Civilian Earnings of Military Retirees

Presents evidence on the civilian earnings of military retirees relative to those of noncareer veterans, with data from the 1970 Census to provide some empirical evidence on the extent of second career earnings loss of military retirees. A commonly cited rationale for the military pension is that skills acquired in the military are not fully transferable to the civilian sector. True second career earnings loss is necessarily unobservable. Veteran status was imputed to retiree and noncareer veterans on the basis of their ages and the conflicts in which they served. On average, weekly wage rates of retirees are 10 to 20 percent lower than weekly wages of noncareer veterans. Roughly half the white retiree differential in annual earnings is estimated to be attributable to voluntarily chosen characteristics, such as years of schooling, hours of work, and location. The residual earnings differential of less than 10 percent is probably an upper bound on true second career earnings loss.