This paper studies the effects of unions on the structure of wages using an estimation technique that accounts for misclassification errors in reported union status and potential correlations between union status and unobserved productivity. The model is estimated separately for five skill groups using a panel data set formed from the U.S. Current Population Survey. The results suggest that unions raise wages more for workers with lower levels of observed skills. Union workers are positively selected from the population of workers with lower levels of observed skill and negatively selected from the population with higher observed skills. Copyright 1996 by The Econometric Society.
David Card (Mon,) studied this question.