What's higher education for?

So this guy has a particular viewpoint and a certain axe to grind, and yet his article brings up some interesting analysis. Let's talk about education, certification and what these things mean for society in general.

Education certifies "brains, work ethic and conformity"
There is a massive gap between school and work, between learning and earning. While the labor market rewards good grades and fancy degrees, most of the subjects schools require simply aren't relevant on the job. Literacy and numeracy are vital, but few of us use history, poetry, higher mathematics or foreign languages after graduation. The main reason firms reward education is because it certifies (or "signals") brains, work ethic and conformity.

You could get a Princeton education for free...without a diploma, does it matter?
If a student wants to study at Princeton, he doesn't really need to apply or pay tuition. He can simply show up and start taking classes. As a professor, I assure you that we make near-zero effort to stop unofficial education; indeed, the rare, earnestly curious student touches our hearts. At the end of four years at Princeton, though, the guerrilla student would lack one precious thing: a diploma. The fact that almost no one tries this route — saving hundreds of thousands of dollars along the way — is a strong sign that students understand the value of certification over actual learning.

Do students who cheat really matter?
Academics and administrators also sense the importance of signaling, even if they won't admit it. Why else would they bother to combat cheating? If school were merely a place for students to invest in their skills, cheaters would literally "only be cheating themselves," spending time and tuition for naught. If, however, school is primarily a place to convince firms you're worthy of employment, cheating has a slew of victims. The cheater who successfully impersonates a good student doesn't just rip off whoever hires him. He also taints the prospects of all his peers who toiled for their degrees.

Flunking vs. Forgetting
Researchers consistently find that most of education's payoff comes from graduation, from crossing the academic finish line. The last year of high school is worth more than the first three; the last year of college is worth more than double the first three. This is hard to explain if employers are paying for acquired skills; do schools really wait until senior year to impart useful training? Or consider how differently employers treat failing a class versus forgetting one. If you flunk a class, plenty of employers will trash your application. But if you pass that same class, then forget everything you learned, employers will shrug.
One of the most glaring perversities of the modern labor market is credential inflation.
These behaviors make perfect sense if — and only if — employers are eager to detect workers who dutifully conform to social expectations. In a society where parents, teachers and peers glorify graduation, failing classes and dropping out are deviant acts.

Credentials are an efficient decision making criteria, even if not perfectly accurate
While the education workers need to do a job is quite stable, the education they need to get a job has skyrocketed since the 1940s. Sure, the average job is more intellectually demanding than it once was, but researchers find that only explains 20% of the workforce's rising education. What explains the remaining 80%? Employers' expectations have risen across the board. Waiter, bartender, cashier, security guard: These are now common jobs for those with bachelor's degrees.
Despite all these tell-tale signs of signaling, many of my fellow researchers refuse to take the idea seriously. Sure, signaling seems to fit our firsthand experience. Yet why would profit-seeking employers base their decisions on mere credentials instead of potential to perform well on the job?
To start, employers can't readily judge your job performance until they actually hire you — and they can only hire a tiny fraction of their applicants. If they ignore less-credentialed prospects, they may lose a few good workers but they save tons of precious time.
And once they hire, it usually makes sense to stand pat. Suppose a well-credentialed worker turns out to be mildly disappointing. Summarily firing him would be bad business, because replacement takes time, and time is money. A subpar worker may therefore profit from his credentials for years. Indeed, because few firms are run by unfeeling robots, even incompetent workers often enjoy handsome educational payoffs because their employers are too squeamish to dismiss them.

On the individual level, diplomas matter, skills, not so much
Education is a weird industry. You study arcane subjects year after year, knowing you'll never use most of what you learned after graduation. Yet parents, teachers, politicians and researchers urge you to finish, promising ample career rewards for your efforts. Despite the many college graduates who end up working as waiters, the experts are, on average, right: Diplomas pay well. What experts misunderstand is why. Instead of scrutinizing what schools really teach, they rush to a just-so story in which schools transform low-skilled students into high-skilled graduates. Students, much closer to the action, see what's going on: As long as they have good grades and finish their degrees, employers care little about what they've learned.

The author's suggested solution: Agree or Disagree?
[The author] advocates two major policy responses. The first is educational austerity. Government needs to sharply cut education funding to curb this wasteful rat race. The second is more vocational education, because practical skills are more socially valuable than teaching students how to outshine their peers.