4 ms·
"The Case Against Education" by Bryan Caplan suggests a framing in which this makes sense. He more or less argues, with data and some interesting points, that t
by SeanAppleby 8y ago
"The Case Against Education" by Bryan Caplan suggests a framing in which this makes sense. He more or less argues, with data and some interesting points, that the barriers to education are not a bug, and not only a feature, but perhaps the primary economic feature of our education system.
A part of his argument is that people who get into elite schools and don't go to an elite school tend to have approximately the same life outcomes as people who go to elite schools, suggesting that the schools are just filtering for people who will be successful anyway and rubber stamping them.
Another part is that, if education's economic value was through adding value to people by exposing them to valuable ideas and skills, then there should be some correlation between how deeply exposed you are to those resources and your life outcomes, but in reality there's really only a binary between whether or not you finish the degree. 7 semesters of college seems to do nothing significant for you if you don't get a piece of paper to show people, which is very strange if college actually adds any substantial value other than as a filter.
There's more too. It's a polarizing book, but I think his framing is at least approximately correct, as it fits better with how the real world is working than any other framing of education I've seen, and if he's right, then the current paradigm of education is just fundamentally not scaleable, and really needs more fundamental restructuring than just giving it to more people.
- wallace_f 8y agoI've heard many economists echo this argument. People who have the opportunity to attend an elite university and decline it would be doing so because their opportunity cost is better. But the question is still open: why should credentials not be open to fair competition? Shouldn't a poor woman in Africa be able to compete for the same computer science credentials, bar exams, medical exams (this would require further vetting of practical lab skills) as a Harvard student, through the same exams and projects? Of course, Harvard could still give its students its own stamp of approval, but the education and credentials should be open to competition, not behind walled gardens.