Replication data for: Earnings Losses of Displaced Workers Revisited

Earnings losses of Connecticut workers affected by mass layoff are calculated using administrative data. Estimated reductions are initially more than 30 percent and six years later, as much as 15 percent. The Connecticut estimates are smaller than comparable ones from Pennsylvania administrative data but similar to those from the Panel Study of Income Dynamics (PSID) and Department of Workforce Services (DWS). Earnings reductions in Connecticut and Pennsylvania are concentrated among Unemployment Insurance recipients. An unusually high proportion of Unemployment Insurance beneficiaries in Pennsylvania explains the larger estimated losses relative to other studies. Fixed-effects, random growth, and matching estimators produced similar earnings loss estimates suggesting each is relatively unbiased in this context. (J31, J 63, J65, R23)