The Puzzle of Credentialism

Why do employers generally require college degrees or high school diplomas to get a job? On one hand, it makes sense: a more knowledgeable man who has demonstrated he can stick with a long-term project is a better employee. On the other hand, it makes no sense: why exclude the legions without such credentials when you can test their knowledge and diligence directly?

There is strong evidence, with Devin Helton having provided perhaps the best summary I’ve found in two blog posts, one from 2016 and one from 2017, that the vast majority of jobs don’t require a college degree to do the actual work, and a very large number of jobs don’t even truly require a high school diploma. Yet employers erect these barriers anyway. Why?

The Dark Side of Signaling?

Obviously the traditional “human capital theory”, that school is the one true place you can get the knowledge to do the job, is wrong, so more recently the idea that a degree signals pre-existing characteristics has been gaining steam. But I’m rather unconvinced by any of these theories.

Signaling theory perhaps proves too much, since literally any job requirement can be construed as being because those who meet it are signaling something or other. In addition, even Bryan Caplan himself seems reluctant to concede that signaling through a college degree might not be only for good-sounding traits like “intelligence” and “conscientiousness” but rather agreeableness and conformity because people who will sell themselves into six-figure debts and slave away for four years just because they’re told to are easier to abuse. The lockdown era has proven beyond a shadow of a doubt that sadism and bigotry is at least as much a driver of businesses’ decisions as profits are.

There has also been a striking lack of discussion about how credentials signal you’re from the social class employers are seeking to hire and promote, a fact that’s blatantly obvious in the case of the white-shoe levels of the management consulting industry, who basically sell a prestigious imprimatur for various plans, schemes, and doctrines and thus require their employees to come from the most prestigious backgrounds possible. In their case that makes perfect sense.

The Puzzle: why only College Degrees?

So signaling theory is obviously true to some extent, but even fewer people have raised the question of why college degrees are the all-dominant form of signal. After all, surely there would be many other ways of testing employees for intelligence, conscientiousness, agreeableness, conformity, or whatever else you want besides whether they have a degree, yet that’s about all we see. This despite an economic incentive to capture for your own firm good workers who don’t fit the mainstream’s requirements.

And it’s not like this general idea is some pie-in-the-sky concept; credential requirements often wildly differ by country. Take investment banking for example; in America the most prestigious banks recruit nigh-exclusively from a tiny fraction of elite schools, but in Switzerland it’s very common for their most prestigious banks to train people as apprentices from their mid teens, with even professionals at the CEO level often having skipped higher education entirely.

The military too doesn’t care as much about credentials, relying more heavily on aptitude tests, examinations, and short vocational courses than the equivalent parts of the civilian sector do. In the United States at least foreign service officers and air traffic controllers are also selected primarily through examinations.

Science contra College

Indeed, it’s striking that one of the few real studies that have looked into whether college degree requirements help anybody was done to answer the question of whether air traffic controllers would perform better if a college degree was required. The answer was no, they wouldn’t; there was not even any correlation at all between college attainment and better performance on the job.

There have also been a few studies done that have exposed “upcredentialing”, the practice of erecting new barriers to entering various career fields. In particular there are several fields where most new job listings require a bachelor’s degree while most of the current workers only have a high school diploma. We’re told the workplace has become so much more complex so more education is required, but if that’s the case how come the current less-credentialed employees can do the same jobs now?

Why isn’t Credentialism long dead already?

This same process has been creeping for over a century; in 1900 very few jobs required even a high school diploma, let alone a college degree, yet now we’re told within a few years two-thirds of all jobs (!) will need post-secondary training. This despite a technologically-driven trend of de-skilling, which one would expect would reduce, rather than increase, the skill level and thus the years of education required to do a given job.

After all, assembly-line workers didn’t need nearly as much education or training as their artisan predecessors did, so why hasn’t the same process continued to overtake the whole economy? Why did that go into reverse in the 20th and 21st centuries? Far from accepting credentialism as normal if anything we should be surprised and puzzled that the required credentials for jobs haven’t actually trended downward from the norms of circa 1900.

Zeitgeistwende

To me the whole thing is rather puzzling. But soon it might not be as relevant; although the turn of employers away from requiring college degrees seems to be more hype than reality it does seem to really be happening. The zeitgeist really has turned against mainstream higher-education institutions and credentialism in general in a way it perhaps never has before. College enrollment in the United States has been shrinking for a decade, with the pandemic only accelerating the trend.

In large measure this is because college is too expensive to be a ticket to prosperity anymore; when simply taking the money you would have sunk into tuition and buying a stock-market index fund instead provides better returns than earning a college degree it’s already getting sketchy, but now the college wealth premium has shrunk to literally zero, as in people aren’t making any more money on net by going to college than by not going at all. As tuition rises but the income of college-graduate careers does not rise it’s inevitable there will be a crossover point where “becoming a professional” means you actually lose money.

This is really what’s driving the current trend from the likes of Apple, Google, IBM, Ernst & Young, and quite a few others to dial back credential requirements. Perhaps the best illustration of the shift is how post-secondary training for all was emphasized as late as the Obama administration as the path to upward mobility, yet now less than a decade later Joe Biden, the man who served as Barack Obama’s Vice President, is now touting a jobs plan where the selling point is none of the new jobs requiring college degrees. To a large extent of course this is because the rise of Donald Trump and right-wing populists more broadly courtesy of the white working-class vote has scared the center-left to death, but nevertheless it’s a finger in the wind.

If not College, then What?

Obviously there’s a huge unmet demand for something superior to expecting everyone to go to college and hope it all works out for the best, but what? That is where ideas have been lacking, aside perhaps from copying the Swiss apprenticeship model.

The most insipid is the current zeitgeist for trade schools; “my father’s brother’s nephew’s cousin’s former roommate became a plumber and he’s earning an easy $100,000 a year!” is practically the background noise on Reddit these days, but as Devin Helton pointed out on his blog 85% of all jobs are not skilled trades. It’s also worth noting that most of them don’t even pay that much either, and it is hard strenuous work. Some people currently steered away from them would undoubtedly love those jobs, but most people just wouldn’t like them much.

Examination and Lottery: the true Solution for the post-College Era?

So if not any of the commonly proposed solutions, then what? I would propose an examination process, where employers have applicants take a standardized test that assesses aptitude for the job, coupled with a lottery, where if you score beyond a certain minimal standard of aptitude applicants are randomly selected.

This exam-and-lottery process, especially the latter component, might sound bizarre, but no examination is perfect; there’s relatively little correlation between the people who are the best test-taking machines and the people who perform the best in the workplace, so you don’t want to select heavily for them, but at the same time you don’t want to select people who are obviously unqualified, hence the lottery (to militate against the former) and the examination (to militate against the latter).

And weeding out the unqualified is about the best you can do; selecting the best qualified people in advance is something employers often at least claim to be doing but in practice it’s been amply demonstrated in study after study that workplace performance has basically no correlation with how employers think the worker will perform in advance of being hired.

Interviews and tests are stupendously powerful at weeding out people who, say, don’t know the first thing about the job or people who just can’t ever be presentable in a professional setting, but the finer distinctions cannot be teased out through such a process, so employers should really stop trying to do so and accept reality: that no matter what they do they have to take the risk of hiring and then having to fire a few bad employees to get a really good one.

Conclusion

You might have noticed a similarity between this and my proposal for college admissions, because they are basically the same idea. I personally believe the exam-and-lottery process has the potential to be a very powerful technique for admitting people into any kind of institution, capturing the bulk of the advantage of a pure standardized test system but without the primary drawback of such a system: incentivizing people to devote their lives to getting the highest score possible (the lottery component means there’s no incentive for doing so!).

Indeed, unless there’s some huge pitfall I’m missing it’s a bit odd that Dutch medical schools are the only place I’ve heard of that employs such a system to select people, as the concept seems rather obviously beneficial to me, far superior to the current system, and it seems to work well enough for the few places that do it.

So why isn’t it much more prevalent? Who knows? But if you’re reading this blog and you’re an employer I strongly urge you to at least consider it; all great social and economic innovations had their beginning somewhere, after all. Maybe this is another one that most everyone besides me just overlooked for all these years; but with the appetite for alternatives to college rising now might be the hour for such a dream of a truly meritistic employment system to take off.

Last but not least, I’m genuinely curious if any employers anywhere have used this hiring strategy I outline, and if so how it’s worked out, so if you have any knowledge of such a thing please leave a comment and let me know!