Bryan Caplan on the Case Against Education
Feb 12 2018

Case%20Against%20Ed.jpg Bryan Caplan of George Mason University and the author of The Case Against Education talks about the book with EconTalk host Russ Roberts. Caplan argues that very little learning takes place in formal education and that very little of the return to college comes from skills or knowledge that is acquired in the classroom. Schooling, he concludes, as it is currently conducted is mostly a waste of time and money. Caplan bring a great deal of evidence to support his dramatic claim and much of the conversation focuses on the challenge of measuring and observing what students actually learn.

James Tooley on Private Schools for the Poor and the Beautiful Tree
James Tooley, Professor of Education at Newcastle University, talks to EconTalk host Russ Roberts about low-cost for-profit private schools in the slums and rural areas of poor countries. Tooley shows how surprisingly widespread private schools are for the poor and...
Tyler Cowen on Your Inner Economist
Tyler Cowen, of George Mason University, talks about his new book, Discover Your Inner Economist: Use Incentives to Fall in Love, Survive Your Next Meeting, and Motivate Your Dentist. Cowen, legendary blogger at, talks with EconTalk host Russ Roberts...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Chris J
Feb 12 2018 at 1:59pm

I haven’t actually even finished the podcast yet, but I am finding it very engaging. I want to comment on the key issue of whether the value of education can be tested. In college, I learned about two historical figures who later had a real impact on my life:

1. A professor in a core English course that I _hated_ recommended that I read John Stuart Mill’s Autobiography in response to my utilitarian arguments against the existence of her course. Many many years later, that book and the discussions I had with the professor, combined with other readings I had done to deeply impact my view of individual freedom, the value of individuals vs. a collective, and basic ethics in how I relate to other people, both in my work and family. That class definitely impacted my daily life, but it took 20 years for the impact to fully evolve. Maybe it is still evolving. You wouldn’t have found it on a test.

2. I learned about both Theodore Roosevelt and Winston Churchill in college, and although I retain very little knowledge about them, or their policies, I came away with a general respect for the ‘never give up/man in the arena’ attitudes they shared. Those attitudes directly impacted my decision making when I was starting a business (which led to a much higher income). Again, there was nothing you could test me on to have captured this learning, and its impact evolved over time in strange ways.

I am actually very, very sympathetic to Caplan’s overall argument. But on this issue, I think Russ was 100% correct. It is incredibly hard to measure the value of education, and the weird connections in the brain that may happen decades later (but, that doesn’t mean that the current education delivery system is remotely optimal).

All that said, you should definitely have Caplan on again and again. Very thought provoking!

Dr Golabki
Feb 12 2018 at 3:18pm

Great podcast – I wish there had been more time to get into k-12 education, but on the topic of college education I had a couple of questions and comment

1. Assuming Bryan is right and the majority of the value of education is related to sorting/signaling for people that are talented anyway, isn’t sorting quite useful in itself? Is there any research on the magnitude of this value generally, and the relative merits of sorting by education vs sorting within the workforce or by a simpler test (like and iq test)?

2. If education is a waste of time and money… I wonder relative to what? Are we really in need of millions of teenagers applying to entry jobs? I’m unclear on what Bryan is really proposing.

3. Bryan seems to conflate the impact of education on income and the value of education. I didn’t go to college thinking, “my goal is to maximize future earnings”, although that may have been part of the push. Education is valuable in its own right. I suppose Bryan would say “fine, but you should think about it as consumption and not investment” but I think that misses something important about what it means to be a complete person… maybe I’m just one of few nerds the found school rewarding.

Feb 12 2018 at 10:28pm

When Bryan Caplan says (at 54:40) Gary Becker should acknowledge the cost of replacement of an average or even subpar worker is probably greater than it’s worth, I had to laugh. Dr. Becker has been dead since May 2014, so unless he can communicate from beyond the grave, I don’t think his viewpoint will be updated soon.

Otherwise, in an interesting and informed podcast, Russ Roberts delivers another virtuoso performance as Dr. Caplan’s interlocutor. The latter’s work continues to push economic research in new and exciting directions. Highly recommended.

Feb 12 2018 at 10:51pm

When Russ is discussing the impact Econtalk and the professors who have stuck with him have had on him what both Bryan and Russ (unless they get to it later; I had to stop listening to make this comment) are not mentioning is the importance of motivation to how something learned stays with someone. (This should have been made more clearly with the million dollar example.) And we know intrinsic motivation matters more than extrinsic in the long run. This is why the answer lies in students having more choice about what to study and when to study it, more control over the skills they develop and how they demonstrate that development, and over the governance of the school itself. (Want to teach civics? Let students have a vote in things that matter.) Until we focus on motivation first and get over the need to program each individual student in exactly the same way and to punish and reward them with grades, we will never make education more than an expedient fixture in our lives. Russ, please have Peter Gray on and let him explain the benefits of self-directed learning.

Thank you for this great podcast.

Feb 13 2018 at 1:22am

[Comment removed. Please consult our comment policies and check your email for explanation.–Econlib Ed.]

Feb 13 2018 at 3:12am

Interesting and thought-provoking as always – thanks. I wonder if Bryan and Russ could perform their current jobs just as well if they hadn’t studied the subjects at college?

Ralph Casale
Feb 13 2018 at 9:12am

Often with your podcasts I find a desire to be part of the conversation in order to bring up topics that are either absent or only addressed tangentially. One aspect of this discussion that I would have wanted to steer the conversation towards is more the ‘what’ rather than ‘whether’ in regards the signaling / sheepskin effect(s). Specifically the concepts of grit / perseverance, as you discussed with Angela Duckworth in 2016. This was brushed past a few times with some comments on laziness, but unfortunately not noted in discussing the shortcomings of IQ (Communication skills. Empathy. Reliability. Integrity. Timeliness.)

There were many points in this discussion [Most people struggle to get through the slog.; many people don’t finish college even though it’s really not that that hard if you just put in the work.] – that could have segued into expanded comments on the signaling of ‘perseverance’. Please do consider exploring this more in future discussions with Caplan.

Feb 13 2018 at 10:36am

Dr. Golabaki makes an important point – a waste relative to what? There may be many good alternatives, but telling people right now to save money and ditch college means they will not even be able to signal. And signaling might be valuable. Maybe we need a cheaper way to do it?

Listening to Mr Caplan, I get the impression that if he hiked the Grand Canyon and years later could not remember the distance he hiked, then the experience would have not happened and nothing was gained. If he hears a symphony nothing of value was learned unless he can recall the names of the movements and the artists performing. I am sure I mischaracterizing his views, but he did seem very persistent. We have a very difficult time testing people for things we train them to do, but Mr. Caplan is certain that he can devise a test to sort out these fine nuances of experience and if people don’t pass the test than nothing was gained. What is learning and forgetting new things just makes people happy? Would that be ok?

Hypothetical questions of nonsense are not that informative. Would you rather have the education at Harvard or the diploma? The answers don’t contain much insight. Its like asking people if they would rather have Supermans powers or Aquamans powers. Neither is going to happen so what insight can be gained asking people to guess. So, let’s ask a question that might be useful – how many people are trying to obtain fake degrees from Harvard and use them to advance their careers vs people who go to Harvard and don’t pick up their degrees or skip their last exams and don’t get the piece of paper? The behavior is far more informative than the theoretical question about two impossible options.

If the government would give me my taxes back that it spends for education I would pull my children out of school the next day, higher a tutor, and allow them to escape the mind numbing life that is K-12. Their teachers are often terrible and even if they were great its very inefficient way to spend time. A job would probably be more useful and enjoyable.

Dr golabki
Feb 13 2018 at 12:05pm

I think the MBA might be the best (worst?) example of education that doesn’t educate.

I have often given people the advice that you’ll learn much more working than doing an MBA. so unless you can get into a top school (where signaling and networking are very valuable) or you have a job that will pay you to go to school part time, there’s no reason to get an MBA.

Pete S
Feb 13 2018 at 5:05pm

Fantastic episode.

Just to put a question up that I put up on Twitter related to diploma with no work or work with no diploma.

Which surgeon do you want to operate on you? The one with the diploma or the one who has done the study? Also pilot / building certifier

The only disappointment was that there wasn’t a discussion of what could be done with online education.

I’ve recently been doing a few online courses, the best of which are definitely better than some of the courses I did at a world top 50 ranked University. They cost zero.

In addition these courses can be while working and shaped around work. A future where better education is available at a fraction of the cost of current university education offers real dramatic chance for improvement.

Feb 13 2018 at 5:43pm

Another very good episode,

I was struck by how heavily weighted income was measured as the “output” of education. If you view education as a factory for creating workers, and the wage of the workers represented the quality of the factory, that might be correct.

However, What if education is actually a factory for making citizens? How would that change your evaluation?

How does widespread illiteracy and innumeracy effect the quality of your democracy?

How does a classical education improve the quality of your legislators and statesmen?

There might not be a lot of money is studying history but can you even begin to imagine the cost to society if we closed the history departments simply because there was no money in it?

JK Brown
Feb 13 2018 at 7:25pm

Sadly, this a problem with education that has been under the same discussion for at least 150 years, based on discussions I’ve found.

Consider this from 1923, by a professor who went on to write the ‘Plastic Age,’ which was the ‘Animal House’ of its time being made into a movie in the 1920s and again in the 1930s. He does offer a bit more to education that strictly signaling, but much of that has been lost in the modern schooling.

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

Consider the productive improvements in schooling. We went from slate to paper in the early 19th century. From handwritten notes to the mimeograph, film strips and the overhead projector in the mid-20th century. The chalkboard to the whiteboard in the 1990s. And digital media at the 21st century dawned. Yet, the ability of the graduate is considered to have declined over this same period.

There was a drive in the first couple decades of the 20th century to improve student learning. Research in how to teach boys and girls how to study was popular. Some efforts were made until the whole promising movement fell to the wayside with the rise of Progressive education promoted by Dewey. The decline in education attributed to the 1960s can be seen to coincide when those professors who were educated before the 1940s reached the end of their careers.

Consider this well regarded book from 1909, How to Study and Teaching How to Study (1909) by F. M. McMurry, Professor of Elementary Education, Teachers College, Columbia University.

The one thing that schools refuse to teach, leaving kids who are successful in school to work out on their own and others to fall by the wayside, is how to study. They will teach any and everything to the students, except how to do their job, i.e., study. And it is only through study that the ideas moves from random bits regurgitated on tests to being incorporated into the student’s useable body of knowledge.

The student has accomplished much when he has discovered some of the closer relations that a topic bears to life; when he has supplemented the thought of the author; when he has determined the relative importance of different parts and given them a corresponding organization; when he has passed judgement on their soundness and general worth; and when, finally, he has gone through whatever drill is necessary to fix the ideas firmly in his memory. Is he then through with a topic, or is more work to be done?

The further elements are learning to use the ideas in other contexts, having a tentative rather than fixed attitude toward the ideas and developing ones own opinions of the ideas before consider the opinions of others.

Mort Dubois
Feb 13 2018 at 8:08pm

12+ years of learning to do what you are told, while getting along with your peers, is pretty good preparation for most jobs. Caplan’s argument seems to entirely disregard the socialization aspects of education. Even if nothing is learned, college can occupy people at an age when they still have diminished judgement.

Also: what was the alternative that Caplan wished to see? Okay, don’t go to school – what happens instead?

And last: There are a wide variety of different education systems in place around the world. Are any of them acceptable to Caplan? It’s my understanding that college in Europe is often more like job training. And there are lots of countries where very few people go to school. Are those places all closer to paradise than we are?

Peter Conrad
Feb 13 2018 at 8:24pm

Wow! I don’t think that I have ever heard a more cynical view of education than Bryan Caplan’s “Case Against Education” as he presented it on the episode of Econ Talk that I listened to today. I will make only a few comments.
Here is a guy who claims to love what he does, perhaps the “research” and book writing part more than the teaching part, who seems not to expect that his students will retain much, if anything, of what he presents to them in class, and then writes a books which, he says, is meant to tell them and their parents how the whole process is a rip-off! I think that you were quite right to call him out on this, but he seems to be quite OK with it if it made him happy.
From the start of the podcast, he made it apparent that he considered the only real worth of an education is as job training, or maybe not even that but rather an avenue to obtaining a job. I began to wonder whether the education that he acquired at Princeton contained the job training required for his present position, or whether it was the “signaling” of his Princeton sheepskin that did the trick.
As to the idea of creating a “test’ that would reveal whether a person has learned any significant information from attending school, and then, perhaps, college, I would say that it would be easy. First, don’t write it down. Then, let the person being examined ask the questions as well as give the answers. Don’t limit the time and allow answers that are not necessarily strictly within the realm of “information”. My guess is that you might think that what you would get would be what comes out in the “cocktail party” scenario that you mentioned at one point. But I think that an amazing amount of knowledge of all kinds of things and the connections between them and the feelings about them would be revealed. One must imagine that being an educated person does not consist in just knowing how many Senators each state has and such trivia, or, for that matter, in having landed a “good” job.
I must add that I was never tempted to take my kids out of school and put them to work at a boring job in a factory. I did, however, spend some time trying to improve the educational environment that they had to endure. My parents did much the same for me, and I value very highly the education that I was given (and found on my own).

John Thurow
Feb 14 2018 at 6:28am

One idea that might have been missed, that is related to signaling – perhaps driving it – is self discovery. There are a lot of activities throughout life wherein you are introduced to new subjects that either capture your interest or not. The University experience is a time when you are introduced to new ideas or activities – a concentrated self discovery wherein you filter out those subjects or activities that don’t interest you and perhaps find out those that do and those that you might be good at. This can occur outside school too: scouts, music, sports, etc. but the university is concentrated and its diversity of subjects the opportunity. Once you find your interst, your passion then that sets your path to doing well thus signalling your ability to future employers. Some can find this in other ways, but do those who went right to work in a factory or other work ever have this self discovery? What would be a way for us to have this opportunity without some type of school?

Daniel Leunbach
Feb 14 2018 at 7:56am

Thank you for a great episode. I recently discovered a paper by Carolina Arteaga at UCLA which lends credence to the human capital story. It may be of interest to others in this forum. Here’s the abstract:

“In this paper I test whether the return to college education is the result of human capital accumulation or instead reflects the fact that attending college signals higher ability to employers. I exploit a reform at Universidad de Los Andes, which in 2006 reduced the amount of coursework required to earn degrees in economics and business by 20% and 14%, respectively, but did not change the quality of incoming or graduating students. The size of the entering class, their average high school exit exam scores, and graduation rates were not affected by the reform, indicating that selection of students into the degrees remained the same. Using administrative data on wages and college attendance, I estimate that wages fell by approximately 16% in economics and 13% in business. These results suggest that human capital plays an important role in the determination of wages and reject a pure signaling model. Surveying employers, I find that the reduction in wages may have resulted from a decline in performance during the recruitment process, which led students to be placed in lower-quality firms. Using data from the recruitment process for economists at the Central Bank of Colombia, I find that the reform reduced the probability of Los Andes graduates’ being hired by 17 percentage Points.”

And here’s the full reference:

Arteaga, C. (2017). The effect of human capital on earnings: Evidence from a reform at Colombia’s top university. Journal of Public Economics.

Dr Golabki
Feb 14 2018 at 9:50am


I don’t think Bryan would say education has no value beyond financial… obviously it had huge non-financial value to Bryan and Russ and you… but…

1. For the majority of students the intellectual self discovery at University is quite over stated and romanticized. Certainly many of my classmates choose their major and their classes primarily based on what seemed easiest at the time.

2. If you are graduating with $200K of debt than you better have some plan for a financial return on investment.

Feb 14 2018 at 10:19am

Loved this episode. So many good points: Sheepskin. The difference in observed professional outcomes in college finishers vs. dropouts. The paradox of preventing kids from working so they can attend school in order to prepare them for working. The predictive aspects of ‘IQ’. Using a degree to signal social conformity. Academic incest! The irrelevance of what is currently taught in school to what is required in modern job markets. The arguments for and against learning how to learn. The importance of repetition for learning. The importance of environment for learning. The depressing consequences of the case against education for educators. The opportunity cost of education. The apparent widespread difficulty firing incompetent employees. Etc. Truly, a wonderful podcast.

One point important to this discussion that was not brought up directly is programmed forgetting. There is a concept called the forgetting curve. The forgetting curve basically states forgetting follows a predictable pattern where roughly 50% of what is learned is forgotten in one hour, 75% by the next day, 90% after one week, and at least 97% one year out. The few studies on the forgetting curve that I can find are woefully underpowered, uncontrolled, unrandomized, etc. But the forgetting curve, if real and universal, would produce exactly the outcomes Bryan Caplan is reporting regarding education. He said, “’Which is better?’ … it’s a head-scratcher: Just that you do have to think about it, and there’s arguments on both sides. And it’s confusing. And it’s interesting that it’s confusing. It’s interesting that it’s not something we’re going to say, ‘Well, duh, of course it’s better to have the education, because then you know stuff and you know how to do things.’” And, “’Does the current education system actually teach people how to think?’ And that’s what I’ll say, ‘Barely.’” If, as the forgetting curve predicts, 97% of what is taught in school is forgotten one year out, then the expected attribution of school content to predicted success outside of school would be 3% at most. Meaning a diploma—in practical terms—could ONLY act as a signal.

A second important concept that might additionally inform this conversation is the function of the frontal lobe. The frontal lobe of the brain enforces discipline on the more primitive and emotional areas of the brain. The frontal lobe is the last part of the brain to mature and does not reach full maturity until—on average—age 26. The primitive and emotional parts of the brain are, relatively speaking, prone to taking mental shortcuts and behaving impulsively. Importantly, exercising any part of the brain makes it stronger. So, even though sitting in school learning to regurgitate facts is uncomfortable and unnatural, that might be the point. Practice focusing, remaining still, and not getting distracted probably counts as exercise for the frontal lobe. A stronger frontal lobe probably would show up as a higher IQ on tests. And it seems reasonable to assume that the ability to focus, avoid shortcuts, and avoid distractions predicts success in most occupations. Students dropping out of college or high school before acquiring a diploma implies strongly that they are impulsive, lack focus, or are unable to avoid distractions. Thus a diploma may ‘signal’ a strong and functional frontal lobe. And a lack of diploma ‘signals’ the opposite. For completeness, these are only signals. Bill Gates dropped out of college.

I usually don’t enjoy when Russ argues with guests, but I take exception when he is arguing with someone in his own field. His arguments with other economists are always delightful and this was no exception. Caplan was better prepared during the arguments, but Russ was making argument in all the right places. I’d like to throw in on Russ’s side of the argument over ‘learning to learn’ that Caplan leans too strongly—entirely in fact—on lack of evidence as evidence of absence. “We’ve been searching for these effects of transfer learning for a hundred years. I thought I could change things around but I’ve been in it for 30 years now; I can’t find anything either. The guess is it’s probably not there.’… That’s the kind of research that I really trust.” That’s flawed logic. Not that Caplan is wrong, just that his argument that lack of evidence being meaningful is not convincing.

Feb 14 2018 at 10:27am

Interesting talk, but I have to say, I find many of Bryan’s arguments totally unconvincing. The core of his argument seems to be based on research by introspection: I can’t think of much that is learned in school that could be valuable on the job, therefore education must only be useful as a signal.

As a labor economist, I would have expected his argument to take several empirical issues more seriously. Here are some questions:

1) How does he account for the large positive effects of compulsory schooling laws (e.g. in Britain in the ’50s) on earnings. If everyone of a particular cohort is forced to stay at school, there cannot be a payoff from signalling.

2) His argument about returns to schooling over the lifecycle are pretty hand wavy. Lange (2007, JLE) shows that employers learn about worker productivity within the first 3 years. Is his contention that labor markets are so uncompetitive that a college grad will continue to be paid almost double relative to his equally productive high school counterpart year after year?

3) Caplan appears to have ignored the large literature on the non cognitive effects of education. We don’t just learn how to read and write at school, but also self motivation, determination, problem solving, etc.

4) Education appears to be an awfully inefficient way to signal productivity. Why hasn’t Bryan started his own “employee rating company”? He could be making millions!

Feb 14 2018 at 11:32am

Very interesting listen.

Here’s what I’ve observed after 20+ years in a demanding and cyclical industry. I have an engineering degree. I wouldn’t have gotten my entry level job without the credential, and I wouldn’t be able to credibly perform the tasks required of that entry level job without at least some of the knowledge I got in school.

However, the further I progressed in my career (I now earn several times what I earned as a recent graduate), the less that knowledge (and credential) mattered. Knowledge, skills and abilities (especially “soft skills”) obtained outside of school have been much more important. That said, the ability to take on additional skills and knowledge necessary for career advancement probably correlates pretty highly with the ability to obtain a 4-year (or higher) degree.

At least in my industry, no one ever asks where you went to college, unless it’s to give you a hard time about their football team.

JK Brown
Feb 14 2018 at 11:53am

In regards to the Princeton diploma or Princeton education question, I was reminded of this bank robber prisoner who had more SCOTUS briefs accepted than almost all credentialed lawyers, but still needed a law school “sheepskin” to get paid for that skill.

Now [Shon] Hopwood was spending his time doing things like reading a 1,650-page textbook on criminal procedure. Twice. And with new sentencing guidelines, he was busy churning out work for other inmates, taking on 10 or more cases at a time. “I was running a law firm in prison,” he said lightly.

A NY Times article and some intervention by a friend “in the business” got him jobs and eventually a full ride to law school, as well as acceptance to the Bar. Now he’s a Georgetown law professor.

He has skils and proven ability, but without the magic paper, he’d just be an ex-con looking for a job.

Mads Lindstrøm
Feb 14 2018 at 3:12pm

Again, it’s not about, ‘When did you learn about how many senators every state has?’ but ‘Do you know that every state, how many senators each state has?’

Russ and Caplan both agree that it is not important to know how many senators each state have. Whether each state has two or three senators is not important, but if you know each has two you also know that each state has the same number of senators. Knowing that states has an equal number of senators is important to understand the American system. And it leads to the important question of why an equal number.

My point is that seemingly non important facts, sometimes hides important facts.

Duane Pontius
Feb 14 2018 at 3:17pm

“After nine or ten nights he understood with a certain bitterness that he could expect nothing from those pupils who accepted his doctrine passively, but that he could expect something from those who occasionally dared to oppose him”
– Borges, “The Circular Ruins”

Feb 14 2018 at 3:44pm

To the human capital theory people:
If there was a separation of education and testing and credentialing do you think school would be pretty much the same as it is minus the tests? I think schooling would be very different.

If I want to learn something I generally do not go to school to learn it. I live in a city dominated by schools (Santa Fe College and the University of Florida) but when I wanted to learn Spanish I didn’t sit in on 1 college class, which would have been free. Even some people in school get tutors to teach them because tutors are much better and not really unaffordable. Even universities hire tutors to teach athletes rather than send them to more classes.

Feb 14 2018 at 11:18pm

Bryan Caplan is one of my favorite economists, along with Scott Sumner, and for similar reasons. Bryan follows the logic of his arguments relentlessly, and seems to have fun all the while. His arguments–and evidence– are not only persuasive but have important policy implications. Our country, and any country, could surely raise human capital far more effectively and with far less waste by encouraging diverse approaches, not least more opportunities for young people to get practical experience in a useful trade or field. Everyone would benefit–most of all those young people who get little out of formal schooling at best and create disruption for some who do at worst. Bryan’s cheerfulness shouldn’t obscure the fact that his message is a serious one, and the dysfunction throughout our education system a massive waste.

David C Morris
Feb 15 2018 at 3:30am

In response to:

Russ Roberts: Well, I suggest–if you talk to an employer and ask them what are the most important factors in job performance and how would they weight them–I’m not sure they–I don’t think they would pick IQ to dominate everything else. I‘m just thinking about so many people who I’ve met outside of academic life, who, I suspect I have a higher IQ than they do; and they are so many times more effective and skilled at what they do. Partly because they’ve specialized in it. But partly because, I’m just not good at those things…

Check out John Holland’s work on Vocational Interests. He posits that personality manifests itself in the type of jobs/occupation we are drawn to as adults. There are six basic types but most people/jobs are a combination of 2 or 3.

As Bryan mentioned in the podcast, cognitive ability (IQ) is the single best predictor of many important work outcomes. Biodata is actually 2nd, but let’s set that aside and say personality (i.e., conscientious) is 2nd. Cognitive ability predicts if a person ‘can do’ the work, personality predicts if they ‘will do’ or ‘are willing to do’ the work.

Walter Clark
Feb 16 2018 at 6:52pm

I was troubled by one thing. They left out what I think is by far the most important reason for college. At least for aerospace engineering. I heard the HR VP say this.
It is a filter the company doesn’t have to administer. Getting in the college is the first filter. Money or scholarship; it doesn’t matter; both are evidence of later success. Those that make it through all the way, show resourcefulness to sneak through or the brains to demonstrate sufficient obedience intellectually. The latter is especially appreciated.
Then there’s the filter for the amount of prestige of the school they went to. That results in a confidence of their superiority and a connection to readily network with others in high places.

Dr Golabki
Feb 16 2018 at 8:06pm

Actually Bryan said that sorting was the majority (55%) of the measured value of college.

Lee O'Neill
Feb 16 2018 at 10:26pm

I found the interaction between Roberts and Caplan to be interest but also more than a little disturbing. They spent more than a considerable amount of time looking for exceptions to prove their points in the space occupied by above average instructors with above average students in above average institutions. Those discussions are interesting but they deftly ignore the enormous number of students that don’t fit into that space.

If every objection that Russ raised to Caplan’s thesis is granted, this still means, at best, is that for many, if not most, students, a college education does not deliver to students anything other than a diploma (assuming that the student graduates – not necessarily a given).

If that is even close to being true, then there are two other issues that really have to give one pause on policy and ethical grounds. The first is that the money spent on higher education is enormous. We’re not just talking about tuition either – just consider the amount of direct government subsidies to higher education AND the amount of taxes not collected because of “charitable” contributions to colleges and universities. What are we getting in return for this investment?

Well, we are getting the research and improved teaching ability of people like Roberts and Caplan. Yet both admit that they are only able to reach a small percentage of their students, and that they are above average instructors with above average students in above average institutions. Caplan also says that he receives an upper middle class income for his efforts. Recognizing that institutions of higher learning tend to lean towards the egalitarian, is the compensation of above average professors at above average institutions like Caplan all that different from the average and below average professors at the same school? In fact, aren’t all tenure and tenure track professors paid upper middle class salaries, regardless of whether they connect to students or produce valuable, time-tested research?

I wish that Roberts and Caplan had directed their conversation towards the implications of Caplan’s conclusion, rather than testing it’s premises.

Feb 20 2018 at 8:01am

I have a question about Bryan Caplan’s views that I don’t think he answered in the podcast. Suppose that it was true that employers were just selecting for consciousness, conformity, and IQ. I don’t understand why employers would prefer college graduates for these things.

But, if that’s true, then wouldn’t employers prefer high IQ people who directly entered the labor force and then showed that they could successfully hold down a menial job for 4 years? That would be far more impressive as a test for hard-work, intelligent, conformity, etc, than a college degree, especially if, as Caplan claims, most college students don’t work very hard.

Caplan’s response would be: but this person signals non-comformity because he didn’t go to college and that’s what conformists are supposed to do.

I find that to be a “just-so” story and, as far as I can tell, there is almost no direct evidence for it in either the podcasts or the book. I see no reason why almost all high-income societies would have fallen into the “conformists go to college” equilibrium as opposed to the “work in a menial job for 4 years plus an intelligence test” equilibrium.

Dr Golabki
Feb 21 2018 at 9:56am

It is a “just so” argument in the sense that it is just an empirical fact that higher IQ people go to college (on average).

If high income societies all developed independently I think you’d have a good point, but societies co-developed together with lots of communication and immigration.

I don’t really see your point on Companies wanting workers with 4 years of work v. 4 years of college. You might be right, but the effect from the students perspective is much larger. Would you rather spend 4 years getting paid 25K per year or 4 years paying 25K per year if you are going to end up in the same job after 4 years anyway.

Feb 21 2018 at 11:05pm

Have to agree with Richard. Was surprised that IQ was such a dominant factor. I guess motivation is hard to measure/quantify. And equally, if not more important, attitude.

You can be the most highly trained,skilled, motivated in your field but a bad attitude will close many doors.

Feb 22 2018 at 3:03pm

To respond to Dr. G:

As far as I can tell, Bryan Caplan never provides evidence that a major reason employers prefer college graduates is that they prefer conformists and that college graduates signals conformity better than alternative measures.

Even if a college degree did signal this, why does a college degree signal conformity? Why not another signal that actually better signals conformity along with the other measures employers care about? These are unanswered questions in Caplan’s book and podcast.

It’s plausible that employers care about conformity, diligence, intelligence, and so on. I have a hard time seeing why a college degree would signal these attributes better than actually holding down a job plus taking an IQ test administered by the employer. At first glance, work experience is more relevant to whether you’re a good worker than a college degree would be, especially if college students hardly work or show up to lectures.

Sure, students might rather spend time partying in college, although Caplan has some interesting evidence that most people prefer to work rather than go to school. But that doesn’t explain why a college degree has a large premium. It’s even more puzzling why many high-income societies have moved towards pushing college. Many societies differ in a lot of ways–but why this common trend if there are many possible ways of signaling conformity and other desire attributes?

There may be good answers to these questions on the signaling model, but I don’t yet see what they are.

Brian Scott
Feb 23 2018 at 5:27am

Dear Russ,
You did yourself a disservice, I think. There are, of course, un-measurable aspects resulting from education. The many boring, and largely forgotten, classes result in changes to our very-plastic brains, especially those of children and young people. People emerge from education as different people, thinking differently and approaching their surroundings and relationships differently. I believe it is this that is the greatest value and impact of education – and not measurable according to Bryan Caplan’s excessively mechanistic information-driven approach.
On another aspect of the conversation: President Roosevelt was described as having a second-class intellect but a first class temperament. Success is due not just to brain power of Caplan’s information-processing type!

Humberto Barreto
Feb 27 2018 at 1:36pm

Bryan said: “I think it comes down to–like, whenever we criticize professors–how lazy they are, whatever–the main thing is: Well, think about what a normal person would do, given these incentives. I mean, I think most people, if they were given a tenure-track job, they would do virtually nothing. They wouldn’t publish anything. They probably would go to class because they’ve feel embarrassed, but then once they were in there I think they would just give practically everybody As. So, again, it does speak something to the character of professors that our consciences won’t allow us to be as lazy as we could be.”

That reminded me of this: “If the teacher happens to be a man of sense, it must be an unpleasant thing to him to be conscious, while he is lecturing his students, that he is either speaking or reading nonsense, or what is very little better than nonsense. It must, too, be unpleasant to him to observe that the greater part of his students desert his lectures; or perhaps attend upon them with plain enough marks of neglect, contempt, and derision. If he is obliged, therefore, to give a certain number of lectures, these motives alone, without any other interest, might dispose him to take some pains to give tolerably good ones.”

Who said that?

Adam Smith, of course.

Earl Rodd
Mar 11 2018 at 10:49am

I would like to add an aspect to how selection bias (those who by personality/ability are high earners also go to college) works. In John Gittinger’s Personality Assessment System (Wikipedia Personality Assessement System), there are traits that are associated with success in college, both qualitatively and from analysis of the PAS interpretation of the WAIS standardization sample. For example, people who are compensated in the Intellectual Dimension of the PAS are more likely to go to college. This trait implies disciplined cognitive activity. Another is PAS Role Uniform which implies relatively poor social intelligence which leads to achievement as a way to gain acceptance. So why does this matter? These personality traits (and others I have not listed) are fixed by early adulthood according to PAS theory. They do not change in college. Furthermore work of the late David Saunders on “Reference Groups” (Facts About Reference Groups) showed that clusters of personalities were highly represented in certain professions. Again, with the personality traits quite fixed by college, personality drives profession. If that profession requires college, those personalities are likely to go to college.

What is further interesting is that with the personalities (in Gittinger’s PAS) which are less likely to go to college, there are careers with very high earnings. But those personalities are likely to have very few people in those high earning positions. For example, we find a lot of engineers in the PAS “engineer” personalities. But we find very few successful (in terms of earnings) in the “poet” personality types.

All of this is connected to IQ in that the theory behind the PAS is that relative abilities (which can be tested using the WAIS) lead to personality. See Krauskopf’s paper A Radical Hypothesis for a discussion of this part of the theory. Thus if we used not just IQ, which is an average of all the WAIS subtests, but personality interpretation of the relative subtest scores, IQ test results with PAS interpretation are likely to be an even better predictor of career performance than IQ alone.

Comments are closed.


EconTalk Extra, conversation starters for this podcast episode:

This week's guest:

This week's focus:

Additional ideas and people mentioned in this podcast episode:

A few more readings and background resources:

A few more EconTalk podcast episodes:



Podcast Episode Highlights

Intro. [Recording date: January 18, 2018.]

Russ Roberts: Bryan Caplan's latest book is The Case Against Education: Why the Education System Is a Waste of Time and Money, which is the subject of today's conversation. This is Bryan's 7th appearance on EconTalk. His most recent was in April 2014, almost 4 years ago, when we talked about many of the topics that are in his book. I'm hoping to take a new approach today to the topic based on the work Bryan's done in the meanwhile, and I'm bringing some other stuff along the way, as well. Bryan, welcome back to EconTalk.

Bryan Caplan: Always a pleasure. I've been away too long.

Russ Roberts: Well, I couldn't agree more. Well, I guess I could. But, anyway. I want to start with the empirical finding which is at the heart of your book, which is that schooling pays; and in particular, for an individual, college, graduating from college creates a large premium--I said that incorrectly. College graduates earn a lot more than high school graduates. The number you quote in the book, I think, is 73%. I want to start by pointing out that's a mean--an average, not a median. Do we have the median number?

Bryan Caplan: Someone has it, but it's not the one that economists focus on, because all of our econometrics are set up around means.

Russ Roberts: Yeah. Which is weird, because of the mean has the people from the far right-hand tail, of which they are somewhat numerous for college grads and much less numerous for high school grads.

Bryan Caplan: Right.

Russ Roberts: Now, you give--as everyone does--you lay out three theories in the book for why college graduates might earn more than high school graduates. And of course you are also skeptical of K-12 education [Kindergarten through 12th grade education]--we'll get to that along the way. But I want to start by focusing on the premium that's earned by college grads. What are the three main theories for why college graduates might earn more than people who only graduate from high school?

Bryan Caplan: Right. So, the first story is called human capital, and just says that people with more education earn more money because they have been trained effectively for the jobs they are going to do. So, school--you go to school, and it gives you more skills; you make more money. Nice simple story. The main story that I am pushing in the book is called signaling. This says that, yes, going to school does cause your earnings to go up; but, the reason isn't so much that you are learning useful skills as that you are getting certified. You are getting a stamp on your forehead saying, 'Great; hey, premium worker. Hire this person.' And then, the last story is called ability bias. This one just says that it's just coincidental that people who have more education make more money; and rather, what's going on is it's the kind of thing that wealthy people--the people who are going to be wealthy--rather do. Just like people who go and ski in St. Moritz--tend to be wealthy. You could go and say, 'Well, people who go skiing in St. Moritz earn 300% more than people in Italy, so that must be a great place to go to get rich.' And as [?] know, it's not that it's a place that you go to get rich. It's a place that you go to spend the money that you have. So, that's another story: it's just that it's coincidence, and education is the kind of thing wealthier people like doing, or people who are going to be wealthy, like doing.

Russ Roberts: So, I just want to reframe that last one just a bit. Because, you've added something there than is usually added, which is the reverse causation: that rich people like to spend money on education. Rich parents, in this case. I want to stick with the simpler idea--simply that people who go to college, rather than who stick with just high school, have more ability--

Bryan Caplan: Right--

Russ Roberts: And their higher earnings just reflect that--

Bryan Caplan: Sure, sure. Of course. Yes, yes, that's actually right. So, they would have made more money anyway, even if they hadn't gone to school.

Russ Roberts: Correct. And we don't observe that experiment. So, we get misled by the premium.

Bryan Caplan: Absolutely.

Russ Roberts: Now, I'm going to challenge Bryan here--because Bryan is, I think a very honest person--

Bryan Caplan: Well, thank you, Russ--

Russ Roberts: I think you are. It's a rare trait in our field. And it's a little bit weird, because this book--it's not a dishonest book, but it does have an agenda, by its--as you say, you focus on making the case for signaling. But I think you are capable of passing the Turing Test and making the--the methodological Turing Test here? I'm going to challenge you to take each of the three cases you've just given us, and make the best empirical case for each one. And let's leave signaling for last. Because that's your offspring, that you care the most about.


Russ Roberts: So, I want you to first make the case for the human capital argument. Then the ability argument. And then I'm going to let you close with the signaling. And this is a prelude: I'd like this to take maybe--I don't know, 10 minutes or so to get us started. Maybe 15. And I may interrupt your story. But, there's empirical evidence for all of these. And, of course, they all, as you concede, can play a role. It's not like all, 100% one or the other.

Bryan Caplan: And in fact, I say all three do play a role.

Russ Roberts: And they do. But you find, in your reading of the literature, that it's overwhelmingly signaling. The number I think you assign is 80%, as your approximate weighting--80% signaling, 20% for those other factors. So it's--

Bryan Caplan: Well, actually, so, that's a little bit misleading, Russ. Because, what I really say is that 80% of the causal effect is signaling. Eighty percent of the causal effect. So, that's really after we subtracted out the ability bias. I do have a big separate section on that--

Russ Roberts: Fair enough. Fair enough.

Bryan Caplan: Yes.

Russ Roberts: So, okay.

Bryan Caplan: And adding in my preferred estimate for ability bias is that subtracts off 45% of the apparent effect. And then, we have to divide that 55% between 20% human capital and 80% signaling. But, so, perhaps we don't disagree as much as you might think.

Russ Roberts: Okay. Got you; good point. But right now, to start with, I want you to put on your human capital hat. You are a mainstream labor economist who thinks that this signaling is overrated. What's the best evidence we have that human capital--that is, knowledge, skills, ability--not ability, but, I don't know? what's the right word?--skill--is acquired in college? Give me the best empirical evidence for that claim.

Bryan Caplan: Sure. So, I would just start off with basic academic skills--reading, writing, and math are useful in a very wide range of jobs. And there's a lot of evidence that going to school raises those skills. So, that would be where I would begin. And then, the improbably, the other thing that I would focus on, is: There is a whole literature in Psychology on the effect of education in IQ [Intelligence Quotient], and the literature review that I build on in the book says that every year of education raises your IQ by between 1 and 3 points. There's a new meta-analysis out there that says 1 to 5 points, which is a little bit higher, obviously, than what I was saying. And then I would say, 'Look, say we got, first of all, like the actual teaching of the basic skills, so reading, writing, and math; and that, you are definitely getting in school to some degree.' And then there's perhaps this unexpected additional benefit of schooling making you smarter. So, that's giving us another channel for education actually be improving people's ability to do their jobs. And there's a whole literature on--IQ is probably the single best predictor of job performance. You know, assuming it is available. So, I'd say that's probably the single best case for human capital.

Russ Roberts: Well, let me--let me challenge that in a different way than you are going to challenge that with signaling. Which is the following. I want to pay you another compliment, Bryan. You are one of the highest IQ people I hang around with in any casual way. We don't, um--

Bryan Caplan: I'm skeptical of that, Russ.

Russ Roberts: But, just take it--

Bryan Caplan: Of course, of course.

Russ Roberts: And I won't, I won't--although I could spend a lot of time on my own IQ, I'll just suggest that it's above average. But, that's, uh--we'll leave it at that. You and I are definitely above average on IQ. Which, you agree with that?

Bryan Caplan: Yeah.

Russ Roberts: So, what I'd also suggest is that you and I are remarkably unfit for most employment. Outside academic life. And so, what I want you to ponder, is your focus in the book on IQ, and raw intelligence, as correlated with job performance. It strikes me as woefully thin. It seems to me that IQ would be only a small part of job performance in many, many, many opportunities. Communication skills. Empathy. Reliability. Integrity. Timeliness. I just--there's so many other factors that I think of as important in real jobs, as opposed to jobs you and I have, that I'm wondering--

Bryan Caplan: You are absolutely right, Russ. But, you know, here again, like I did a lot of reading in Industrial Psychology, and, you know, like, the standard view there, coming out of a lot of empirical work is that the most, the single most important predictor of job performance is indeed IQ. And, while you and I might not be qualified for other jobs, that's because we have some other weird defects, and [?] about us--

Russ Roberts: Yep--

Bryan Caplan: But, in general, it's very hard to find any occupation at all where higher IQ doesn't predict better job performance. But, if you know, nevertheless, they also will say the second best predictor of job performance is conscientiousness, which captures all these other things you are talking about. So, you know, again in the book, I do talk about those things as well. The reason why I spend more time talking about IQ, the conscientiousness, in the book, is just because there has been so much more measurement of IQ. And so there's been much more research in the conscientiousness stuff is just much thinner, unfortunately. But, again, the industrial psychologists who really do work on this will say, you know, very much what you are saying. And I do try to really relay their message. Which is: Sure, this is like, IQ is by far the only predictor of job performance; but it is the single best one, although, of course, tons of other stuff is important, too.


Russ Roberts: Yeah. I just think that that says more about industrial psychology than it does about how the job market works.

Bryan Caplan: These guys are really empirical. They actually--

Russ Roberts: Well, that's a shortcoming? I hear you that that's a shortcoming. They are stuck with this really incredibly precise, 3-digit number that lulls them into thinking that they have something.

Bryan Caplan: I think they are also going in front, looking at every possible measure they can get of job performance that, once we can actually measure, like number of widgets per hour, they do that. They do employer ratings. Job performance. You know, they are really are looking at it from every angle. It's the kind of thing where, you know, economists have been kind of chuckling at psychologists' replication problems lately. Likely, we are picking the worst stuff. And when, like when you go, and least for me, when I go and read these industrial psychologists, say, 'Wow. I'm ashamed that economists are so ignorant in what they are doing.'

Russ Roberts: Well, I suggest--if you talk to an employer and ask them what are the most important factors in job performance and how would they weight them--I'm not sure they--I don't think they would pick IQ to dominate everything else. I'm just thinking about so many people who I've met outside of academic life, who, I suspect I have a higher IQ than they do; and they are so many times more effective and skilled at what they do. Partly because they've specialized in it. But partly because, I'm just not good at those things. So, I just want to mention that. I think it's quite an important issue, actually, this issue of what we can measure, what we can't measure. We'll come back to it perhaps when we talk about some of the other aspects of schooling.


Russ Roberts: But, let's go on now to the Ability argument: 'I'm an ability guy. I think this human capital argument, this signaling argument is nonsense. All the return to education is, is just the fact that we are skilled people, the ones who go to college.' And therefore--and this is a key point--and therefore, if you encourage more people to go to college, they won't get particularly higher income, because they are not like the people who chose to go now. And the analogy here would be, which I use all the time on the program, is, since basketball, professional basketball, pays more than being an economist, I should just--I made a mistake; I should just be a professional basketball player. Obviously that's not going to work out. I'm missing something important there--I'm 5'6".

Bryan Caplan: Right, right, right, or just a slightly--you know, to tweak your metaphor, I would say that it's like saying that because professional basketball players practice basketball all the time, if you, too, practice basketball as much as they do, you would be as successful as they are.

Russ Roberts: Yeah. Good one. So, what's the case--so, make the case for the Ability Argument.

Bryan Caplan: Right. So, this just says that the people who graduate from college were more talented all along. And if they had decided to go and skip college, they could have gone straight into the labor force and done a lot better. And again, I'd say that the best case for this is when we actually go and get measures of ability; and then use them as control variables for, control variables in the kinds of statistical equations that we use to predict income. It turns out that when we go and control for these measures of ability, that a lot of what appears to be paying for education really hasn't paid off for ability. And then I think you can strengthen this [?] by saying, 'Well, that's just for the abilities that we can measure.' And then there's probably a bunch of unmeasured abilities. And if we could get measures of those and put those in, then maybe the payoff for education would be minuscule.


Russ Roberts: And that's very disheartening, of course. It's somewhat unlikely, in that people do go to college. You could argue they go for consumption reasons.

Bryan Caplan: Right. Or because they are confused.

Russ Roberts: Or they are misled.

Bryan Caplan: Yes--they are children who have been brainwashed by interference in teachers--

Russ Roberts: you have to go to college--

Bryan Caplan: Yeah. Maybe. So, but there are some, as you say, there is evidence that people who do go to college are not the same as people who don't. It's not a random draw from each pool. What's left for signaling to explain? Why is signaling--why are you persuaded by the argument behind signaling?

Bryan Caplan: Right. So, you know, Step One is--the way that I do it in the book, and I think the way that makes the most sense in terms of comprehension is just to start with Ability Bias, to see how much of the apparent effect of education on career success is genuine. Right? And this is where economists have these standard things where they say, 'All right, you don't think it's genuine? Why not? Tell me why not?' Or, you think that people that go to college were smarter to begin with, and that smarts matter in the labor market, or that maybe you think they are hard-working and work ethic matters? Or let's go and get measures of these things, you think that matter; put them into operative[?] equation and then re-do the equation and see whether education still is as lucrative as it first appears. All right. So, that--you know, I go and review all of the research on that. So, like the main one that really works really well, again, is IQ. So, if you go and put IQ in as a control variable, then you'll generally see that the payoff for education will shrink by--something like 30%. Sometimes a bit less. Other times a bit more. But, roughly by 30%. And then there's other things that people put in order to go and capture the effective work ethic--punctuality, conscientiousness in general, those kinds of things. So, that literature is quite a bit thinner. My best guess is it adds on another 15%. So, in the book I mostly work with this 45% of ability bias estimates. So, really what I do is I just look at the raw--at the raw, the observed effect, the observed difference in our [?] between people who have more education or less. I subtract out 45% of that. And then, that leaves us with thing that really--the [?] right. So this is the part that's causal. This is, if you go to college, this will in fact cause you to get more money. But then, there's a question about why. So, why? And again, that's where human capital and signaling disagree. So, human capital says that the why is because you receive training and useful job skills of one kind or another. And, signaling says that it's because you got this big stamp on your forehead saying, 'Grade A. What a Grade A quality worker.' Now, why do I put so much more weight on the second story, the signaling story, than the human capital story? A whole bunch of reasons. Again, the first one is just to look at the curriculum and to see how irrelevant to the modern labor market so much of the curriculum is. So, if you just look, if you just go down, and you just look at, like, what is the way that high school students spend their time: it's shocking how what a small share of it is actually spent on reading, writing, and math. So, like under 50% for most students. So, what's going with the other stuff, the stuff, like: Gee, are you really going to use your knowledge of history? Or, you know, like, Is a non-native speaker of Spanish who gets 3 years of high school Spanish really going to get a job as a translator, or like any job that requires knowledge of Spanish? Seems pretty hard to believe. So, there's that. And just looking at the curriculum. And then you can do the same thing, or if you'd like to look at college majors, you'd see that while there are some college majors that seem vocational, that's a small minority of them. So, electrical engineering seems pretty vocational. But, you know, communications is a much bigger major. Right? History? History, you know, I think, [?], right? Yeah, yeah, so--college is--

Russ Roberts: Psychology is much--

Bryan Caplan: Yeah. Psychology. Yeah. Communications, Psychology, both much more common than Engineering. And yet, we still see that people in these majors, they make less money than the engineers. They still make a lot more money than high school graduates--your people who have only graduated from high school. So, and this is true even after going and making all the adjustments that people have made in the research for, for pre-existing ability and for everything else. So, there's that. I think, and to my mind, that's really some of the most convincing stuff. It's just looking at: What is it the students are studying? Because it seems so irrelevant to what they are ever likely to do in the future. So that's pretty hard to believe. Then, I go over, a bunch of the counter-stories. So, like one of the most popular ones is that, 'Well, sure you are not going to lose, use, user-knowledge of history on the job. But it taught you how to think. You learned how to learn.' And that's where I go and turn to Educational Psychology, and say, 'Hey, these guys have been studying this, learning how learn, learning how to think, idea for a century.' What they conclude. And what they conclude is, there's very little to it. It's mostly, it's mostly wishful thinking. So, again, like this is a highly experimental field, but there's also some other ways of doing it, too, where [?] see that people are really bad at taking something they learned and then transferring it over to a different area. People are really bad at taking something they learned in the classroom and transferring it to a practical task. You teach someone the Pythagorean Theorem, and it doesn't mean that when they are building a birdhouse that that they will actually use the Pythagorean Theorem. And so, people are really bad at that. Then, on this IQ point that I mentioned, education raising IQ: So, you know, while there is definitely something to this, the main thing, though, is that the gains look very hollow. Which means that it looks mostly like it's just teaching to the test--you know, like you, if you give someone a test, you can approve them in anything. But, you know, so it seems like of what of what school does, it functions like IQ test prep. But to the idea that it's genuinely making them smarter in some practical way, is, you know, like New York City[?], the reason, it at least seems like at least seems like the reason that--it seems like a lot, a lot, a lot of the gain that we are talking about, like 1-3 points, a lot of that is probably just hollow. Doesn't really mean anything in practical matters.


Russ Roberts: So, I want to challenge--there's one other thing I think that's very powerful, which is the Sheepskin effect. We talked about that last time we talked.

Bryan Caplan: Oh, yeah--

Russ Roberts: We're going to come back to it. But I want to talk about this issue of the learning how to learn. Which, I think, is really interesting and really important. And really hard to measure. So, some of the kind of evidence that you use in your book, I find totally unpersuasive. So, I want to mention what that is, and then you can defend it if you want. You give a number of examples where people take exams after--as adults, they don't remember anything from their history class--

Bryan Caplan: Right, right. That's not retention, which is a different subject.

Russ Roberts: Yeah; it's totally different. And I just--I don't think the goal of: Let me say it this way. I think it's extremely difficult to test or measure the impact of K-12 or college major on one's neural networks. Your brain pathways, in the future. And one reason I'll say that, is I think--I've probably mentioned this before on the program, but I think it's remarkable how little we remember about anything that we did in K-12. You know, if you said, 'Remember, try to conjure up anything you learned,' you know, I'll bring up 4 or 5 memories that have--and that's about it. I might have 10. I have one teacher I've mentioned on here before, Miss Kineen. I have more of her than anyone else. I have more of Mr. Smyth [Richard Smyth], my philosophy professor in college. I have a few from economics. But, it's shocking how many hours you spend in those classrooms and how little you remember. And I would say that's the same of EconTalk. I know there's a number of listeners here who have listened--over 600 hours; and God bless all of you, as well as those listening for the first time. But, I doubt those people who have listened over 600 hours could tell me a lot of things about what they remember. They say a few things--they could say, 'I remember that time you got into that fight with so-and-so.' They might say, you know, 'Bryan Caplan's a fantastic guest; you should have him on more often.' But, if I said, 'I've had Bryan on 6 times before this. What can you remember?' 'I remember one story, or one insight.' They might remember your argument in favor of immigration. And yet, does that really--do those memories really capture what has happened through those 600 hours, or the similarly large number of hours we spend, K-12 or in college? I'd say the answer is No. There's all kinds of subtle things going on there that we certainly can't remember. And you couldn't tease out of me if you tried. But to suggest that nothing has happened because I can't remember it, or it can't be revealed on a test, seems to me a very poor measure.

Bryan Caplan: I am going to have to disagree with you on that, Russ. So, I would say if there is no designable test that can show that people learn something, then they haven't learned it. You might say the test is bad, in which case I would say, 'Fine. Design a better test, and then show it to me.' But, if you want to say that people have been transformed but it's a way that no one can actually show, no matter how hard they try, then I'm going to say, 'No. That just sounds like wishful thinking.' But, so--there's a big difference between being able to have specific memories of someone teaching you how to read, say, and whether you can read.

Russ Roberts: Yeah; that's true. I can read.

Bryan Caplan: Now, you might say, 'Maybe you could read if you hadn't gone to school. So, what I do in the book, actually--

Russ Roberts: I could. I could read before I went to school. But that does not mean that I got nothing out of English class--

Bryan Caplan: Yes. Maybe you learned how to read better. But what I do in the book, is I try to actually bend over backwards to be fair to education. I say this: Let's just give education credit for 100% of what people know. All right? Now, that's an overstatement, of course. It can [?]--like, people learn in other ways. But I said: Let's just give education credit for everything people know, and then see how much they know. So, and that's where I think those tests of people's historical knowledge are directly on point. Because, these are very basic questions. Again, it's not about, 'When did you learn about how many senators every state has?' but 'Do you know that every state, how many senators each state has?' And when half of Americans can't do that, I want to say they don't know it. In any sense. And all the efforts to teach them about the number of senators per state were in vain, if they do not know it as adults.


Russ Roberts: I disagree. And so let me take another shot at it. I don't disagree with that last point. I don't think we are very good at teaching people how many senators there are. It just doesn't matter much. Most people don't care. So, that's not so surprising to me. But here's the thing I would give as an example. I have interviewed Nassim Taleb a number of times on this program. In the course of doing that, I've read his books. And I've interviewed a number of people related to the issues that came up in those conversations--Philip Tetlock on forecasting. I'm blanking on the same right now. But, a number of other people have spent an hour with me. And I've spent more than that with their books or their articles or their essays, trying to figure out and understand randomness and uncertainty and probability. And, like you, I took econometrics in graduate school. I was formally trained in that--to the extent that I was. I spent a lot of time on it. I certainly--I took statistics as an undergraduate. And, I've learnt a ton from those conversations and that reading after graduate school. And I think you'd be very--and I have a much different--maybe I'm fooling myself, but I much different and subtler appreciation for randomness, for the nature of uncertainty, for decision-making under uncertainty. And, I don't think you could test that. I don't think you could design a test that would pull that out of me.

Bryan Caplan: Yeah. I say it would be easy; and in fact if we go back to the first person, the first minute of this show. Remember when the very first minute of this show, you caught yourself, and you talked about how we see that education causes a 73% rise in earnings? And then self-corrected, Russ.

Russ Roberts: Yeah.

Bryan Caplan: You said, 'Oh, no, no: what I should say is they earn 73% more'? That is your knowledge of correlation versus causation, which you spontaneously used and apply. And I could give you a test. And you would blow it away if I gave you a bunch of questions like that. You would give great answers on these things. And yet, most of our students would not. Right? And, there is experimental work on this. Like, I cite this one study where they asked, you know, advanced science students in college about the effect of sleep on student performance. And, you know, like all these kids could do was say, 'Well, it can't hurt to get enough sleep.' Like, only garbage like that. So, like, even though they spent years studying the experimental method, when you get them outside of the exact subject they are studying and just give them a chance, an opportunity, to apply their knowledge--I can't remember whether it was sleep or nutrition, actually--but to apply their knowledge of experimentation and correlation versus causation, to a practical issue: Almost all of them completely blew it. Because, they, unlike you, have not internalized this stuff. And probably never will.


Russ Roberts: I have a different perspective. I'll just push back one more time, and then you can respond if you want. And then, we have plenty of other things we need to talk about. But, my thought would be the following. Which is that: These issues are really hard, a lot of them. History--the subtlety of history, which I happen to enjoy--I think the biggest impact of studying English is not how to "read better." It's not how to analyze a novel. That's occasionally what's done, but it's how to think about human beings. And I think that's what we get out of learning, reading fiction. And I think that's nontrivial. And I think it would be extremely difficult to tease that out. And I'll just mention--

Bryan Caplan: But you could totally do that. You can write a test to see whether English professors--

Russ Roberts: You could try--

Bryan Caplan: Russ. All right. No, no, no. You are getting me excited here. So I have to say--you don't really think that English professors have an especially profound knowledge of human nature because they've been reading a bunch of novels? They don't.

Russ Roberts: I think they have more than they would have had if they hadn't read them. You've got to hold constant the ability by--

Bryan Caplan: So, the kind of people that study English? Who--you think it's socially inept people--

Russ Roberts: It could be--who get a Ph.D. in English. It could be.

Bryan Caplan: Like, the stereotype is that the English Professors are the cool, charismatic ones, and we're the darks.

Russ Roberts: That's because we've--that's a very low bar. Sorry. The point I want to make--I think about the famous Monty Hall problem, the 3-door problem--we'll put a link up to that. Statistics professors messed it up. Have messed it up. Had trouble with it. Doesn't mean they don't know any statistics, they haven't learned anything. Sometimes a quirky problem will come up--even a regular problem--out of your experience will be challenging. But you will eventually--it's not easy in that case for people--but you could eventually learn how to think about it. And I think the other aspect of that, which I just want to defend, is that on the job, a lot of what we know, learn on the job, we did not learn in school. Absolutely. There's an enormous amount of job-specific human capital that we have to learn. And, to acquire that, I think being able to learn is really important. And people who have been able to show they can learn and pass exams are maybe, are better it--and they are signaling, of course. I don't deny the signaling, like you would point out. But I do think you can get better at that; and I think that's part of what school does.

Bryan Caplan: Well, let me put it this way. So, I mean, it's one thing where there's an academic field where they find exactly what they wanted to find. And then you could say, 'Ehh, I don't believe those guys. Of course, that's what they found.' What's amazing about Educational Psychology is the people who go into it are educational idealists. They love education; they want to find out that education is wonderful. And yet, you go and read the Educational Psychology, and then they are glumly saying, 'We've been searching for these effects of transfer learning for a hundred years. I thought I could change things around but I've been in it for 30 years now; I can't find anything either. The guess is it's probably not there.' You need--it's sort of like, to me, when academics find something very unwelcome to them, and they twist and turn it every way they can and they still keep getting the wrong answer; and then they just say, 'Okay, I guess that's the answer.' That's the kind of research that I really trust. And that's what it's like.

Russ Roberts: That's a fair point. It's an excellent point. And again, it could be that they just struggled to deal with some of these measurement issues that I think are non-trivial.

Bryan Caplan: Well, of course, they are absolutely not trivial. But, you know, if people have twisted and turned it every way, like, 'Can we measure it in all these different ways? Can we do it differently?' If there's a whole bunch of people making their careers if they could become the most famous Educational Psychologist by coming up with a way of showing that what everyone else thinks is wrong--it would have happened by now. I think it's as reasonable to say, 'Look: Unless you are willing to go with it's just inherently ineffable, untestable, that--then of course you can never show it's wrong.' But, you know, that just seems crazy to me to say that you have a great scale that can never be demonstrated on a test no matter how hard you try to design the test to pick it up. I mean, yeah: there's crummy tests, totally.

Russ Roberts: I'd even say it more strongly than that. I'd say: Life's nothing like a test. And, of course, that's the irony. It's an ironic comment, right? Because, obviously, it's a strange way that we've produced this signal, this credential, by having people sit down and do this thing that they're never going to do ever again. Right? There is some work under a deadline in the real world. But, most of the work in the real world is nothing like sitting at a desk with a piece of paper and a clock ticking. So, I will concede that, but I also would claim that at the age of 63, when I think about the things that I understand and learn and how many of the things I've learned--outside of school, for sure--but I think school helped me learn them, for starters. That's my first argument. And my second argument would be that the things that I have learned--and, by the way, I think the thing I have learned the most from, for sure, is EconTalk. I think being able to talk with smart people has been incredibly useful to me, and kept me intellectually curious and thoughtful, I hope. But, I think it would be really hard to show that in an exam. I really do. And, I struggle.

Bryan Caplan: And that's really a disagreement. I think, Russ, that I can write an exam that would actually show in a bunch of ways that you've improved by doing EconTalk. I really do.

Russ Roberts: I think you'd show that I'd learned about the existence of some things. Let me try two levels here. One I would call cocktail party. So, EconTalk listeners are, I like to think, very good at cocktail parties, because when someone says Bitcoin, they think, 'Oh, I learned about that years ago.' Someone mentions some other thing that's going on--I try to have a diverse program and I think people are--there's a cocktail-party return from listening to EconTalk. But, if that's all EconTalk was, I'd be very disappointed. It certainly is not the only thing it is for me. I actually think I've gotten smarter. I don't know--again, I think it would be very hard to show that. But the cocktail party part, you could. You could say, 'Bitcoin is a)., b)., c)., d).' You could have a multiple choice. I think about some of the stranger episodes we've done about things that turned out to--of course, many of them dropped off the radar, but some of them did arise and become popular and talked about; and people would email me and say, 'I learned about doing this years ago. Thanks for doing EconTalk.' That's lovely. That's great for an exam, if life were an exam. But that's not the main return.

Bryan Caplan: Right. So, just to mention someone that you mentioned earlier, Phil Tetlock--I'm a huge fan of him. Now, he, I will say, actually genuinely has improved my thinking. But, I don't think that most people read his books, have had their thinking improved. And again, I say: There actually are tests--really all that Tetlock does is he tries to come up with testable ways of distinguishing good and bad prediction, good and bad [?] insight. And, I will say from reading him, I've actually improved. But, that's because I deeply internalize this--I am a Puritanical Tetlockian: I think about it all the time. And, like, whenever I say something, like: Were Tetlock the angel on my shoulder, would he be [?], would he be frowning in disapproval and I must repent? But, I'm weird in this way, in that when I read something that seems true to me, like I just feel this incredible, this weight on the world: 'I must repent. I can't keep living the way I used to live any more. I've got to go and incorporate this knowledge into my decisions, day after day. And, I'm a sinner if I don't.' But even that is such a weird response to a book. Most people read Tetlock's Superforecasting and say, 'Oh, yeah. So interesting. Some people are really great at this stuff. Yeah. Right.' And then they go back and live their normal lives.


Russ Roberts: Well, that's my other point: this stuff is really hard--learning, knowledge, wisdom. It's easy to learn how many senators there are. I think I could--if I told you that I'd pay you in 5 years a million dollars to tell me how many senators there are, I think you'd keep track of it.

Bryan Caplan: Uhhh--

Russ Roberts: If I told you in 5 years you'd need to explain to me the Central Limit Theorem, I think you'd struggle--a lot of people would struggle. You wouldn't, but a lot of people would struggle.

Bryan Caplan: For a million bucks, I think we could get up to 10% of the population--

Russ Roberts: We might. We might. For a million bucks. But, my point is that these things are really hard; and the process by which they are acquired is not straightforward. And, I often use the metaphor of drops on a rock--that, for something to finally go in and to be accepted and to be really in your bones, you have to hear it a lot of different ways, a lot of different times. Some people, like you--you are unusual; you concede that you are unusual--you are going to grapple with it and struggle with it till you master it, and it's going to bug you until you don't. So you do. But most people aren't like that. You're right. And they acquire--

Bryan Caplan: Yeah. [?] That's exactly my point. And since they don't do that, they don't really learn it.

Russ Roberts: Well, I don't think they learn it right away. I think they learn it over time. I think they learn it through the accumulation of those lessons, those observations. I think that's the way--and through stories and narratives that grab our attention that otherwise we didn't even note--literally didn't even notice something that we've actually seen 5 or 6 times. But, finally the 6th time, something clicks and something gets embedded in our understanding, something deep. Not a fact. Facts are cheap. Facts are easy. That's not the goal of education, and never should have been. And even less so now with the Internet and Google, it's silly to memorize.

Bryan Caplan: [?] you know what? Like I say, given that we're teaching facts, that people don't even learn the facts they are taught. Why should we be optimistic that they are learning some deeper things that they are never tested on and don't have to do? And don't [?].

Russ Roberts: We shouldn't teach the facts. And you and I don't. I've been in your class. I don't think you've been in mine. But, a good economics class doesn't give a multiple choice test that says, 'The ratio of the price is equal to A. The ratio of the marginal rate of substitution, B. The slope of the production possibility frontier is C.' Etc. That's a bad class. You don't teach that way.

Bryan Caplan: Yeah. Most classes are bad classes.

Russ Roberts: Well, that's a separate issue. I agree with that.

Bryan Caplan: Well, yeah--so, in the book, I'm talking about education as it really is. I'm not talking about what education could be--which is much harder. I'm saying, 'Look, does the current education system actually teach people how to think?' And that's what I'll say, 'Barely.'


Russ Roberts: Well, I think there's plenty of problems with our K-12 and our college education system. I like your point, which you made before, that people cheer when their class is canceled, which is somewhat unintuitive for the human capital story. But, I've thought about that. That made me think a lot, as did the Sheepskin Effect--we'll get to that next. But the point I would ask you to think about in the cheering when the class is canceled is: It's hard. You are on the way to work, you are on the way to the gym, and you get there and it's been, it's temporarily closed--there was a pipe broken, there was a water leak--and so you can't do your workout. And a part of you just goes, 'Oh, thank goodness! I don't have to do that again.' Now, some people get good at working out. And they love it. And after a while it becomes easier and a pleasure. And that's true for education, remember[?]. For a lot of people it's just hard. And it's not fun. And I think one of the great illusions that you don't poke fun at--you poke fun at a lot of them, and very well in the book--but, one of the illusions you don't poke fun at is that: 'Learning should be fun.' Well, it should be fun when you can make it fun. But a lot of the time it's not fun. It's hard work. It's a slog. And, you're right: Most people struggle to finish the slog. Most people struggle to get through the slog. But, I don't think we should ever pretend it's straightforward, like teaching facts. It's not.

Bryan Caplan: Right, right. So, on the class cancelation point, I do consider your story, but there's a problem with it, Russ. Which is, at least in college, the students don't have to come to class. And, as we know, they often don't. Right? But you can always not come to class any time you want. What students like is when the whole class is canceled, so no one has to go. And then keep you, unilaterally skipping a class and having the whole class canceled--it's not like going to the gym. It's more like, if the whole gym is closed and then you [?]--that would be weird, right? Like, well, 'What difference does it make?' But, for school it matters, because you say, 'Well, now, [?]--' [?]will never know--, employers will never know.'

Russ Roberts: They are not going to get ahead of me.

Bryan Caplan: Employers will never know.

Russ Roberts: And my fellow classmates aren't going to get ahead of me, when I cancel class.

Bryan Caplan: Yes. They are not going to get ahead of me. Yes.

Russ Roberts: Yeah, there's some truth to that. And there's a weird thing--

Bryan Caplan: Which we can fix with signaling more than this myopia story that people are just lazy. Again, of course, the laziness is a big factor, too, which is why so many people don't finish college even though it's really not that that hard if you just put in the work.


Russ Roberts: Well, I'm just going to make an observation, which is not necessarily related to this, that comes to me now, which is: It's a very strange thing, the way we grade. There's sort of grading on a curve and there's sort of an absolute standard; but it's never really articulated; and everybody does their own thing--at college. In high schools, there could be an oversight where somebody says, 'Don't do this. Don't do that.' But we have a lot of freedom as university professors to grade the way we want--

Bryan Caplan: Of course. We're artists, Russ. You don't tell an artist what to do.

Russ Roberts: It is an interesting thing, that the professor, who has standards, so-called standards, and doesn't grade on a curve, could eventually find no students attending her class, as well as having, at the other end, having really wonderful course evaluations because everybody gets an A. And yet, there's no really global oversight of that. And, of course, we've seen grade inflation. To me it's surprising we haven't seen it just go to a 4.0 all across the board.

Bryan Caplan: Yeah. Of course. Yeah. I think it comes down to--like, whenever we criticize professors--how lazy they are, whatever--the main thing is: Well, think about what a normal person would do, given these incentives. I mean, I think most people, if they were given a tenure-track job, they would do virtually nothing. They wouldn't publish anything. They probably would go to class because they've feel embarrassed, but then once they were in there I think they would just give practically everybody As. So, again, it does speak something to the character of professors that our consciences won't allow us to be as lazy as we could be.

Russ Roberts: Well, that's kind of the--that's the ability issue again: It tends to attract people who like--

Bryan Caplan: Who like their subject. When I first became a professor I was dismayed at how many professors didn't seem to love their subject any more. But, in hindsight, I said, 'No, you should be amazed at how many people do love the subject. Still, after all these years and all the disappointments and they still, still, they wake up in the morning and say, 'Ah. History. It's great.'


Russ Roberts: So, I'm going to turn to the Sheepskin Effect. Before I do, though, I have to get in a comment that--I was having dinner last night with my son, and I mentioned to him that I was doing an EconTalk the next day with Bryan Caplan and about education. And I laid out the case--I laid out the different arguments and got some of his reaction. And I wanted him to understand those issues; I thought it was really interesting. And I do. And he said, 'Boy, he must,'--speaking about you, Bryan,--'he must be awfully depressed, doing something day in and day out that has no value--having people wasting years of their life sitting in a classroom.' So, I just wanted to pass that on. And I have to say, by the way, he didn't realize at first you were a professor. So, his first thought was you were just having to argue that people were wasting their lives. When I told him that you actually had a hand in it, he was even more worried about you. So, I'd like to get your reaction to that.

Bryan Caplan: Uh, sure. So, really what I want to say is that I get meaning from doing high-quality work that I think is intrinsically good. And, I may [?]--the research that I do, you know, I have some hope that it will exert some small influence on public policy. But, I'm in no way convinced that will occur. But, you know, to me, I get a lot of meaning out of just writing a book that I think is excellent and did a great job and that a small number of people who appreciate that kind of thing and who savor it, like. You know, so like, if there's 10 people who read my book and say, 'Wow. This is a great book. It's so insightful,' that, to me, justifies my, justifies my existence. Like, I feel like what I am doing is worthwhile. Of course, it doesn't hurt that--

Russ Roberts: Do you know about this opportunity cost thing, Bryan?

Bryan Caplan: Yeah, well--

Russ Roberts: Ten people? You can do better than that. Come on.

Bryan Caplan: I also get paid, I also get, I also have a nice upper-middle class or higher lifestyle from doing all this stuff. I enjoy it a lot. So--

Russ Roberts: When you earn your--just like we talk about the students who cheer when class is canceled, when you are on your way to class, do you say, 'Here I am, teaching another hour of meaningless nonsense'?

Bryan Caplan: Well, so, what I am thinking about is there are two or three students in the class who like it, and that's good enough. And I'm happy to talk to them. And those students give it meaning. Yeah, if they were all just like, hysterianist[?] based, then that would bother me. But--although, honestly, I just enjoy talking about it so much, and as long as there's one person in the room who seems to be looking in my direction, I think I can actually have fun doing the performance. Like, an actor might perform Shakespeare for themselves and get something out of it. But, yeah. To me, what gives me what gives me satisfaction is the small minority of students who, you know, like who do genuinely enjoy it. Yeah. So, actually two days ago I went back to my high school and talked to their economics class. And, like most of the students there were typical high school students. This is a required class. They have to be there. Unsurprisingly, no matter how hard I tried, they were not very interested in what I had to say. But, afterwards, one student came up and said, 'Wow, that was fascinating. Where can I learn more?' And like, that student alone gave me a high for an hour or two. I was like, 'Wow, that kid, he seemed like really sincere. And he knows I'm not giving him a grade or anything. And so maybe he got some value out of it.' And again, to me, that's plenty. So, I'm a big fan of the Greek philosopher Epicurus, whose recipe for happiness is: Set your expectations super-low. And I try to do that. And it works. And my expectations are really low, and I'm happy with what I get.

Russ Roberts: Well, I'm going to quote Mel Brooks, which you know I don't get to do that often on this program. His masterpiece in my view, underappreciated, is The Twelve Chairs--which is a phenomenal movie. I hear laughter. I hope that's encouraging laughter, Bryan, not skeptical laughter. But in that movie, the theme song is, "Hope for the Best, Expect the Worst." Which is very consistent with your motto, there.

Bryan Caplan: Yeah. I mean, just to sum it up, like, why am I not depressed? Let's see. Because I get paid a comfortable income to do whatever fascinates me. So, like, why would that be depressing? I mean it's--I would just say that this job is so good for the professors that, if you would have told me what my life was going to be like when I was 12, I just would have said, 'You're lying. There's just no way that there can be a paid job this good.' Right? But, yanno, people who take my book seriously, it will be terrible for me, because I absolutely love this job. It's fantastic for me. But, I do at least feel like I need to be a whistleblower and let taxpayers know they are getting ripped off.


Russ Roberts: Yah, and so I disagree with you. But that's okay. I loved to go into class, when I was in the classroom. And I think more than 3 people were paying attention. And I think there are more than 3 people paying attention when you speak. And I think what they learn, actually--

Bryan Caplan: Well, you taught big classes, though, right? I'm talking--10%. I'm going to say 10%.

Russ Roberts: No, I could do better than 10%. I think you can, too. But the point is, I think even the 90 that weren't riveted to my every word like they are to yours, that 90% I think got something out of the class. And I even think it could have helped them in life. And perhaps even in the job market. Which is always interesting.

Bryan Caplan: It could have. But it did[?]. You think, like, what fraction of your students walk out of movies after you talk about opportunity cost? Or, what's the change?

Russ Roberts: That's more complicated.

Bryan Caplan: That's one of the easiest applications of opportunity cost in the universe. And it doesn't happen.

Russ Roberts: You mean, to walk out in the middle of a movie.

Bryan Caplan: Yeah. Because, like, you've already paid the money; you're not enjoying it. Leave. That's like, basic opportunity cost. And yet, how many students are persuaded to do that because they took an economics class?

Russ Roberts: So, I'm going to digress for a minute; and this is off subject. I think that's a total misunderstanding of the lesson of opportunity cost, because it--

Bryan Caplan: Ohhh, Russ.

Russ Roberts: Sorry. I agree with you that I don't finish every book I start. And it did take me a long time to learn that lesson, and despite the fact that I taught people about opportunity cost. But, I think if we stop doing the things we don't enjoy, we would lead a very bad life. Because I think there are many things in life that we don't enjoy that turn out to be worth it. And I think that's the--it's ironic, because that's what I've been suggesting is true about education.

Russ Roberts: But I want to move on. I want to move on to the Sheepskin Effect, and I want to introduce it with a quote from your book, which is a great quote. It says,

Imagine this stark dilemma: you can have either a Princeton education without a diploma, or a Princeton diploma without an education. Which gets you further on the job market? For a human capital purist, the answer is obvious: four years of training are vastly preferable to a page of paper. But try saying that with a straight face. Sensible versions of the signaling model don't imply the diploma is clearly preferable; after all, Princeton teaches some useful skills. But you need signaling to explain why choosing between an education and a diploma is a head-scratcher rather than a no-brainer.

And I think that's a fantastic example that forces you to think about it. You, of course, have a Princeton diploma, a Ph.D. from Princeton, in economics. I want you to make that case. Lay that out a little bit more clearly than I just read it quickly. What's the issue there? Why is that a head-scratcher rather than a no-brainer?

Bryan Caplan: Well, because when you pose a hypothetical, people at least think about it. They are saying, 'Well, which one would be better?' And say, 'Well, on the one hand, I could have that Princeton education, but then people wouldn't know that I had it. And, how would I get people to actually give me a chance or opportunities based upon that?' Say, on the other hand, 'If I had that Princeton diploma, then that would open a lot of doors; but maybe they would find that I was a fraud. Hmmm. Which is better? Which is better?' So, that's the sense in which it's a head-scratcher: Just that you do have to think about it, and there's arguments on both sides. And it's confusing. And it's interesting that it's confusing. It's interesting that it's not something we're going to say, 'Well, doh, of course it's better to have the education, because then you know stuff and you know how to do things.' Whereas, you know, if you are on a desert island, you say, 'Would you be better to have a diploma in Survival Studies right now,'--that you would without the survival skills--or the survival skills without the diploma? On the desert island, you go of course, 'I want the survival skills.' Duh. 'I'm talking about food.' But, in society, it is a [?] question, because education is so much about convincing other people that you are worthy of opportunity. Convincing them that you are worthy of receiving training. And the actual job.

Russ Roberts: Yeah; once [?] that other person who looks around and can look up to you've got the degree, then it's a totally different--

Bryan Caplan: That again. It's like: Which one is better? Is it better to be perceived as this, or to actually be this? I mean, like, would it be better to be perceived as healthy, or actually to be healthy? And, you know, like you said, probably actually healthy. But, you know, what if people have the plague and you don't? Like, well like then no one talks to you, and everyone stays away from you, and no one wants to hire you. Maybe it would be better to have at least a more minor sickness. But to mistakenly thought to be, thought to be fine. And then other people treat you nicely; and then your life is great.


Russ Roberts: So, I think it's a fantastic example, because it forces me--who is less of a signaling guy and more of an ability and human capital guy--to concede, I think immediately, I think that, 'Oh, yeah; there is a signaling effect of that credential.' The credential is certainly a way that people in a world of imperfect information, and asymmetric information, economize on that problem, fix that problem, by using this piece of paper--this Certificate, this credential. It does raise the puzzle--and again, I promise we're going to get to Sheepskin, which is a harder situation for purists to explain on human capital grounds. But, it does raise the question which is: I can understand how that can help me get in the door, the Princeton piece of paper. But, as you point out, if I don't have the skills, I'm going to maybe struggle. Unless I've got native ability, to hide that. Should the signaling effect diminish over time, when people get more information about me?

Bryan Caplan: Right. So, there is a little bit of empirical evidence saying that's actually true, and that the signaling effect does diminish over time. I would say that the--

Russ Roberts: You'd think it would diminish a lot, though, if you were probably a skeptic.

Bryan Caplan: Yes. Right. Right. Yes. So, I mean, yeah. So, in the book I go over a bunch of reasons why you should think, why you shouldn't be so convinced that it would diminish quickly. Again, so, one of them is just that there's actually a lot of resistance to firing people in modern economies. Some of that is legal: You are worried about getting sued. But a lot of it is just basic human empathy, where once you name the puppy, once you get to know another human being, you don't want to go and fire them. Economists hardly ever talk about this, but I found there's a whole literature in Sociology on how firing actually happens. And--but you can also just go and ask a roomful of anyone and say--you know, this is a question I do like to ask my students, and it does engage with students, by the way, more than most of the other stuff. I say, 'First of all, how many people here have a job?' And at Mason, I'd say it's 80% of the students raise their hand. All right, 'How many of those people who have a job--how many people are in a job where there's at least one worker that everybody knows is incompetent?' And almost all the hands stay up. Everybody knows this person is incompetent. And so, why hasn't this person been fired yet? And, let's see, assuming--and so, the answer is, 'Well, people feel sorry for this person.' And like, it's not that--maybe the person's nice, but also, you know, or maybe the boss just doesn't have the guts to go and fire the person. And then--so, like, once you realize that, you know, contrary to, say, Gary Becker, at least what Gary Becker suggested that people who are not, who are below expectation don't immediately get fired, then there's also the fact that it's very common that when you search for other job, that your employer doesn't like you helps you--by lying, or by hiding the truth.

Russ Roberts: Because they feel guilty.

Bryan Caplan: Yeah.

Russ Roberts: There's some truth to that.

Bryan Caplan: Yeah. So, there's actually a whole specialty of people who try to help employers with termination. And they have a special word for this that I use in the book. They call it 'de-hiring.' 'Don't fire workers. Dehire them. Help them to become somebody else's problem.' And then, by the time that that employer realizes that they hired an incompetent worker, they once again name the puppy and they are sort of in the same situation that you were. So, it's actually quite reasonable to think that once you've got the right--once you've got your foot in the door, and as long as your personality does not make people hate you, that you really can, you know, coast on fumes for a very long time. And again, you're not going to be getting great promotions; you're not going to be [?] the best worker; but still, it makes sense that actually a fake diploma could keep helping you for a very long time, although probably it does [?] it becomes a bit less valuable over time. Again, not getting promotions that you would normally expect for, say, a Princeton graduate. But still, it makes perfect sense. And then on top of that, there is a reason that even Becker should acknowledge, which is: Suppose that you are only at the 40th percentile of what the employer expected you to be when they hired you. They probably won't fire you because the cost of replacing you is high enough that it's easier to just settle for moderate disappointment.

Russ Roberts: Yeah. But you don't have the same--as you say--you don't have the same job raises. You'd think that would show up in the data more [?].

Bryan Caplan: It's not that it doesn't show up in the data--as, there hasn't been that much that--the research is pretty thin. There is one paper on the sheepskin effect over time that finds that over the course of a few decades, a lot of it goes away. But, yeah, we're talking decades.


Russ Roberts: So, let's talk about the sheepskin effect. We were talking about it a little bit in passing. The sheepskin effect is the idea that--the sheepskin is your diploma, it's just a fancy, old-fashioned name, because, as we learned in the last episode, it used to be written on sheepskin. Which is really cool. That's one of the least important cocktail party things I've learned with you, Bryan. One of the more important cocktail party things--and I'm going to now make your case for you--is that, I was surprised to learn, and I think it's actually true, that people who have some college, even some reasonably large amount of college, like 3 years, don't do that much better than people who didn't go at all. Suggesting that the credential itself--the finishing itself rather than the knowledge acquired--is what's rewarded in the labor market. Summarize that literature, and make the case for signaling there.

Bryan Caplan: Sure. So, back in the 1970s when signaling first came along, there was an immediate debate about it, saying, 'Well, if signaling is true then we'd expect a lot of the payoff for education would come from finishing.' And then, the data was sufficiently crummy that the debate went on for about 15 years. Then, finally, there were some papers based upon very good data, that very strongly confirmed that the sheepskin effect was truly--namely that most of the payoff from education comes from crossing the finish lines, from diploma years. Especially for college, the sheepskin effect is enormous. For high school, it's not as big but still very large. And, from a human capital point of view, this is bizarre, because, it's like, 'Why would the last year pay so much more than the earlier years? Is it because we save all the useful job skills for senior year? And we only teach them then?'

Russ Roberts: Well, it's like all the calories are in the last brownie; if you don't eat that one, it's not fattening.

Bryan Caplan: Yeah. Yeah.

Russ Roberts: I've probably made that joke before. I apologize.

Bryan Caplan: No, it's a good joke. So, [?], again, that's pretty weird as to why that would be. Now, so there are some economists [?], just to back up--once the data became totally clear, then rather than there being a massive surrender to the signaling model, there were new models that came along and said, 'Ah, well, you could totally get that in a regular model. All you need is some ability bias.' Right? Which was in principle, it was true. So you could say, 'Maybe the graduates just had some pre-existing massive advantages over the non-graduates.' This does mean, of course, that if you randomly forced someone to not take their final exams in the last year of senior year, they would still do virtually as well as someone who did do those exams. But, again, we hardly ever observe that, so it's hard to use that thought experiment and actually thoroughly disprove it. But, anyway, so, the economists who are aware of the sheepskin effect will often just say, 'Well, yeah, this could all just be ability bias.' Although, here what I'll say is, 'Look, some of the papers do have measures for ability and it doesn't seem to change the ratio of the payoff for graduation to a regular year at all.' So, all you can really say is that, 'Sure. There's enormous effect; I don't believe it. None of my concrete efforts to debunk it work, but one day it's going to get debunked somehow.' So, of course, you can always say that; and you can't convince a person like that. But I would just say this is the kind of thing where, if you combine common sense, the common sense that every parent, every teacher, everyone who has ever advised a dissertation, like, 'Does it matter if I actually finish?' 'Yes! It matters. The world will not forgive you if you don't finish. You better finish.' Your kid says, 'I've decided to drop out,' one semester from graduation: 'Don't you dare. You finish that degree. I don't care how you do it. You better finish it.' This is--so, given that we've got common sense from almost everybody who doesn't have some dogmatic reason to disbelieve it, combined with all the evidence says it's real and none of the debunking efforts work, I say what you see is what you get. Sheepskin effect is there; it's huge. There is an interesting question of, like: Does the signalling model necessarily imply it? And, no, because the signaling model could just say every year that you suffer gives you more certification. But, this is where it's very important to think about: What exactly does education signal? And I say, the sheepskin effect is a clue that one of the main things that it signals is conformity to some social expectations. Which isn't quite the same thing as being hard working.

Russ Roberts: Yup. Very different.

Bryan Caplan: We [?] people who are very hard working, but they are defiant personalities. And, you know, I love defiant personalities. They are fun people. But I don't want to hire them.

Russ Roberts: Right. Well, you might. I just have to interject here, because a CEO [Chief Executive Officer] once told me that his best workers who are responsible for an enormous portion of the revenue of the company were the people that HR [Human Resources] always tried to keep him from hiring. Because they had, he said, they had hard edges. The HR people liked people who had everything that had been kind of smoothed over and weren't going to cause any trouble, and that defiant thing, that contrarian, outside-the-box thing. But he said: You really need those. You've got have some of those people, because they generate the ideas that make the business successful.

Bryan Caplan: Right. So, [?] we assume [?] is one where it applies to a wide range of people. So, there's the person who puts the job first but who is nice; and those kinds of people can be great workers. And then there's the kind of person who, the boss says, 'We really need you to do this,' and they say, 'Screw you. I don't want to do that.' And, that kind of person, you know, I don't think anybody wants to hire those people--like, you say, 'We're here to make money.' 'I'm here to make money for myself. I don't care what happened to this firm.' 'Well, then, you can do it someplace else.'

Russ Roberts: So, I concede that that's, all of this we're talking about, is evidence for the signaling model; and I think there is definitely some measure of conformity that is being signaled, that is of value to employers. And I encourage people to go back and listen to the episode of EconTalk we did with Nassim Taleb on Skin in the Game, one we did where he talks about the difference between and employer and a contractor. And, he suggests that employees are much like a slave, and you'd much rather have a docile slave than a non-docile one. So, that's an interesting side note.


Russ Roberts: Before we finish, I want you to defend child labor--which, you don't see that often in a book published by Princeton University Press. You make the case for child labor. I want to hear it again and let the audience hear it.

Bryan Caplan: Uh, sure. So, of course, in the 19th century kids had a lot of jobs that would seem terrible to us. And, in Third World countries, that remains true--although it's overstated as to how bad, you know, child labor in the Third World generally is. But, there is an idea that it's somehow intrinsically horrible for a kid to be working at all. And, my question is: Well, why? Aren't they learning--so it seems like they are learning some useful skills. They are making money. Why is that so bad? You know, it's one thing to say that you don't want them working in jobs where they are going lose fingers. But, why not have them work at really, work a few hours a week in a job that is, that they don't mind very much? And there is this idea, 'Well, that's terrible because it's distracting them from their school.' And, I say, 'Well, but maybe they're actually learning something more useful on the job than in school.' So, maybe it would be better if they just move some of their time over from working from school. And then there's an idea of, 'Well, but it's just a shame that a child doesn't get to enjoy childhood, and that they are there, like on an assembly line, doing something boring.' And like: Have you looked at school? Like, all of these complaints apply to school, too. So, like, why is it okay for a kid to suffer and be bored in school, and to have their movement confined and not to be free to act like a kid; but they can't do it for money? Right? And, that to me makes, really, no sense at all. I just have no idea where people are coming from. It just seems like there's this huge double standard: Where kids suffer in school, then they should just toughen up. Tough luck. Life's not fair. But, if they have a somewhat unpleasant experience when they are actually getting job experience, that's something terrible and it's a sick society that allows something like that. So, you know, what I say is, just like schools used to do terrible things to kids--you know, as I mentioned in the book, my mom went to a Catholic school where the nuns did hit children with sticks on their hands. Like, so, to say, 'Well, that used to happen, so school is bad today,' seems dumb. And similarly to say, 'There used to be kids getting black lung from treating chimneys,' or '[?]Let's stop[?] the kind of work we are talking about for kids.' So--and then from a signaling point of view what I say is that, it is much better for people to actually get practical work experience. So, even if the effect on earnings of work and school is that work experience and school are the same, still, from a social point of view it's better that they are getting work experience, because that is primarily actually acquiring skills, rather than just showing off.

Russ Roberts: Yeah, so, where we agree on this, Bryan--we agree on a lot here on this [?], actually: I'd like to see kids, at least teenagers, working more. And I like the idea that--I do think there are opportunity costs--is, small. Not as small as you think it is. But I think it's small-ish. I do think we should make school more interesting. My wife's a math teacher. Her kids actually say, 'I love calculus.' And more than 2 or 3, even. So, I do think it can be done. I think we have a lousy education system. We agree on that. And you have some very interesting things to say about separation of school and state that we've addressed on this program before. And I hope we'd continue to address in the future--have an episode just devoted to that.


Russ Roberts: But I want to close with the following. I'm surprised--because this is not my field any more. I actually was a labor economist in graduate school--meaning, I took a field in it. And, I knew about signaling. I went to Chicago. You know, Becker and Heckman were my professors. They are not big supporters--they are big human capital folk.

Bryan Caplan: Sure.

Russ Roberts: So, they are not so sympathetic. And over the years, I've become more sympathetic to the signaling idea. Partly through your work. I'd say, mostly through your work. So, I salute you for that.

Bryan Caplan: Thank you, Russ.

Russ Roberts: But, I'm surprised--

Bryan Caplan: You just made my life meaningful.

Russ Roberts: There you go. It's easy. It's so easy.

Bryan Caplan: You're one of the 10.

Russ Roberts: One of the 10. Yeah. Or, yeah, I'm one of the 2 or 3 in your virtual classroom. But, I'm surprised the profession didn't join me. I'm surprised, at least in your book, you portray mainstream labor economists as being so much more skeptical of signaling as they were in the early days. So, close, talking about--the sociology of that phenomenon? the empirical part of it? Obviously, they think they have evidence for their view. And you've suggested the evidence is not as convincing as it should be, as it might be. But, why do you think more people haven't come over to this viewpoint? And, why are you--you paint a picture in the book: I don't think it's just marketing, I think it's true, of being intellectually lonely. That you are on a bit of an outpost here. Why do you think that's true?

Bryan Caplan: Right, right. So, just to back up a little bit: I don't know that signaling was ever a popular theory, even in the 1970s. So, I think it's important to distinguish between the high status of the theorists--like Michael Spence and Stiglitz who are doing signaling models--and yeah, they got Nobel Prizes.

Russ Roberts: And Arrow--

Bryan Caplan: And Arrow, of course. But, I don't think there was ever a take in that empirical series--it was more like, you know, a gadfly on the side saying, 'Oh, yeah, there's that signaling thing we might have to deal with.' Now, why is it that opinions changed, as the evidence has mounted? So, I think there's a few things going on. One is that a lot of the best evidence comes from outside of economics. And economists, like almost everybody else, having not invented [?] syndrome, where they don't actually care to study educational psychology. They are not curious about what goes on inside of the black box of education. They don't read sociology. So, again, a lot of the best evidence is just by disciplinary grounds not their area and they just ignore it. And just tend to be dismissive of it. Or not even be aware that it exists. So, I think that's one big part of the problem. Another thing that's going on, I think, is that--signaling, it is harder. It means that you can't just go and do a regression of income and education and say, 'Aha.' You know, 'We're getting job training out of this.' So, you know, like distinguishing the two requires additional work. And again, at least for economists, the economists tend to this functionalist view of the world: the world makes sense not just for individuals but from a social point of view. So I have gotten this, 'So, if you're right, how come every country does it this way?' Right? And, you know, I have this other book, The Myth of the Rational Voter--

Russ Roberts: I hate that argument.

Bryan Caplan: Yeah--

Russ Roberts: That argument is so despicable.

Bryan Caplan: You are absolutely right.

Russ Roberts: Why does that have evidence for anything? It's like saying, you know, 'In 1500, all the countries of the world tortured their citizens, so it must have been--what's wrong with it? How could you be against torture?'

Bryan Caplan: Eyeah.

Russ Roberts: It's a bizarre logic. The logic of it--sorry. I interrupted you.

Bryan Caplan: Well, like, and Stigler--I don't know? Was he one of your teachers, too?

Russ Roberts: No. Oh, Stigler, yes. Stigler, absolutely.

Bryan Caplan: So, he's more to blame for this than almost anyone, probably, this functionalism of just rationalizing whatever exists as really being a good idea.

Russ Roberts: Some. A little bit--

Bryan Caplan: And then I think that, you know, there is just the romanticization of education. So, you know, you always have to remember that anyone who was a researcher, a professor, always, probably a good student for their whole lives. And, it's just hard for them to get inside the head of someone that didn't find it to be a rewarding experience. And, either--I didn't actually call it this in the book, because it would gross people out. But, I think a lot of what's going on with professors is what I call intellectual incest, where you go and study something and then it keeps the very thing that you studied, which then creates the illusion that what you learned is useful. Because you use it. But that doesn't mean--what about the students that don't become professors? When are they actually going to use this stuff? And then I think on top of everything else, I think there is a bit of leftwing bias, of, just, 'We love government. We love the idea of government, going and supporting this thing.' Although, what's interesting is, this is one where it's easy for economists to be bipartisan, because Democratic economists can love education because it's government social engineering; and Republican economists can love it because it's not a handout: you have to work to get the value of the education, so there's sort of a Puritanical aspect of it. And at least it seems more like it's focused on increasing the size of the pie rather than just redistribution, so it appeals to Republican economists more for that reason. And, you know, I think also Republican economists want to have one thing they can say government totally should do. And education is that thing. Right? So, I think all these factors are at work. Of course, I think they would say that it's just the strength of the evidence. But, you know, I would just say that the evidence that they have, you know, like, the papers they are publishing--I don't say that they're wrong. I just say that they don't say what they think they mean. And, of course, in economics, it's a lot easier to advance by going and solidly proving a fact than by convincingly arguing that the fact means something important.

Russ Roberts: Well, that's for sure.

Bryan Caplan: And I think that's what's going on. So, you know, I'm not saying that their work is wrong. I'm just saying that they are--you know, ignoring or unaware of other work that's also good, that, when you snap it all together provides a totally different story of what's going on. And that story is--signaling. Very important.