Bryan Caplan on The Case against Education

CaplanDespite being immensely popular—and immensely lucrative—education is grossly overrated. In this explosive book, Bryan Caplan argues that the primary function of education is not to enhance students’ skill but to certify their intelligence, work ethic, and conformity—in other words, to signal the qualities of a good employee. Learn why students hunt for easy As and casually forget most of what they learn after the final exam, why decades of growing access to education have not resulted in better jobs for the average worker but instead in runaway credential inflation, how employers reward workers for costly schooling they rarely if ever use, and why cutting education spending is the best remedy. Romantic notions about education being “good for the soul” must yield to careful research and common sense—The Case against Education points the way.

The “signaling model of education” is the foundation of your argument. What is this model?

The standard view of education, often called the “human capital model,” says that education raises income by training students for their future jobs. The signaling model, in contrast, says that education raises income by certifying students for their future jobs. Doing well in school is a great way to convince employers that you’re smart, hard-working, and conformist. Once they’re convinced, career rewards naturally follow.

Could you give an analogy?

Sure. There are two ways to raise the value of a diamond. One is to hand it to an expert gem smith so he can beautifully cut the stone. The other is to hand it to a reputable appraiser with a high-powered eyepiece so he can certify the pre-existing excellence of the stone. The first story is like human capital; the second story is like signaling.

Is it really either/or?

Of course not. The human capital and signaling models both explain part of education’s career benefits. But I say signaling is at least half the story—and probably more.

And why should we care about signaling?

Key point: In the human capital model, students go to school and learn how to produce the extra income they’ll ultimately earn. They get more of the pie because they make the pie bigger. In the signaling model, in contrast, school raises students’ income without raising their productivity. They get more of the pie without making the pie bigger. How is that possible?  Because in the signaling model, education is redistributive; it’s a way to grab more for yourself at the expense of the rest of society.

Selfishly speaking, of course, is doesn’t really matter why education pays. But from a social point of view—a public policy point of view—it makes all the difference in the world. If the signaling model is right, education enriches the individual student, but actually impoverishes society. Using education in order to spread prosperity is like telling the audience at a concert to stand up in order to see better. What works for the individual fails for the group.

But isn’t assessing workers’ quality socially valuable?

To some extent. But once workers have been ranked, giving everyone extra years of education is socially wasteful. Furthermore, since the status quo is supported by hundreds of billions of dollars of subsidies, we’re probably underusing alternative certification methods like apprenticeships, testing, boot camps, and so on.

In 2001, Michael Spence won a Nobel Prize for his work on educational signaling. Can the idea really be so neglected?  What is your value-added here?

Signaling enjoys high status in pure economic theory. But most empirical labor and education economists are dismissive. Either they ignore signaling, cursorily acknowledge it in a throw-away footnote, or hastily conclude it’s quantitatively trivial. My book argues that there’s overwhelming evidence that signaling is a mighty force in the real world. There’s strong evidence inside of economics—and even stronger evidence in educational psychology, sociology of education, and education research. And finally, signaling has abundant support from common sense.

You say that both common sense and academic research support signaling. What are the top common sense arguments?

First and foremost, there’s the chasm between what students study in school and what they actually use on the job. How many U.S. jobs actually tap workers’ knowledge of history, social science, literature, poetry, or foreign language?

Signaling also explains why students are far more concerned about grades than actual learning. They want “easy A’s”—not professors who teach lots of job skills. Signaling explains why cheating pays—a successful cheater profits by impersonating a good student. And signaling explains why students readily forget course material the day after the final exam. Once you’ve got the good signal on your transcript, you can usually safely forget whatever you learned.

And what are the top research arguments?

First, there’s the diploma or “sheepskin” effect. Fact: Graduation years are vastly more lucrative than intermediate years. This is hard for human capital to explain: Do schools wait until senior year to finally start teaching useful job skills?  But it flows naturally out of conformity signaling. If your society says you should complete a four-year degree, anyone who only does 3.9 years looks weird, and hence bad. It’s just like going shirtless to a job interview: Either you don’t understand the social convention, you don’t care about it, or you’re actively defying it.

Second, there’s credential inflation. Fact: Over the last century, employers have dramatically increased the amount of education you need to get any given job. In the modern U.S. economy, many waiters and bartenders have college degrees. This would have been almost unheard of seventy years ago. This is hard for human capital theory to explain. Why would employers pay extra for workers with superfluous credentials?  But it’s simple for signaling to explain: When overall education rises, you need more education to distinguish yourself—to convince employers you’re worth hiring and training.

Third, there’s the employer learning/statistical discrimination literature. That’s rather wonkish, so anyone curious should just read the discussion in my book.

Finally, there’s the contrast between personal and national payoffs for education. Fact: Researchers have never found a country where education fails to noticeably raise individuals’ income. But there’s a messy debate about the effect of education on nations’ income. Plenty of researchers find that raising a country’s national average education level has little or no economic benefit for the country as a whole—precisely as signaling predicts. While others find modest national payoffs, the average estimate of the social return for education is far below the average estimate of the selfish return.

Wait, what’s the difference between “selfish” and “social” returns?

The selfish return evaluates educational investment from the point of view of the individual student. The social return evaluates educational investment from the point of view of the whole society (including all benefits the individual student enjoys, of course). A standard example: If taxpayers provide free tuition, your selfish return will generally exceed the social return, because you invest only your time, while society invests your time plus taxpayers’ money. The primary wedge between selfish and social returns for me, however, is signaling: If education boosts your salary more than your productivity, the selfish return exceeds the social.

Suppose you’re right. How should the education system be reformed?

Above all, we need far less education. And the cleanest way to get far less education is to sharply cut government education spending. Won’t this make education less accessible? Absolutely. But if I’m right, employers will no longer expect you to have the education you can no longer afford. In other words, spending cuts will cause credential deflation. You’ll once again be able to get low- and middle-skill jobs with a high school degree—or less.

We also need more vocational education, especially for early teens. Most researchers detect solid selfish payoffs. And if you take signaling seriously, the social advantages of teaching plumbing instead of poetry should be very large indeed.

If you’re right about signaling, should students drop out of school?

No. The whole point of the signaling model is that school is selfishly rewarding but socially wasteful. Although I also argue that even in the current regime, weaker students’ odds of academic success are so slim they’d be better off just getting a job (and job experience) straight out of high school.

Aren’t you being too much of an economist?  Isn’t the real point of education to spread enlightenment and sustain civilization?

For an economist, I have broad interests. Ideas and culture are my life. But if you look at the data, there’s little sign that education causes much enlightenment or civic understanding. Even at top schools, most students are intellectually and culturally apathetic, and most professors are uninspiring.

Given today’s political climate, who do you think will be most receptive to your message?  The most hostile?

Support for education is bipartisan. Most people, regardless of party, favor more and better education. It’s no accident that both Bushes wanted to be known as “education presidents.” That said, I think my biggest supporters will be pragmatists and fiscal hawks. And my biggest opponents will be ideological fans of education and fiscal doves. Most progressives will probably dislike my book, but they really shouldn’t. If you care about social justice, you should be looking for reforms that help people get good jobs without fancy degrees.

You’re a full professor at George Mason and a Princeton Ph.D. How can you of all people possibly challenge the social value of education?

I see myself as a whistleblower. Personally, I’ve got nothing to complain about; the education system has given me a dream job for life. However, when I look around, I see a huge waste of students’ time and taxpayers’ money. If I don’t let them know their time and money’s being misspent, who will? And if I wasn’t a professor, who would take me seriously?

Bryan Caplan is professor of economics at George Mason University and a blogger at EconLog. He is the author of The Myth of the Rational Voter: Why Democracies Choose Bad Policies. He lives in Oakton, Virginia.