Before this conversation ends, I’d like to press Charles Murray and Pedro Carneiro on signaling:
For Charles: In your view, why precisely does the market financially reward students for taking lots of classes that at best seem distantly related to job performance? You don’t seem ready to sign on to my signaling story. Do you have an alternative? As I’ve said before, I agree with most of your conclusions. It’s your model that leaves me wanting more.
For Pedro: Yes, I am familiar with Fabian Lange’s work on signaling. But to deny the importance of signaling in education goes against decades of my first-hand experience. When I mentally review my 21 years as a student, I just don’t see that many of my classes caused my marginal productivity to go up. Foreign languages, history, physics, physical education… in my job, they rarely come up.
You could say that I have an unusual job. I do — I’m an economics professor. But that just reinforces my point. Since a major part of my job is teaching the material I learned as a student, there is an unusually strong connection between what I learned in school and what I do in my job. If I hadn’t become an academic, I probably wouldn’t be writing about the return to education and calling it “work.”
The other response I’ve heard to my skepticism about the practicality of education is that so-called “useless” subjects improve job performance by “teaching us how to think.” But educational psychologists have been testing this hypothesis for over a century, under the heading of “transfer of learning.” (See Haskell 2000 for a survey). The punchline of this massive literature is that learning is highly specific; if there is such a thing as “learning how to think,” it occurs too rarely to see it in the most of the data. 
So my questions for Pedro are: If you mentally review your years as a student, can you honestly say that your typical class raised your marginal productivity? If you adjust for the fact that you’re an academic, doesn’t the weakness of the connection between what you learned and what you do trouble you?
 There is another major debate in this literature about whether we can improve education so that transfer of learning does occur. But virtually no one in educational psychology claims that significant transfer of learning is occuring now.