In last Sunday’s NYT, economics columnist David Leonhardt had a piece entitled “The College Calculation.”

It starts off well enough, posing the question of “whether the college makes the student or the student makes the college.” To be more precise, does college education add to an individual’s human capital and thereby make him more productive? Or do innately more capable and ambitious people tend to go to college, and does the fact that they tend to be more successful have little to do with their education? How much does college education, Leonhardt asks, “really matter”?

Leonhardt glances at some evidence and arguments on both sides, but it’s evident that he’s in the “yes, it really matters” camp. At the end of his piece, he writes, “at the very moment when education would be of most value, money for it is disappearing.” Well, the money is not “disappearing,” but more to the point, how do we know that education — presumably he means increasing the number of Americans enrolled and graduating — would be of “most value” during our current recession? Leonhardt merely assumes the point he’s trying to prove.

It’s an article of faith for many people that having more formal education under your belt means that you’ll do better, and that the higher the level of “educational attainment” in a nation, the more prosperous it will be, but both of those propositions are highly questionable. Many people with college degrees end up in jobs that most high-school students could easily learn — unfortunately, Leonhardt does not look into that phenomenon. Nor is it true that a country can pull itself up by the bootstraps economically by increasing its “investment” in higher education, since most of what people need to know in order to succeed in their work is learned on the job, not in classrooms.

Leonhardt also completely misses the screening argument. He thinks it telling that a study of identical twins where one went to college and the other did not shows higher earnings for the college educated. That, however, does not prove the “human capital” hypothesis. Employers tend to only offer better-paying jobs to people with college credentials not because they’re necessarily more capable, but because the degree is an easy (and legal) way of sorting people. As more and more employers have taken to using college degrees as a screening mechanism, fewer and fewer job paths leading to high incomes are open to people without them.