Wage analyses indicate that white men in the U.S. who are married earn between 8 and 15 percent more than white men who are not married, even after controlling for education, experience, and occupation. The precise nature of this wage differential has not been fully explained, although considerable effort has been expended to determine how much of the differential is attributable to selection and how wages change following marriage.
We expand upon this literature in several dimensions using data from the National Longitudinal Survey of Youth. First, we extend the analysis to nonwhite men, finding that the marital wage effect is indeed not race specific. Second, we introduce a control for cognitive skills to determine if the selection effect is brought about because more able men are more likely to marry. While our measure of cognitive skills is associated with higher wages and is higher for married men than for unmarried men, controlling for such cognitive skills has no significant impact on the marital wage differential. Finally we test the hypothesis that the marital wage differential reflects a differential in formal training between married and unmarried men. There exists some evidence, both theoretical and empirical, that married men are more likely to receive formal training than unmarried men. Our formal training measures are significant determinants of wages, but again have no impact on the estimated marital wage differential.