The oversupply problem is one that doesn't seem to be going away ay time soon, especially in the pure sciences. From my anecdotal evidence, it seems to be harder and harder to get into academia or cross into industry each year. I have multiple bio and chemistry PhD friends who have been either stuck in limbo at a postdoc position or who have left the field in frustration. I wonder if it will correct over time, or if the glut or PhD's will continue to grow.
Meanwhile, there's been a trend of major corporations closing their R&D facilities because they aren't as sexy on the quarterly balance sheets, which clearly isn't making things better.
CS and Econ PhD programs both benefit from an extreme amount of cross-pollination between academia and industry. The ability to jump ship from the academy isn't nearly as large of an issue as it is in the sciences (with the exception of physics perhaps). The humanities are mostly screwed.
Not from what I have read or heard. There still seems to be a need across the board for those with deep knowledge, especially when it comes to certain topics. The people I have known who have had trouble, and studies I have read about the topic point mainly to biology and chemistry as the two hardest fields to break initially with a PhD. But of course even that could change depending on specialization, etc.
Among those doctoral graduates who went to North American
industry and for whom the type of industry position was
known, about 60 percent took research positions .
>I largely agree with this, but in some fields I also think it is an over supply of PhD problem as well.
Looking abstractly at the "market", you are right. Asking a person on a tenure track committee why they're forcing newer hires to do a lot more than they ever had to do to get tenure, and getting that answer, is problematic.
Nothing in the market is forcing the hand of the person on the committee. Nothing. It is the equivalent of "I'm being a nasty jerk because I can." The market helps in him being nasty, but it is his/her choice to be so.