It’s time for the start of another school year, which means that, along with sales of funnels and plastic tubing, it’s a good time for thinkpieces about the state of our education system. One frequent theme of these perennial exercises is the rising cost of tuition at 4-year colleges and the subsequent interminable debate over whether or not, in this day and age, a college education is “worth it.” At first the click bait-y thing to do was to argue that a the high price of a bachelor’s degree, combined with the high unemployment rate for recent college graduates, proves that college is no longer a good investment. These kinds of takes became so popular that the pendulum swung back the other way, and it started to look like you were bucking conventional wisdom if you made the case that a 4-year degree actually is a good investment.
But these kinds of stories are driven by people looking for an argument to back up what they already believe, so it’s easy to see why people would be eager to draw conclusions beyond what’s actually supported by what we know. Because the true answer is that, even with all the data we have about education and incomes, we have no idea whether or not college is “worth it” at the price it costs. And I mean this even if we treat a bachelor’s degree as purely an investment in one’s future earning power and productivity, excluding any consideration of the other ways in which an education might be valuable. Even if we measure the value of a college education completely in financial terms, there’s still no way tell whether the amount that it costs now is too much or too little.
There are two basic issues that prevent us from having clear data about the value of a college degree. First, although college graduates on the whole earn much more money than non-college graduates over their lifetimes, we have little reason to think those increased earnings are caused by having a college degree. Matt Yglesias of Vox has ably debunked this argument, pointing out that saying college is worth the expense because college graduates earn more is like saying BMWs are a good investment because people who own BMWs tend to make a lot more money than people who don’t.
Because Yglesias does such a good job pointing out this statistical fallacy that I would hope anyone with a college degree could spot, I’ll spend a little more time explaining the second reason we don’t know whether a college degree is worth it or not, which is this: even if we could prove that the average student more than made up for the cost of their degree in future earnings, we still wouldn’t know that the money we spend as a society providing higher education is a good investment..
To see why this is, consider the question of who ought to receive a college education as simply a management issue. I think the obvious answer would be that someone should receive an education if and only if the amount that it will increase their productivity is more than the amount it costs to educate them. In our society, however, decisions about how to allocate higher education are not made by a single authority. Instead, it’s mostly left up to each individual to decide whether to invest in a degree to boost his or her earning potential. Given the existence of a gap in earning between degree holders and non-degree holders, more people pursue degrees in pursuit of higher wages. This competition pushes down the wages of non-degree holders, which in turn increases the wage premium workers expect to get from their education, encouraging more people to pursue degrees. But all of this could be true even if degree holders are only scantly more productive than non-degree holders. As John Cassidy put it in the New Yorker, this can turn education into more of an arms race than a productive investment.
Consider an analogy to illustrate how this might be the case with college education. Imagine someone invents a technique to make doughnuts more delicious, and it costs $1000 to buy the machines to make new, more delectable pastries. Imagine there are 100 bakeries in our market, each of which has to decide whether or not to invest in the new doughnut technology. At first, new doughnuts are selling for much more than old doughnuts because few people are producing them and some people have the extra money to spend on a doughnut that might be only a little bit better. This gives many bakeries the incentive to invest the $1000 to switch over to new doughnut technology. As new doughnuts become more prevalent their price goes down, but so does the price of old doughnuts, because consumers buy about the same number of doughnuts on average and everyone prefers the new variety. This means that it’s still a good investment for each bakery to buy the new doughnut machines, even if they can only sell new doughnuts for the same price they used to sell old doughnuts for, because now old doughnuts sell for much less. So each of the 100 bakeries spends $1000 on new doughnut equipment and old doughnuts more or less go extinct. It’s easy enough to imagine this happening, even if the new doughnuts are only a little bit better than the old kind, but from a public choice point of view, it’s only a good use of resources if the total social value derived from everyone eating new doughnuts rather than old doughnuts is worth $100,000 or more. For our story to be true, new doughnuts have to be better, but they don’t have to be $100,000 better.
In the same way, clearly people who have a bachelor’s degree are, on average, somewhat more productive than they would be without one. But the question we need to answer as a society is whether they are, on average, $50,000 or $100,000 more productive. That’s a much dicier proposition. If we knew that we were talking about a competitive marketplace we could draw some conclusions about how an overproduction of college graduates would work itself out, but higher education is almost the definition of an imperfect market. People making the decision to go to college can have only the foggiest idea of how it will impact their employability after graduation, and firms hiring college graduates have only the most general idea of a correlation between bachelor’s degrees and employee productivity.
There are a lot of specific issues facing higher education today, and many of those issues have dedicated advocates on one side or another. Rising tuition costs are one of those issues, but it’s ridiculous to suppose that we can treat a college education as merely an investment in human capital and cut through the all the substantive disagreements by just drawing up a balance sheet. Colleges and universities are a cultural institution in this country, while at the same time being a means of vocational training and an important force influencing the character of younger generations. We shouldn’t focus on one aspect of education and ignore the others, but even on that one aspect we don’t know enough to draw a definitive conclusion. More information is good, and we should welcome things like the Department of Education’s College Scorecard, which attempts to bring more data to the public broken down by specific schools and programs. But we have to remember that data-gathering is the middle part of problem solving, not the beginning or the end. information by itself is no substitute for determining the problem correctly or making justified inferences. And I would hope that point is something a highly educated population can appreciate.