Economics professor Bryan Caplan, an elite practitioner of formal education, has shit in the nest of formal education by writing a book excoriating almost everything about it. When it comes to formal education, I’m also an unusually harsh critic. For example, I have previously written about the futility of teaching calculus in the modern world. When I heard that someone had written a whole book calling bullshit on education, The Case Against Education was at the top of my reading list.


My years of formal "education" were painful. With decades of hindsight I am now more sure than ever that the problem was not simply my deficiencies. But go to university I did and I clearly remember sitting in my (required) economics classes and having a very profound stroke of insight: if this class is not bullshit, my time spent in it should be fungible with money. What I discovered over the years is that economics courses do not magically unlock secrets that increase personal wealth but that the magic piece of paper which these courses were required to obtain do. My degree, industrial engineering, is about bringing engineering methodology to bear on maximizing returns and minimizing investments. It didn’t seem unnatural at all to turn this ROI maximizing analysis back onto the educational process itself. It was hard not to notice the extensive room for improvement.

That there is immense room for improvement in modern education is basically the entire thesis of Caplan’s book. Like Caplan, I also have a lot to say about the topic of education and for the purpose of remembering this book I’ll say much of it here. I made a lot of notes while reading this so I apologize in advance if this is overly long and disjointed.

Rather than calling it the "The Case Against Education", a more accurate title would have been the phrase used on page 239: "The Case Against Actually Existing Education". Caplan uses "education" the way it is used in the phrase, "increased funding for education". This is the "provide schooling for" definition of education. He’s not talking about the particular dictionary definition of education that is "to develop mentally, morally, or aesthetically especially by instruction". What Caplan is arguing is that all of the ostensible goals of that second definition of education are thwarted by how the first is actually practiced.

I feel like the provocative title may actually detract from efforts to improve the situation. If people are so repulsed by the apparent argument against the idealistic notion of education as they imagine it, they may not investigate beyond the absurd sounding title. Buried in the book, Caplan does make it clear that he does truly care about real education. He makes a fuss because education humbug is actually detrimental to true education, something you could say he is making the case for. That is definitely how I feel about it.


This was on my car for nine years. I wasn’t happy with the undignified cartoon Einstein that was the official logo of my son’s school so I printed my own enhancement. Despite my sardonic edit the school was excellent because it did do something properly: they taught in German. Teaching young kids a foreign language when they’re naturally receptive to it is something education can do. But the actual classroom content mostly doesn’t matter to accomplish that worthwhile goal.

The book asserts that current educational regimes are easier than they used to be and getting easier. Physics degrees, however, are never out of fashion because they are simply a proxy for any random instruction that is genuinely difficult. "As long as you avoid rare, demanding paths like engineering and premed in college, you bask in the warmth of a four-year vacation." As someone who completed one of those more challenging majors (some time ago) and who until recently provided engineering tutoring to computational molecular biophysics grad students professionally, I’ll have to take his word for it. He does provide data.

One damning point he correctly brings up is that the internet is the embarrassment of "education". What does formal education purport to achieve that the internet is not utterly upstaging? Here, for example, is an astonishingly good explanation of Fourier transforms which live instruction could not dream of bettering. Caplan points out that critics point out that this only serves self-motivated people. To which he says, to ask for more is petulant; I completely agree. This leads me to believe that if education were obvious, everyone would do it. I don’t just mean attend college, but also compulsively read Wikipedia, have a big library agenda, watch educational videos on-line more than TV, audit courses from MIT and Stanford, etc. But as far as I can tell from the popularity of general internet Prolefeed, most people are just not truly into education despite what they may claim.

Homer: Marge, I’m bored…

Marge: Why don’t you read a book then?

Homer: Because I’m trying to reduce my boredom.

Some problems with education seem obvious but they are far worse than you might imagine. For example, the staggering costs — if Caplan is to be believed, education costs vastly more to society to provide ($11e11) than defense ($7e11). Let that sink in. Read the book if you’re doubting the numbers. And there must certainly be similar or worse opportunity costs — students sitting there wasting potential en masse.

Part of what makes education so futile is that retention is almost nonexistent — he cites a study where 25 years after learning algebra and geometry in school, almost everybody had forgotten almost everything about them. He uses the idea that schools can’t be responsible for more than 100% of what people generally know about subjects they’ve been "taught", i.e. disturbing levels of post graduation ignorance show how sad the ultimate effects of education are.

Here is an email from my son’s math teacher (who I am sure is a hard working competent person using best practices). Note how they struggle to hold the house of cards together until they reach the safety of The Final Exam.

The best way for your student to prepare for their Final Exam is to practice Math each and every single day (including weekends!). I recommend they practice 1 problem from every unit every single day — this amount of practice will not take more than 15 minutes, and it will ensure they do not forget any information when they take their Final Exam.

No mention of ensuring recall for the (basically nonexistent for most people) opportunities to apply this learning in one’s actual life. As you can see, this "education" is designed to be forgotten.


I’ve always felt that homework was an intellectual failure of the education process. It’s a bit like saying that flat pack furniture provides quality of life by giving you something to assemble. Worse, I remain unconvinced of homework’s efficacy. I am now not alone. On page 241 Caplan calls the majority of schooling "insipid busywork". I pretty much agree.

It is now easier than ever to properly criticize traditional education because there are now much better options. I hope that cost-effective, convenient (to everyone) on-line courses which are focused on functional useful skills that employers truly want employees to have (e.g.) will start to eclipse the fluff and bother of a misdirected, onerous, unpopular educational process in the traditional format that Caplan is criticizing.

Caplan’s main thesis is that education is mostly signaling. He thinks the ratio is around 80% signaling and 20% genuine "human capital", i.e. skills of value to society. He basically believes that current educational practices give kids all of the fun and joy of child labor with almost none of the productivity or practical vocational training. Kids today learn what is valuable in a perverse artificial marketplace of grades and other nonsense while kids working jobs in the past at least were educated about what society found truly valuable. That’s right, Caplan flirts with advocating for child labor and I have to say, he made a pretty good case (unless you hate children).

Caplan believes school in general is getting easier and that there is grade and degree "inflation" where students attain higher grades and degrees than in the past for the same amount of work. Here is a NYT article about how much good grades matter; short answer, they don’t. Caplan argues that this just wastes students' lives.

"Intellectual inbreeding" is Caplan’s term for the fact that professors teach what they have been taught, and "relevance" is only attended to in that academic context. Otherwise intelligent educators may be largely blind to the deficiencies of current educational practices because, as the saying goes, "It is difficult to get a man to understand something, when his salary depends upon his not understanding it!"

If most real and valuable learning is done on the job (and Caplan’s mountains of data say that is true), formal education is actually robbing us of valuable education.


Caplan really is convinced (again, mountains of data are provided) that 80% of the schooling experience is dedicated to signaling. Four days a week kids are figuratively growing peacock feathers at school.

Remember, the scarecrow got a diploma, not a brain.

— @chrisxed

Signalling is a zero-sum game (if I’m the best, you’re the worst) but real skill development is not (if I’m skillful and you’re very skillful, together we are even more skillful). Caplan points out, "Conventional education mostly helps students by raising their status, but average status cannot rise." That is an interesting thought.

Absurd increases in education costs may allow one to signal the utter lack of common sense which seems required for today’s jobs.

— @chrisxed

On page 240 he reiterates exactly what I’ve said are the two essential ingredients for any effective presentation: a presenter who is genuinely interested in the material and an audience who is. Caplan adds a third essential ingredient — actual interesting content. While somewhat true, I would argue this mostly sorts itself out automatically with the other two. The important point here is that to say the typical classroom, at any level, is filled with people genuinely interested in learning (or teaching!) what is being presented is nonsensical.

Of course education does have some noticeable effects. For example, this article explores the question: is too much indoor time causing myopia? And in an astonishing and evolutionarily stupid twist, according to the book the main quantifiable result of education is to limit fertility. Yes, the truest clear actuarial thing that can be said about the way we practice education now is that if you subject your kids to it, you’ll have fewer grandchildren. Awesome.

Caplan does a pretty thorough job of thinking of ways that current education practices are deficient but I thought of some more. For example, I wonder if there is anti-signaling — normal people dumbing it down to gain status in some social strata that I am not a part of (perhaps related to this topic). Remember, some people do want grandchildren!

Another problem I easily perceive is what if you’re a generalist and want to octuple major in mechanical engineering, mechatronics, technical writing, urban planning, operations research, information theory, statistics, and computer science (with a "focus" on programming, graphics, networking, and machine learning)? Not only is the current configuration designed to not accommodate that, it is designed to eradicate any such thinking. To the education business, the idea of a generalist is literally worse than the idea of an ignorant person. Taking cues from education, how then does industry value these kinds of people — pretty much zero. (I can assure you!) Then consider the kinds of people who actually bring profound change to the world — they are often people who can fuse disparate fields like Ben Franklin, John Von Neumann, Paul Erdos, DaVinci, Bertrand Russell, Isaac Asimov, Leibniz, etc. Today any one of those guys would have been diagnosed as having a mental disorder and pumped full of magic beans to cure their ADHD. I don’t have any obvious solutions but the intense focus on a "major" with "high" achievement funnelled into absurdly specific PhD research certainly isn’t helping.

I’ve talked about the problems Caplan highlights in his book, but now let’s take a look at the problems with those problems. What are the reasonable counterarguments? First, here is an excellent critical reveiw of the book.

For me, it is too difficult to imagine that literacy is not helpful and attaining it is not important. Caplan is trying to shake such core beliefs but as a natural reader, that one is tough. But I’ve known illiterate people who got along just fine except for the stigma of illiteracy. The number of books those people read were pretty close to the national average of almost zero.

It is possible society needs to train 80% of kids to understand trig so that you can eventually produce society’s necessary 5% of engineers. Another thought I had is that since one of the best ways to learn something is to teach it, perhaps a captive audience of students is required to get grad student TAs up to the very top levels needed to sustain our technological civilization. That’s a messy trolley problem for you!

Caplan says, "And if we know anything about the future of work, we know that the demand for authors, historians, political scientists, translators, physicists, and mathematicians will stay low." I could argue for the (increasing) utility of these but let’s focus on the last two. It could be that very hard subjects like physics or math, besides signaling difficulty, are good at giving young people a practice domain that is broadly applicable. I know for certain that many very good practical programmers started out as physics and math majors. Caplan would ask, wouldn’t it be better to just have them study programming? Perhaps. However, given the programmer inbreeding problem, I could see how fresh perspectives might be valuable. Still, counting on that is tenuous.

One bonus of education that is not appreciated by the book — do not underestimate the confidence that you can do complicated stuff which comes from having done different complicated stuff in college, a relatively safe practice environment. Here’s an exact example between minute 1 and 2 of this talk. This may be borderline vocational though, which Caplan believes is perfectly valid and underappreciated.

A thought I had in defense of education was prompted by the question of whether reading itself was valuable. My readers (who are all highly literate and lovely and intelligent and good-looking and all around great people) may be shocked by the question but apparently it is very debatable. I wondered what I got out of reading. I certainly can completely forget everything about a book in a couple of years (hence reminders like this post!) but I rationalized that it might be more of a kind of neural network training than a "memorize these facts" kind of thing. Certainly you can read a humor book and find it funny and have memorized none of the gags. Reading does more broadly rewire your thinking in subtle ways that probably are beneficial. I think some of that reasoning applies to general education too. Of course some would argue that kids in school are getting their neural nets primed to submit to dull oppressive boring jobs as professional conformists.

I kind of get the feeling that the whole education circus might be a way to give natural nerds a chance to do a lot of reading and nerd stuff. But today, with near-compulsory signaling, people who do not enjoy reading (or writing long blog posts about the subtleties of education) or self-directed study of any kind are swept up in the "education" frenzy. That would be fine if it wasn’t such a futile waste of people’s lives.

A fun question I had — is Caplan trying to elucidate something with this book? Or is he merely signaling? I’ll arbitrarily give it a 80/20 split to highlight the arbitrariness of such a conclusion. ;-) Caplan has more support for his numbers, but not much more. Although I’m giving him a bit of a hard time, I did actually come to accept his 80% signaling as probably pretty close and certainly much closer to reality than 0% which is the ostensible figure. I don’t know if he properly does the math, but he certainly does quite a bit of math.

Education seems like unquestionable anti-evil magic. To just say the word puts everyone on the same page. People may quibble about how much to spend on the details of the education budget, but if cost were no object, it is taken for granted by pretty much everybody that any "education" is good education. Caplan is pretty brave to be the guy — in education no less — who thinks the matter is at least questionable. He questions it. He discovers that when looked at analytically the same way people make value propositions about most other facets of life, education comes out quite badly for both kids and society in general. He finds that taking the status quo for granted has produced muddled thinking about education and its worth. He finds that there are much better ways to invest our energies that better satisfy education’s presumed goals.

As someone who immediately suspected the same theme throughout my formal education, I am gratified to read such strong evidence that I wasn’t simply the one being petulant. Of course I am saddened by the current state of formal education which seems rote and somehow not genuine to me. I can only hope attitudes change and people start to question the true practical real-world effectiveness of schooling done in the ordinary way. The good news is that for those with an intrinsic motivation to learn who desire to be intellectually challenged thereby, there has never been a better time to be alive — despite the prevalence of schooling.

UPDATE 2019-07-16

Here’s an interesting article for people who have misgivings about school. Turns out that Steven Spielberg was rejected from two film schools despite being an obvious precocious prodigy. The reason: he simply hated school and got bad grades.

UPDATE 2019-07-22

I only read the abstract, but this paper seems to suggest there are diminishing returns to education: The Remarkable Unresponsiveness Of College Students To Nudging And What We Can Learn from It. I love the ironic "what we can learn from it" subtitle — might as well parenthetically add, "if we’re self-motivated."

UPDATE 2019-11-09

I just came across the best defense for the prevalent tepid system of education I have yet seen. It is made by Garry Kasparov inadvertently in this interview (@13m35s).

Because the level of concentration at the chessboard [during a match] is different from [practice exercises]. But somehow if you spend a lot of time at the chessboard preparing — in your studies, with your coaches — hours and hours and hours, and nothing of what you found had materialized [during a match], somehow these hours help — I don’t know why — always helped you. It’s as if the amount of work you did could be transformed into some sort of spiritual energy that could help you come up with other great ideas [during the match]. Again, even if there was no direct connection between your preparation and your victory in the game, there was always some sort of invisible connection between the amount of work you did, your dedication and your passion to discover new ideas, and your ability during the game, at the chessboard when the clock was ticking… to come up with some brilliance.

UPDATE 2020-03-11

Here’s a wry paper titled Teacher Effects on Student Achievement and Height: A Cautionary Tale. The idea is that many metrics are used to measure putative improvements to student outcomes caused by "value-added" teacher factors. The authors were skeptical and tested to see what effect these teacher qualities had on a ridiculous student outcome - how tall the students eventually were. It seems that the statistics supporting teacher input’s influence on educational outcome were no stronger than on height. Ouch.

UPDATE 2020-03-11

Here is a very well written review of John Holt’s "How Children Fail". Lots of comments by intelligent people going back and forth reminding us that, like most things, it’s complicated.

UPDATE 2021-08-17

Scott Alexander is not terribly worried about kids missing school. And he brings a lot of factual sounding points to the discussion. If you’re interested in this topic, worth a look.

UPDATE 2021-09-04

More from Scott Alexander and feedback from his pretty smart readers.

And I just had an interesting (original?) thought: Could the absurd grade/degree inflation and educational gluttony be the result of the Baby Boomers? When they were kids one might suppose education infrastructure had to be axiomatically ramped up quickly just to keep the profusion of kids all somewhat literate. By the time my (subsequent) generation came along, they were removing schools (e.g. I was the last middle schooler to go to a dedicated middle school in that district.) But I could imagine the optics of scaling education back down made it harder than scaling it up. "We need to close schools, fire teachers, and teach fewer kids less stuff" may have been the correct message but who would dare have proposed it?