First, to Tom Kruse, an excellent little text on inequality and discrimination in the labor market, covering both neoclassical and heterodox, is Albelda, Drago, and Shulman, _Unlevel Playing Fields_. I don't know if you are looking for something for classroom use or not, but this is perfect for undergrads.
Regarding Kliman's questions about the human capital studies, yes, it is true, that "Nowadays, it seems . . . that many of them don't even bother." Many of the neoclassicals argue that any wage differential left over after all measured HC is accounted for is not really discrimination, but is due to unmeasured productivity characteristics. [Please excuse me if any of this has a snide tone to it, I find this line of arugment so appalingly wrong, I can't help it. If neoclassical economists took some of their skills and applied them to actually looking at the world around them they would notice that capitalism creates inequality rather than elminates it.] These "unmeasured" characteristics are things like school quality, "soft skills," etc. Thus, even if African American's have equivalent schooling, it is of lesser quality and therefore this explains their lower earnings at similar levels of education.
Cheers, -- Heather Boushey, Ph.D. Research Department NYC Housing Authority 250 Broadway, Room 711 New York, NY 10007 Phone: 212-306-3372; Fax: 212-306-6485 hboushey at csi.com