Suppose college graduates out-earn high school grads by $30,000 a year.  Naive analysts will tell you, “Finish college and you’ll get a $30,000 raise.”  The clever, however, will warn you about ability bias.  The kind of people who become college grads usually have pre-existing advantages – intelligence, work ethic, conformity, and more.  Some of the $30,000 earnings gap reflects the pre-existing advantages college grads bring to the table, rather than their education.  In The Case Against Education, my best guess is that only 55% of the earnings gap is causal.  Given an apparent college premium of $30,000 a year, your true college premium is 55%*$30,000=$16,500.

Now suppose that as a result of going to college, you end up marrying a fellow college graduate.  How much does this raise your expected family income?  Assuming your spouse works, the tempting answer is, “$16,500 – the same as it does for me.”

On reflection, however, this is a serious mistake.  When you marry a college grad, you wed the whole college package – extra education PLUS the pre-existing ability that typically accompanies that extra education.  So your family income doesn’t rise by 55% of $30,000; it rises by 100% of $30,000!  Key insight: College graduation helps you marry into the high-education, high-ability pool, changing the identity of your spouse.

The precise mechanism is immaterial.  Maybe you directly meet your spouse in college.  Maybe your college degree gets you a job where you meet fellow college grads.  Maybe college grads only date their own kind.  As long as your education somehow causes you to marry a college grad, you’re in the money.

Notice: This argument for ignoring spousal ability bias does not apply if you are already married.  When you tie the knot, your partner’s pre-existing ability changes from a variable to a constant.  So if your current spouse decides to pursue a college degree, you should only expect your family income to rise by $16,500 after graduation.

In the real world, of course, there’s a large (but shrinking) gender gap in the marital return.  The average male gets a bigger absolute payoff in the labor market, and therefore gets a smaller absolute payoff in the marriage market.  The average female gets a smaller absolute payoff in the labor market, and therefore gets a larger absolute payoff in the marriage market.  As I’ve previously explained, though, switching from two one-person households to one two-person household saves so much money that even the higher-earning spouse normally profits from marriage.

In any case, most people plan to marry someone.  Given this preference, the marginal payoff of your education in the marriage market would be large even if marrying per se didn’t save you a dime.  If you’re going to split your income with another person, you want that person to be rich.  And in our society, extra schooling is one of the most effective ways to make that happen.

Lest I be misunderstood: I’m not claiming that gold-digging is a common motive for education, and I’m certainly not advocating gold-digging.  My claim, rather, is that regardless of their conscious intentions, the well-educated strike even more gold than you’d think.