Why the college bubble won’t pop

From the American Conservative: The College Bubble Won’t Just Pop

Although taken for granted by millions of Americans, the choice to obtain a college degree mystifies economists. Teenagers making the first significant financial decision of their lives are paraded through prospective student tours—essentially timeshare presentations for minors—and finally make an arbitrary pick where they will spend around a quarter-million dollars of borrowed or parental money. For some, the degree results in open doors and respectable careers. For others, it’s a scam that wastes their time and money.

To say something is a scam implies that the buyer derives no benefit. However, the data still shows that college grads earn more than non-grads, and have a much lower unemployment rate, and that this wage premium holds even in spite of inflation and student loan debt. And it holds for virtually all majors and institutions (although it’s highest for STEM degrees and top schools). So it’s not mystifying or irrational that college continues to be popular given that it still yields definite benefits. Perhaps college could be a scam if someone fails to graduate because they derive none of the wage benefits, but the debt of having attended. Also, the $250k figure he cites is wrong, being much closer to $30k/student. For some reason, perhaps for dramatization, the media and pundits keep citing these 6-figure sums when the most basic, cursory research shows this to be a huge overestimation.

But I agree that the bubble has resisted all predictions over the past 12 or so years of its demise, as I correctly predicted many years ago. I also predicted that Covid and remote/online learning would not put a dent in the wage premium either. My guess is, this is due to a combination of the loans being federally backed, but also because college is still mutually beneficial for both parties: employers get better employees, and grads make more money even if tuition keeps going up. For the bubble to burst would require that one of these fail. Somehow employers would have to realize that they are vastly overpaying. But given how obsessed companies are with profits, makes this even less likely. Labor is a major cost, and it’s not like employers want to spend more on employees than is absolutely necessary. When a company like Google pays $300k/year for an engineer, it’s because if they didn’t the engineer would go elsewhere, not because Google is feeling really generous. When a company is struggling or if there is possible trouble on the horizon, the first step is always layoffs, so keeping labor costs low is a top priority. It’s not going to come from defaults, like the 2008 subprime crisis.

The con game’s most important element is its referees: the managerial bureaucrats who fill the cardboard tube received at graduation with its “fiat” value. As long as the diploma earns the respect of hiring recruiters, graduate faculty, and other industry functionaries and patronage networks—and as long as the system remains under the thumb of such gatekeepers—colleges will serve a similar function as the Uchraspred of the old Soviet nomenklatura. For young people without money or clout, an endorsement from the politburo managerial class is crucial for future career success.

The common narrative is that the degree does not signify any special intellectual or meritorious ability of the recipient. It’s just a very expensive ticket, as the argument goes. So employers are voluntarily paying a huge wage premium for this hard to quantify attribute(s). In my article last week, I pushed back against this assumption. I think these are mutually inclusive: college is a ticket to the middle class, but at the same time, it signifies above average ability, too, not just conformity, contentiousness, or obedience.

College grads, especially from decent schools, tend to more competent overall at many things, and even if far from perfect (cue the articles about how college grads can’t write well), there is still a cliff-like differential or gradient of ability between college grads and high school grads, especially given how inflated high school GPAs are (much worse than college grade inflation). Even in spite of supposedly dumbed-down courses, the 4-year college dropout rate is 60% (and 40% for those who somehow need six(!) years to finish), whereas the high school graduation rate is approaching 90%. If college grads cannot write that well, do you think the high school grads will fare much better?

So we’re talking a 6-fold difference (10% dropout rate for high school vs. 60% for college), which is huge. Conditional on graduating from high school, the overall college graduation rate is 36% of the adult population. This is about half a standard deviation of ability (z=.5), which isn’t much as far as outliers are concerned, but is a powerful filtering mechanism nonetheless given that just 40/100 college students finish vs. 90/100 high school students.

In a recent article for RealClear Education, I argued that Republicans can use their majorities in state legislatures to restore a job-focused middle-class mission to state college systems. Such action could differentiate institutions offering practical training from those offering ideological indoctrination. It would only be a first step, as public colleges account for less than half of student enrollment. But if these colleges begin providing education with objective value, they would become a clear alternative to our current system of credentialism.

Republicans have completely caved to the left as far as higher education is concerned. Two possible solutions include a bipartisan bill that makes it illegal for employers to require degrees unless it can be demonstrated that the the degree is necessary, such as being a doctor. Or, better alternative/substitutes for degrees. The second suggestion is obviously much more viable than the first, but probably much less effective. If businesses derive value from hiring only college grads, then that ought to be their right, is the argument.

But the higher education monopoly on career credentialing cannot continue forever on the backs of a tiny network of elite schools. Nascent alternatives to the average four-year degree are already cropping up. Some, like the Bloom Institute of Technology, offer skills certificates in programming and data science that charge tuition only as a percentage of a student’s first well-paying post-graduation salary. This “payment on delivery” model eliminates the need for student loans and prioritizes personal responsibility over limitless government subsidy. Others, like Praxis, help young people launch their careers through paid apprenticeships and on-the-job training without the need for college.

Creating more alternatives to degrees won’t change anything if those alternatives are inferior and or if are not adopted by companies. Certificates are inferior to degrees in this regard. Degrees take much longer, have way more courses, but also the courses are harder. A physics course will have difficult problem sets and exams, for example. From my own and other’s experiences, I can attest that college courses are harder than videos. Watching 3blue1Brown will never compare to actually doing problem sets. Certificates cannot compare to the timed, high stakes college environment. Hence it’s not surprising that most good-paying jobs have not embraced certificates or other alternatives. Google may claim that their career certificates takes 6 months to complete and that some recipients find employment, but what about overall success rate instead of just cherry picked examples? Good luck with that.