Common Sense and the Marginal Student
Common sense says that marginal consumer of a product – the person who say “Eh, why not?” before he buys – benefits less from his purchase than the product’s average consumer. This is clearly true for consumption decisions: The typical person at the opera clearly enjoys the show more than the reluctant newbie whose friends talk him into going.
This principle – the marginal person gains less than the average – is equally plausible for investment decisions. The typical college student will earn a larger return from his education than the kid who grudgingly enrolls to get his parents off his back. When I share this claim with practicing labor economists, however, they’re quick to rebuke me. Don’t I know that instrumental variables methods show that the marginal student gains at least as much – if not more – than the average student?
My main technical objection is that labor economists inappropriately focus on completed education rather than attempted education. But that aside, I simply consider common sense more reliable than instrumental variables. Fancy econometrics reduces my confidence in my original position, but only slightly.
Now just yesterday, I learned that Heckman and co-authors have a new NBER working paper vindicating the common sense view that marginal students earn a lower return. Their evidence against all the fancy econometrics is… even fancier econometrics:
This paper estimates the marginal returns to college for individuals induced to enroll in college by different marginal policy changes. The recent instrumental variables literature seeks to estimate this parameter, but in general it does so only under strong assumptions that are tested and found wanting… Our empirical analysis shows that returns are higher for individuals with values of unobservables that make them more likely to attend college.
It would be tempting, but dishonest, to claim that the latest research “proves me right.” I was convinced that school helped the marginal student less years before I heard about Heckman’s results. I doubt that reading his research in depth will make me any more confident. Yes, I’m glad to see Heckman lending his Nobel status to the common sense position. But in a perfect world, the common sense position wouldn’t need his status to survive.
P.S. Dan Klein and I are speaking in Madison, Wisconsin tonight (Monday 10/25). If you make the talk, please say hi.
The post appeared first on Econlib.
A few thoughts here:
- Your description of the "marginal student who just signed up to college to get his parents off his back" looks very different to the "marginal student whose parents only just managed to scrape enough money to get their daughter to go to university instead of an early arranged marriage". Whichever represents the marginal student depends on your context.
- There's an obvious story where the former kind of marginal student gets less or even negative value from higher education. They can't be bothered, they learn less than the average, they feel less competent, (possibly) get in debt and are more likely to drop out.
- There's also a plausible story where the positive effects of higher education for the marginal student involve "entering the middle class", which might mean marrying and having kids later, getting a peer group of slightly smarter friends who are a bit more ambitious, drink less, take fewer drugs, and are less likely to get into crime etc. This effect is greater for the marginal student because the average student is probably already in the middle class.
- There are a bunch of RCTs that look at marginal students who would only continue education given a financial nudge. e.g. Pascaline Dupas looked at prospective high-school (16-18, I think) students in Ghana. Ten years later, the treatment group were doing a lot better in the obvious ways - later childbirth, office jobs, more financial stability; the control group were far more likely to have gone back to their village to farm. My sense was that the marginal student benefitted more than the average student (in comparative rather than total terms), who was middle class to begin with.
There's clearly some common sense on each side of this debate, and it wouldn't surprise me that you get different results with subtle variations in the context.
Nice example of the usefulness of basic price theory in understanding the world.
Note that it can even help you see where your econometrics wasn't fancy enough, because it gave the wrong answer.