Part of a new paper on SSRN - thoughts appreciated

I'm done for now with the first section of the NBER chapter on the supply of new engineers. I've posted that and the introduction on SSRN here. The rest of the paper should be quicker to write - it's in a detailed outline, but I didn't include it here.

Any thoughts would be greatly appreciated. The audience is intended to be fairly broad - not just economists. Also engineers and policymakers. So I tried to explain things like self-selection models in clearer, non-technical language. If you can think of any way I can improve that I'd appreciate it.

Thanks for any thoughts. Of course I post more as I get it cleaned up.