better way to maximize) their human capital. It also makes it easier for employers to find them and identify this value, leading to a larger wage premium.

"I suspect if you exclude the top 5%-10% of wage earners, you’ll find the college premium will shrink dramatically but not entirely as a portion of it is made up of people who found something really different an unexpected in college and that altered their life path dramatically."

I'm not sure if I am understanding you correctly here, but if you are suggesting that the effects you argued for just above (and that I am agreeing exist) might be more common in the bottom 90-95% of college-educated wage earners, I don't think I agree. My educated guess would be that the largest impact of the effects you are describing would be among the top 5-10% of wage earners. Because they will often have the most potential to begin with (because of other issues that lead to a wage premium, like selection bias), mis-matching them due to a failure to explore their options and interests would be the most costly.

I am a great fan of self-study, and I can recognize that signalling is part of it, but too many commentators fail to differentiate between narrow and well rounded education. It is tedious for the student that old guys define a curriculum, required classes, but that is what you need for true preparation in a field.

Maybe an ____ has no interest in statistics, but maybe they should really have it to be certified with a degree in ____.

Meanwhile, in worst case scenarios for curriculum design:

If that is true, and Wolfers is not jumping the gun, very bad. I hope the meta on this whole education theme at MR is that young earth creationists are harmless, because signalling.

Outside of specific professions, the signalling occurs at hire time. Past that point it is about being able to pick up the specific requirements of the job quick enough to keep it.

There are already people taking advantage of the lower costs of non credentialed. It is called the economy. Anyone with some sense, the ability to work hard, to take tasks to completion etc. is working. The college educated ones are simply more expensive.

Take a relatively soft degree like b.a. in business administration. Yes, it signals sticking with it. It signals institutional compatibility.

It also tells a hirer that you have exposure to basic accounting, information systems, economics, marketing, management .. actually a pretty long list.

A view that is too signalling focused may forget that flexibility comes from that kind of foundation.

Again, a great deal of manpower in academe is spent teaching and learning things without the practical application of business administration and some of the notionally vocational programs are just paper hoops.

I am definitely not a fan of low value degrees, and I think we should ferret them out by statistical methods, reduce their role, especially in public institutions.

On the other hand, an a.a. in restaurant management probably does more than signal. It literally keeps down the vermin.

Exactly. And there should be more a.a. programs out there.

A few general thoughts:

(1) The extent to which any premium from education is selection bias vs. signaling vs. skills acquisition vs. something else almost certainly varies across disciplines.

We all need to be more careful about generalizing too much from our own experiences in specific fields to other fields. And we need to be sure that when we argue a point, we are talking about the same thing. If one person is arguing a point assuming a history major, and the other is assuming an EE major, there will be a lot of pointless arguing past each other.

(2) We should not treat "education" as a black box, nor any of the types of signaling or skills acquisition etc. as black boxes.

Post-secondary education has many components, and they can be unbundled. Signalling and skills acquisition also have many components (and I suppose selection bias too), and we can be more specific. Perhaps "education" signals underlying potential as well as the willingness to slog through boring work. Which parts of "education" signal what, and to what extent? Is the "potential" signal mostly the admission itself, or is it the performance in college courses? Is the "slog" part the number of years, the number of courses, the amount of pointless BS needed to attain a certain grade in each course?

The answers will tell us a lot about how the system would ideally change. It isn't just a question of should the current higher education system exist in perpetuity vs. should the universities be burned and everyone sent into the army. It is, among others: what parts of "education" are useful; which parts are most useful; can we achieve those useful parts more efficiently; and are different parts more useful to do different people?

(3) We need to separate the selection bias vs. signaling vs. skills acquisition vs. whatever else issue along the "going to school" vs. "going to a better/worse school" axis.

We should be trying to determine, say, what part of the measured value of an EE degree attained at the local state school is selection bias vs. signalling vs. skills acquisition etc. Then we should try to determine, to the extent there is a wage premium for getting the same degree from MIT, to what extent that premium can be attributed to each effect.

For law school, for instance, I tend to think that the difference in wage premium between going to Harvard/Yale/Stanford and whatever school rounds out the Top 100 is almost entirely signaling/selection bias/network effects. The practical "lawyering" skills acquired at Harvard, say, are probably if anything worse than the practical lawyering skills the same person would be forced to acquire at school #100. So not only is admission a signal of ability, but the school's curricular focus is an additional costly signal: hey, look at us, we are so smart that we can be less prepared to be lawyers than most students coming out of lower-ranked schools and not even care; that shows how great we are.

An attorney I occasionally correspond with had this to say about law school:

1. Law school curricula are designed to train appellate judges, which most lawyers will never be.

2. He was told by the senior partner of the firm which hired him "young lawyers are useless". He offered that the senior partner was right . He and his classmates had no clue about how to go about the day-to-day business of law practice.

The lawyer in question has suggested that law school be reduced to one year of study followed by apprenticeship in law offices.

I've know people admitted to law school whose antecedent preparation came from the following fields: teacher-training, theatre, art history, history, and international relations. The BA as preparation for law school is a paper hoop.

Also, currently, awards of JD and LLB degrees are running at 49,000 a years. You have 625,000 law jobs in this economy. Even allowing for some career shifting over the course of one's work life, the number of law degrees awarded as we speak is nearly double the number necessary to staff the legal profession.

Pretty much correct on all accounts, though I would say that 1 & 2 are probably more true the higher up you go in "Law School Rankings." Which can be slightly justified by saying that a lot more students at Harvard/Yale/Stanford will become appellate judges or law professors than those at lower-ranked schools. But the portion is still so small that it isn't really justified on that basis (plus a lot of students end up clerking for appellate judges anyway, so they get that experience right after law school).

A one year program plus apprenticeship would make a lot more sense, and lead to fewer heavily-indebted young people. I had friends leaving school $200k+ in debt. And that was almost five years ago, so I assume it is even worse today.

Though three years of law school was a nice break from the real world, and it came with a socially-approved narrative for why I was taking that break. Not a break I would have paid for if it didn't provide so much signaling value, however.

About the mobsters, a possible explanation could be that people with high academical achievement have better job prospects in the non-criminal world, then they only go to organized crime if the prize is high?

I agree there is a certain conformity and conscientiousness screen/infusion represented by going to college.

The question is just how valuable this is. College is awfully expensive in time and money. It's hard to see it offering that good of a filter in non-elite circumstances. I think the game theory incentives of HR departments have to be a part of that equation.

This is all beside the point. There's selection bias. Educated people can make more money outside the mob, and presumably those who choose to go into the mob will demand more money on average to be part of it. If they don't get paid more, they don't join the mob and aren't counted in the sample. This holds regardless of what is the reason underlying why they might be paid more.

The paper also supports a traditional, skills-based argument for education:

Within that, those charged with complex crimes including embezzlement and bookmaking have the highest returns.

Familiarity with finance/accounting can definitely be useful in a criminal context, both in the '40s and today (e.g.
If the title of the paper were "Returns to Education in Criminal Organizations: Did Going to College Help Tom Hagen?", the answer would definitely be "Yes".

I made this point yesterday. Are we really surprised that Tom Hagen, the guy Don Corleone sends to law school, is making more money than the guy the Don helped to steal a rug at the risk of his life?

When you think about it, the Don basically has a white collar job. There is a lot of violence, which he delegates. Aside from that, he runs bookies, has an import/export company, and has political / legal / union connections. He views himself as someone who does favors and calls them in.

Don't forget, a big part of Bryan Caplan's signalling model of education is that conformity is part of what you're signalling. The more meaningless, expensive, and inefficient college is, the more powerful the conformity signal. Ordinarily, markets reward mavericks with great foresight into efficiency, but try to sell yourself as an education maverick these days: "No, sir, I don't have a formal university education, but I did go to a coding boot camp and I'm 3/4 of the way through the entire list of course offerings at Marginal Revolution University."

Now suppose it's not a job interview, but rather dinner with your daughter's new live-in boyfriend. Get the picture?

We had large industrial concerns in 1928 with stricter work rules than you see nowadays. Somehow, they coped without putting 43% of incoming entrants to the labor market through 5 years of higher education. What was the signal then?

There were tons of other signals back then. For example, back then employers were allowed to discriminate on the basis of things that are prohibited these days. Interviewers could ask questions about an applicant's marital status, number of children, etc. etc. The work was also less specialized and relied more on on-the-job training and/or apprenticeship, so pre-qualifications were far less important. Personal references actually meant something; they were more like recommendations rather than whatever they're supposed to be now.

All of these varied social institutions combined to form a signal of the applicant's ability to perform the work. And if the applicant failed to do it, it was far easier to fire him and hire someone else.

Totally different world, not easily compared.

"Signaling" is turning into a dangerously defocused word. Are we supposed to believe that everything which is learned in college but not a day to day requirement for your job was just signaling? Yesterday we had someone claiming that most of engineering education is signaling.

For any job more complicated than professional tire changer, there will be a big difference between the everyday, routine requirements of the job and the occasional, very difficult challenges that you encounter. Preparing yourself for the challenges will require learning things that don't come up very often and may require some formal theoretical background. That's not signaling so much as reasonable preparation for the unexpected.

Members of criminal organizations steal from the legitimate world. It would help to understand that world well, and four years of college might do the trick. College graduates probably have an easier time gaining the trust of bankers, lawyers, politicians, and businesspeople. It isn't signaling, it is surface acculturation.

This is where the networking is truly useful. You meet a lot more suckers in college.

Education is very important factor towards success, if we have education then we can get whatever we wish quite easily, but if we aren’t educated enough in whatever field we pick than it will be struggle. I am fortunate enough to work with OctaFX where their educational guidance, articles and daily market updates helps me a lot and makes it so easy for beginner to work out perfect way in generating revenue and also being able to do it with consistency.

Comments for this post are closed