Book Review: Bryan Caplan’s “The Case Against Education”

-reposted from the old blog, which I'm never posting on again.

I read Bryan Caplan’s book, The Case Against Education [1]. I thought it was good, and I recommend that people read it. The book is quite evidence-oriented, and it examines questions from multiple angles with multiple lines of research, which is very nice. The book mainly does two things with the knowledge it lays out:

1. It makes policy recommendations for basically every nation of the planet, but especially the United States.

2. It outlines to the reader how to make the most selfishly lucrative possible educational investment decisions with its advice hand-tailored to multiple different types of people (❤ thanks fren!).

The most important takeaway from the book is its answer to the following question: Educated people are paid ~70% more, but why? There are three things which could plausibly be responsible for this:

1. Education increases individuals’ productivity (human capital).

2. People who are innately more productive get more education (ability).

3. Regardless of productivity, dumb employers pay for diplomas (signalling).

“Human Capital” (1) and “Ability Bias” (2) are imo much worse names for their respective theories than are “signalling” (3) and “sheepskin”effects” (3), but whatever.

Caplan, imo, convincingly makes the case that #3 is 60%–80% of the story; this figure is the most important takeaway from his work. The book changed my mind substantially: Before, my opinion—under my unnoticed, incorrect assumption that the market is 100% accurate in judging productivity—was that #2 was ~75% of the reason, with #1 explaining the rest. Now, Caplan and I would probably both agree on the 60% to 80% figure for #3, with #1 having a small effect, and #2 explaining the rest of it. The Book provides 8 reasons to believe Caplan’s 60%-80% thesis:

1. School is filled neary to the brim with useless classes, and the irrelevant classes are valued about as much as are the relevant ones.

2. Anecdotally, professors are flattered when somebody who hasn’t paid tuition is interested in attending their lectures; colleges don’t card because the value is in the “sheepskin effect” of the degree, not the education.

3. Employers don’t punish applicants who forgot all of the material. Rather, they only punish people who didn’t get credentials in the first place.

4. On the top website for judging professor quality (ratemyprofsessors), professors are judged by how easy it is to pass their classes rather than by how marketable & lucrative their training is.

5. Graduation years in the GSS are shown to be 6–7 times more financially lucrative to a student than are non-graduation years, and Caplan shows these to be true sheepskin / signalling effects by illustrating that correcting for innate ability affects the profitability of every school year equally rather than disproportionately affecting graduation years. (He words this section a little strangely imo and it’s more clear that this is the point he is making when you read his 9th/10th/11th notes. The real value of this supporting argument is in the references rather than Caplan’s attempt to replicate these findings with the lame measure of IQ in the GSS).

6. Janitors, cooks, baristas, bartenders, security guards, etc get the same relative pay premiums for degrees that editors, programmers, etc do. If the value were in the training, then this wouldn’t happen.

7. When a candidate is hired, the pay premiums they get for their educational status shrinks for 10–20 years before stabilizing. (Not that stabilizing necessarily means that employers ever actually discover true productivity)

8. The effect of education on an individual’s wealth is consistent and substantial while by contrast, the effect of education on a nation’s wealth is inconsistent and weak (For this, he only reviews correlational evidence in order to see if more educated nations are wealthier). As Caplan points out, reverse causality is definitely possible; spending more on education is an immensely-popular proposition, so perhaps it’s the case that increases in national wealth merely enable more wasteful spending on education. Indeed, education is highly prioritized by the laity [3]:

   

   Caplan was not equipped with anything to inform us about causality on this question, but I am. At one point, Russia once had a 5-year degree system rather than the standard 4-year system; this has since changed however, and since the change applied to every student, average market performance was unaffected [2] (shame on Caplan for ignoring any evidence which did not yet exist at the time he wrote the book).

Employer Rationality:

On its face, when I don’t know about any of the evidence and I just hear that “most of the reason educated people make more money is that employers are just dumb and give them money without regards to merit”, my knee-jerk reaction is to just retort that “Businesses care about profit, so why has every employer on the planet made such imbecilic financial decisions since at least as long as we’ve had the statistics to record it?” However, this response is detached from the actual reality of business decision making: It is common for applicants to put their degrees on their resume to try to beat the competition even when nobody asks them to. When applicants mail this in on their own volition, using their own paper and their own postal stamps and everything, a hiring employer gets this information for free. Even though there is a great deal of variance in productivity within any degree level, and even though there are many “diamonds in the rough” scorned by the system, all a hiring manager has to know is that the average person with a generic master’s degree is more productive than the average highschool dropout. They don’t have to read up on the efficacy of any new measures, and more importantly, they don’t have to spend thousands of dollars buying the rights to print out a bunch of testing materials to give out to every single applicant. Even if all of the testing would technically be a great return on investment, it’s a hard sell when the profitability calculation is, by its nature, much less straightforward than an asking price of “10k for copyright, 100k for materials” (plus the decline in the number of applicants who are willing to wait through all the testing). If we pass on the testing fees to the applicant, then the applicant is just put in the unenviable position of having to explain to a hiring manager why they should accept this unconventional new test rather than just going with “the industry standard”. The benefit of this is not infinite however. During labor shortages, employers are given greater incentive to go through the effort of finding these diamonds in the rough rather than increasing salaries [4]; immigration actually fuels this sheepskin effect by making the degree premium more affordable for employers.

Positing of plausible explanations for employer behavior put aside, we don’t need to know why something is true in order to know that it’s true. Apples fall from trees whether or not we’ve thought up the theory of gravity, and employers hire educated people—largely regardless of productivity—whether or not we can explain why they do this.

The Football Stadium Analogy:

Given that the financial compensation of the educated is mostly just due to empty sheepskin effects rather than productivity increasing due to training, what happens when the public cries out that “we should invest in our children, not fighter jets” and that it’s the military that should be made to rely on bake sales? Well, what happens when you stand up at a football game? You get a better view in return for the toil of standing for an hour. But what happens when everybody stands up? The average person ends up with the same view, but now everybody is needlessly toiling, and a few unlucky short people have their view completely blocked off despite them having paid the same ticket price as everybody else. Moreover, education is a much more horrifying example because everybody can just keep on standing higher and higher until they’re 35 years old with their youth and their money poured completely down the drain. Then we cry out to pour even more funding into it! We can spend as much as we want on education, but it will never make the average person smart enough to see how awful this whole system is. If anything, giving somebody a degree will give them an ego and make them want to justify the decision post-hoc.

Caplan says the solution is simple: cut education funding (in the very specific ways that he methodically outlines). However, I disagree. I think I have a better approach: Do not cut education grants by a dime, but ban applicants from putting degrees on their resume. If the education is really that valuable, then they’ll take the money with glee, learn their stuff, get their job knowledge and intelligence certified by third-party test proctors, and put their certifications on their resumes; the companies will have higher productivity in exchange for lower employee compensation, and the applicants will have their debt cut by tens of thousands of dollars. Of course, everybody with a scrap of sense knows that the value of the schools is in their multi-thousand dollar sheets of sheepskin degree paper, so people will just do whatever is necessary to learn the job they’re shooting for, take the test, and get into the job market 4 years earlier. The government would have to give all the money back as tax refunds because nobody in their right mind would take it unless their aim is to socialize with rich people in Harvard.

It once was no big deal if you were a highschool dropout: If you were good enough, you could get a job in high-status occupations as long as you had the stuff. An employer could ask you why you have a certification instead of a degree, and you could just claim that since you couldn’t afford it, you toiled in the public library without the luxury of paid instruction, and then went hungry for a day in order to pay for a test proctoring service. This was a great excuse to an employer; rather than signaling a desire to subvert social expectations, it signals responsibility and a can-do attitude. However, I reckon that an even-more understandable excuse would be that you’d get sent to prison and raped if you were caught putting your degree on your application.

Critiques:

1. Structurally, the book isn’t put together as well as it could be. Overall, I think the book’s persuasive power comes more from the overwhelming evidence of the literature than it comes from the author doing a particularly-good job writing it.

2. The different lines of evidence for the importance of signalling disagree with each other to a slight degree in terms of the magnitude of the signalling effect that they suggest. Discussion easily could have been given as to why it is that they differ, and as to how they could be rectified with each other. IMO, the more foolproof approaches are the ones which suggest signalling to be a bigger piece of the picture.

Sauce:

1. Caplan, Bryan. The case against education: Why the education system is a waste of time and money. Princeton University Press, 2018. Retrieved from http://not-equal.org/content/pdf/misc/Caplan2019.pdf

2. Avdeev, Stanislav. “Zero Returns To Higher Education: Evidence From A Natural Experiment.” Higher School of Economics Research Paper No. WP BRP 236 (2020). Retrieved from https://wp.hse.ru/data/2020/09/22/1584505319/236EC2020.pdf

3. Pew Research Center, January, 2019. Public’s 2019 Priorities: Economy, Health Care, Education and Security All Near Top of List. Retrieved from https://www.pewresearch.org/politics/wp-content/uploads/sites/4/2019/01/PP_2019.01.24_political-priorities_FINAL.pdf

4. AP News Editorial Staff. (2022). Amid Global Labor Shortage, Businesses Rethink Talent Strategies. AP News. Retrieved from https://apnews.com/article/cb2739df66fea98e4e017f25e114dfb6