College and the job market

Can a poor job market cause rising college attendance?

Of course. Especially when financial aid is easy to find. If someone can’t find a job but can easily finance an education that improves their value in the job market, they’ll be more likely (on the margin) to go back to school and put off their job search until later.

But to what extent does rising college attendance indicate a poor job market for young and young-ish people? For example, is the rise in college attendance over the past seven years mostly due to a poor job market for 18-30-year olds?

I think that could definitely be true, but the key word is “mostly.” We’d have to know what percentage of students wouldn’t have gone to school had they been able to more easily find work instead. Do these students account for more than 50 percent of recent increases in college attendance? I doubt it, but I really don’t know.

I do know, though, that the number of people attending college has steadily increased over time. And I have every reason to believe that a long-term upward trend in per capita GDP (which exists) should correlate with a long-term upward trend in the number of people attending college. I’ll call this the “natural” rise in college attendance over time.

The key to knowing whether a poor job market for young and young-ish people since the 2008 financial crisis is a big reason for increased college attendance is finding some way to subtract the portion of the increase caused by this “natural” upward trend from the total increase in attendance. The number we’re left with should theoretically equal the number of students who would rather be working than in school, but could not find work and so chose school instead.

But even that is pretty iffy, because whether one wants to be in school or not depends, in part, on the differences in the type of work available to degree- and non-degree holders, which in turn is partly determined by the health of the labor market. It’s all very confusing.

Regardless, this has implications for discussions about the true value of figures like the unemployment rate, which I’ve seen attacked lately. The unemployment rate doesn’t include people who’ve quit looking for work as “unemployed”, so the story goes, and thus is really just a big lie.

In more precise terms, it’s the drop in the labor force participation rate (LFPR) that is not well-reflected in the unemployment rate. In fact, dropping out of the labor force entirely (that is, unemployed people quitting their job searches) lowers the unemployment rate, ceteris paribus, by removing people from the equation altogether. For example, if the labor force includes five unemployed people and 105 people total, those five people dropping out of the labor force would leave 100 people employed and 100 in the labor force—a zero percent unemployment rate.

But a falling LFPR doesn’t happen only when people get so discouraged that they quit looking for work and “drop out of the labor force.” It can happen for several reasons, of which a “natural” rise in college attendance rates is one. I discuss other reasons, like an aging population and planned early retirement, here.

All this to say, things like rising college attendance and a falling LFPR cannot, in themselves, be considered a gauge for the health of the labor market. These metrics are influenced by a variety of factors, and we shouldn’t necessarily be suspicious when they move one way or another. I do think college attendance is higher and LFPR is lower than levels we’d see if we hadn’t had a recession, but knowing whether the portion of these figures’ movement directly attributable to a poor job market is significant is a matter of subtracting their “natural” movement from the total movement—a difficult task, indeed.

Posted by Nick Freiling

Founder/Director of PeopleFish. I write on technology, market research and economics. Bylines at Startup Grind, FEE, the American Enterprise Institute and the Mises Institute.

Leave a Reply

Your email address will not be published. Required fields are marked *