Kevin Lang and Michael Manove offer very interesting evidence about the links between ability, education, and race in their new American Economic Review article Education and Labor Market Discrimination. My personal excitement about the article comes from the mathematical theory Lang and Manove present. As the authors put it, their work “stands on its head” some of the theoretical results on discrimination that my wife and I published 30 some years ago.
You may be more interested in what the data shows about link between ability, race, and schooling choice. As “everyone knows,” black Americans get considerably less schooling than white Americans–about three-quarters of a year less on average. This is true, but Lang and Manove show that
- Conditional on measured academic ability, black Americans get about 1.2 years more schooling than do white Americans.
The theoretical argument is that being successful in school is one of the few things that able black students can do to signal their ability to employers. Empirically, the authors used AFQT scores, which are sometimes thought of as a rough measure of IQ, as a measure of academic ability. Then they plot years of education attained for different levels of AFQT. Here are the plots:
Note that the white line (solid) is well below the black line (dashed).
[Back to the personal part for a sec. If you’re interested in economic commentary, especially with regard to family and labor economics, take a look at my wife’s (Shelly Lundberg) Google+ posts at https://plus.google.com/circles?tab=G#113224326270049840069/posts.]