This is a video of Richard Freeman speaking on "Innovative Measures of Innovation" from August. I don't really get into this stuff, although if one of my dissertation chapters is on new growth theory and I do anything empirical with it I may have to. As you might imagine, it's a tough thing to get at. Almost everyone knows that patents are pretty useless for this purpose. Bibliometric statistics (publication and citation) seem like a non-starter for the exact same reason as patents, but that is fairly popular right now.
Freeman's talk with Paula Stephan at the American Chemical Society, which I listened to online, went very well. They mentioned a recent National Research Council report on the science and engineering workforce employed by the Defense Department as an example of how views are changing on the "shortages" assumptions. That report used some graphics I put together on the relationship between computer science majors and IT wages and unemployment (no fancy analysis, but pretty clear descriptive statistics showing that students respond to wage signals).
Speaking of no-fancy-analysis-but-telling-descriptives, it looks like I won't have time to get the labor supply elasticity estimates into the petroleum engineering chapter. I have too much work to do to get it out in time and haven't touched it since the summer. That's alright, because we have lots of interviews with faculty and administrators from petroleum engineering departments and other good info in the chapter. That means that the labor supply models could go into another paper over the next couple years.