Is it still worth it to go to college?

In recent years, this has become a contentious issue. The latest volley comes from Bloomberg, which pointed out last week that the "college premium" — the difference between what workers with a bachelor's degree or higher earn, versus workers with just a two-year degree, some college, or just a high school diploma — was as high as $17,500 for younger workers in 2013. That's according to research by Pew, which also found that the college-educated can expect to make around $500,000 more over their lifetimes.

The pushback tends to come in two flavors: that the "college premium" is undone by massive student debts, and that college promotion is just a product of navel-gazing elites thinking the way to fix the country is to make everyone else more like them.

Neither gripe is correct, though the second arguably gets closer to the truth. The data is pretty clear that even after costs and debts are accounted for, college degrees are still a great bet for the people who can get them. But it’s important to understand the underlying economic mechanism that produces the college premium. From there, you can see what’s wrong with suggesting we must "improve" workers with more education, and why promoting college degrees as an answer to stagnating incomes is so fundamentally weird.

Let’s start with a thought experiment: What would happen if everyone in the country got a four-year college degree or higher? Would we then be a nation of well-paid tech-industry yuppies and creative-class artisans? Of course not. Toilets and hallways would still need to get cleaned. There are plenty of noncognitive, manual labor jobs that are difficult if not impossible to automate. So we’d have lots of janitors and such with bachelors degrees — the most well-educated working- and service-sector classes ever.

But there would be no intrinsic reason these workers would make higher incomes. What drives the college premium is supply and demand; even today, only about one-third of the American population has a bachelor’s degree or more. And work by M.I.T. economist David Autor does indicate the demand for college-educated workers has been outpacing our ability to supply them. Hence the higher incomes.

If everyone is college educated, the script flips: The supply is as big as it could possibly be, so there’s no reason for employers to pay a premium for it. Education level becomes a moot point.

The real problem is that, while improving education might help low-education workers' position vis-à-vis high education, it’s not clear how it helps all workers bargain with capital and business owners. That’s the real trick. As income inequality scholar Branko Milanovic pointed out, the popularity of the term "human capital" — the term of art often used for a workers' skills and abilities, often provided by a college education — has served to obscure the basic conflict between actual capital and workers. People who own capital don’t have to work to get their human needs met. They can just sit on their laurels and their capital produces an income for them.

Workers, by contrast, must labor to get an income and thus meet their needs, no matter how great their "human capital." This creates an inescapable power imbalance, which in turn decides who gets what slice of total economic output.

The fundamental point is that higher incomes don’t arise from a worker’s education level, they arise from how their education level is positioned vis-à-vis other factors in the economy: supply, demand, bargaining position, and so forth. Change those circumstances, and the college premium changes. As my colleague Ryan Cooper argued, promoting college as a path to a higher standard of living rests on the fallacy of composition: "Education surely increases productivity [of the economy as a whole], but it has little or nothing to do with who gets those productivity gains."

If we're looking to change the arrangement of factors in the economy to drive up low-income wages, there are lots of ways to skin the proverbial cat besides education.

We could alter immigration law to bring in way more high-skill, high-education people from other countries. Just like educating more native workers, bringing in immigrants would increase the supply of high-income workers relative to demand, driving the costs of that labor down. That would make the goods and services those high-education workers provide cheaper, making everyone else's incomes go further. We could also alter regulatory policy, monetary policy, and other factors to drive up the supply of jobs, which would increase worker bargaining power and drive up wages.

But, as Cooper also notes, it’s really difficult to run the economic engine hot and hard enough to maintain a tight labor market over any extended period of time. So we should also beef up unions and the social safety net to bolster workers' bargaining power and put a floor under their incomes even in the troughs of the business cycle.

In sum, there’s nothing "wrong" or "lacking" with non–college educated workers that leads inexorably to their lower incomes. Rather, their incomes are the result of how we, as a society, have collectively chosen to arrange the economy around them.