The problem is not that there are too many PhD. The problem is that there are much less jobs outside academia, in North America specifically, that require a PhD degree. In the past many bright PhD went not to academia but to research divisions of big corporations, where they contributed to development of novel innovative products and made good science at the same time. In my area lots of fundamental work was done by people from places like IBM, RCA, etc. And where is RCA now?
Now corporations do not do R&D. They watch the bottom line and move production to China and other countries. Little wonder there is such a demand in China, the R&D jobs follow the production facilities. It makes little sense to keep an R&D department 10,000 miles from the production. So these jobs disappeared.
Now the new breakthrough research is supposedly being done in universities, by people like me (I am a mid-career prof in Canada, materials research). However, I am forced to work with graduate students and (very rarely) postdocs because I do not have sufficient funding. This is partially a Canadian problem where an average grant size is slightly over $30,000 per year. Well, most of us have more than one grant, so we survive, but there is no way we could create permanent scientist positions, there are neither sufficient funds nor indeed room in the university structure to accommodate such kinds of jobs. Our main granting agency in sciences and engineering in fact specifically forbids funding of a pdf position for more than two years from its funds (!). This lowers dramatically the productivity of my research. Having experienced stuff scientist positions available at the universities would enable me to increase the efficiency of my research (and investments into my group by the granting agencies). This would also solve, at least in part, the problem of having not enough academia jobs for all recent PhDs. Now I have to take new students because I need people to work on my projects, but as soon as I have trained them, I have to kick them out and start all over again, while they are hard pressed to find good jobs. I, or my colleagues, do need them, we would have taken them, and our research would benefit enormously, but we have no way to do so. We would also then take less graduate students thus reducing the glut of PhD graduates we have to produce to keep our funding and increasing the overall PhD level, which indeed becomes somewhat low in recent years.
There are a few professors in my Department who are somehow able to support stuff scientists, but only because they have long-term industrial funding. However, these are really exceptions. There have been comments on another thread about a professor who was doing the same and having hard time justifying her budgets for grant agencies. The university system in North America is not equipped for this. And this may be a bad thing because the current level of complexity of the cutting edge research requires more, not less qualified people to develop novel breakthrough technologies. So the fault is not with how we train PhDs, the fault is how we fail to use them properly.
