To estimate the magnitude and temporal pattern of displaced workers' earning losses, a study used an unusual administrative data set that included employees' quarterly earnings histories and information about their firms. It created a longitudinal earnings file for a 5 percent sample of the Pennsylvania wage and salary work force. These data had two principal advantages: a large sample of quarterly earnings histories extending from 1974-86 for both displaced and nondisplaced workers and a merger of workers' earnings histories with information about their firms. To develop a statistical framework for estimating and summarizing the evidence on the magnitude and temporal pattern of displaced workers' earnings losses, the study specified a definition of earnings loss associated with worker displacement and developed a basic statistical model to represent workers' earnings histories and identify the displacement effect with some subset of the model's parameters. Findings indicated that high-tenure workers experienced substantial long-term earnings losses when they left their jobs, representing 25 percent of workers' predisplacement earnings. For workers displaced from distressed firms, these losses were characterized as follows: they were long term, with little evidence of substantial recovery after the third year; they began even prior to workers' separations; they were not limited to workers in just a few industrial sectors; and they were substantial even for those who found jobs in similar firms. (Appendixes contain notes on the data and 25 references.) (YLB)