Bowles & Gintis argue that it's very difficult to explain the education wage premium on the basis of things people learn in school. Instead, educated people earn more in large part because employers think if you could stick it out through school, you're likely to be a more docile employee and make more money for the boss. Of course this is an entirely separate question from the effects of racism and sexism on hiring and pay.
Greg:
Having gone from an undergrad major in literature to graduate work in poly sci, with 0 courses in poly sci, and thence to work for an engineering company for nearly 2 years, I have a somewhat different perspective on this issue. Yes, the "worker discipline" issues are significant, but if that were the only issue, than the same premium ought to apply to army retirees with no college at all. My hunch is that there are more potential careers than any one kind of training institute can provide. The university atmosphere really works by combining selectivity with a notion that if you toss people into an environment and make 'em work they get good at it. The process of getting good at something new is itself what is developed. I'm not saying that there is a 1:1 correlation and that there aren't capable people w/o college degrees. But if for example sixty percent of the college educated are better than typical non-college educated folks at learning new things, that is a significant edge, on average, from the employer's viewpoint.
An unrelated example is landlord use of credit card checks for determining who gets a pad. It is probably often, not always, but often, the case that people with bad credit histories have other kinds of problems that *might* make them bad tenants. Even if it works only 60% of the time, it might be a rational selection process from a landlord's point of view. There may be, for example, women who are perfectly good tenants and relatively solvent but for the fact that scuzzbag ex-boyfriend ran off with a card. But given a choice between the unknown hazards of someone solvent with no scuzzbag boyfriend history and someone solvent with scuzzbag ex boyfriend, the unspecified risks of association with scuzzbaginess may deter the hesistant landlord. The process is not fair, and perhaps should even be illegal, but triage methods of many different kinds a re a fact of life, in almost any kind of setting. -gn
-- Gregory P. Nowell Associate Professor Department of Political Science, Milne 100 State University of New York 135 Western Ave. Albany, New York 12222
Fax 518-442-5298