The 1980's witnessed rapid changes in the structure of wages in the United States. In this study, recently developed techniques for quantile regression are applied to every March Current Population Survey since 1964 and changes in the return to schooling and experience at different points of the wage distribution are examined. The quantile regression technique parsimoniously describes the entire conditional wage distribution. This conditional distribution is used to examine changes in within-group wage inequality as measured by the difference between conditional quantiles. Two linear models are considered: a simple one-group model, and a 16-group model. The results suggest that the returns to schooling and experience differ across quantiles of the wage distribution but their patterns of change are similar. Significant differences in wage inequality are also found across the various skill groups. The paper also presents a new imputation method for CPS data.