Quantile regression, Box-Cox transformation model, and the U.S. wage structure, 1963–1987

This study examines changes in the increments (‘returns’) to education and experience at different points of the wage distribution. It also examines changes in within-group wage inequality. Newly developed quantile regression techniques are applied to the Box-Cox transformation model using the Current Population Survey data. The increments differ, but they changed in similar pattern at different points of the wage distribution. They are larger at the low end of the distribution for the less skilled workers, and larger at the high end for the more skilled workers. Wage inequality increased differentially for all skill groups. These increases result mainly from a longer tail at the lower end of the wage distributions.