Estimating the union earnings effect using a sample of displaced workers.

This paper improves on past longitudinal estimates of the union earnings effect by using a sample of workers for whom the error in measuring changes in union status is minimized. The author uses a sample of workers displaced by plant closings from the 1994 and 1996 Current Population Survey Displaced Workers Supplement files to estimate the effects of union membership on weekly earnings. When models are estimated using the entire sample of displaced workers, longitudinal estimates of the union earnings effect are quite similar in magnitude to estimates from cross-sectional regressions. In models estimated separately by skill group, the author finds some evidence of positive selection into unions among workers with low observed skills and negative selection into unions among workers with high observed skills. (Abstract courtesy JSTOR.)