...For the same reason we are over/underestimating all sorts of things: Correlations don't mean causation. Even when they do, subsidizing the cause can break th

...For the same reason we are over/underestimating all sorts of things: Correlations don't mean causation. Even when they do, subsidizing the cause can break the causal link. Taking english classes and learning english might be causally related, but paying people to take english classes may not have a predictable result. The current disadvantage=discrimination debate has the same problem. So do many of the Scandi-inspired policy suggestions, assuming that policy X caused result Y, and will again.... If we apply this logic in other places we'd conclude that homosexuality increases salary. Evidence for these things is probabilistic, and we suck at statistics, especially when we're being political. Anyway, I found this interesting (by ommission): Part of the reason why university graduates earn more is because they are brighter and harder-working to begin with Not that they were wealthier to begin with? Not for class reasons? and... humanities graduates, ...tend to earn more if they come from more prestigious institutions suggests that one reason ... get ahead of peers in the job market. Get ahead how? Knowing people who are ahead? Being associated with people who are ahead? Talking like people who are ahead? They already conceded that it isn't the education.... Humanities (and educations generally), was the domain of the educated classes. Lower class people sent their kids there for class mobility reasons. It was a way for the more talented middle class kids to join the upper-middle classes, by joining them physically. Kids from those classes, went to college to strengthen the class association. In the UK, they were generally explicit about this. The economist knows this. Why beat around the bush. Higher class seeming people (yes, this still exists) do better in job markets. ..also in marriage markets, probably get loans easier...