Return to Revisions

I think the why part is easy. University administrations like cheap labor and are risk averse when it comes to giving permanent jobs, and the people who already have permanent jobs like cheap labor because then their teaching load goes down. The real question is why people take post-docs. I think the answer is that people really want to do good mathematics and they think (often correctly, often not) that having access to the top people will help them do that, so they're willing to accept all the sacrifices that come with post-docs.