The Estimation of Training Premiums for U. S. Military Personnel.

The paper examines the effect of military training and experience on the civilian earnings and employment of persons who have served one term of military service. Since training effects are specific to individuals and to civilian occupations, empirical problems in precisely defining the meaning of training effects arise. Additional problems in understanding and estimating training effects result from selectivity bias. Preliminary assessments of the effects of military training on civilian earnings indicate that when civilian occupational choice is ignored, military training does not seem to significantly affect civilian earnings. However, within occupations, military training does sometimes have such an effect. Having discussed these theoretical pitfalls and preliminary findings of the project, the paper develops a model of choice from which it is possible to derive empirically estimable relationships to calculate military training effects and to impute the total returns to training in a military specialty from the occupational choices actually made by enlisted personnel. Finally, the paper discusses four problems which need to be solved before the model can be applied (treatment of unemployment, choice of critical sample sizes, survey nonresponse, and selection of civilian alternative categories), and outlines a plan for proceeding with the analysis of military training premiums. (JR)