He has an outline here. I've been tangentially interested in the rational voter stuff, but I should probably actually pick this one up and read it more carefully. I agree marginally on some of this stuff, but have hesitations. One of the things that's bothered me about the way Caplan blogs about education is that he often treats human capital theories and signaling theories as an either/or issue. This outline seems more nuanced, which is nice.
The point I strongly agree with here is chapter 8. I think that's more because we've underinvested that and the way we do education here has resulted in some serious bifurcation in the skills distribution.
One of the things I wonder about his emphasis on signaling is that a lot of the things that are being signaled are actually traits you probably pick up in college. College helps turn dependent kids into independent adults, and it's those qualities that make an independent adult that play a large role in signaling. I'm not saying that's the only way to impart those skills, I'm just saying I wouldn't make such a strong case against formal education as Bryan does.