A model of occupational choice in the U.S. Navy

Patrick C Mackin, Fordham University


This study examines the occupational-choice behavior of enlisted personnel in the U.S. Navy. It predicts the probability that a sailor will reenlist in the Navy for three years or longer; extend his present contract for two years or less; or leave the Navy immediately. The analysis develops a conditional logit model in which choice probabilities are determined by the relative pecuniary returns to each choice and a vector of other variables, including unemployment and the sailor's individual characteristics. The model estimation used a sample of sailors making decisions in Fiscal Years 1978 through 1989. Sailors were divided into samples by reenlistment zone and occupational (skill) group. It was the first study to examine decisions among sailors at the third reenlistment point and sailors who were eligible for retirement. The results confirm that relative military pay has a significant effect on retention behavior. The implied pay elasticities, however, were smaller than those found in earlier studies. This finding may be evidence that the period covered in this analysis covers a less elastic portion of the labor supply curve. Unemployment also had a positive effect on reenlistment and extension probabilities, although its impact was not as strong nor as universally significant across all samples. Expected sea duty, which affects taste for the Navy, did not have a clear relationship with retention. An alternative specification of sea duty did not improve the results. Demographic and service-related variables picked up other non-pecuniary effects. Relative promotion speed had a positive retention effect, but its impact was relatively greater for extension probabilities than for reenlistments. Likewise, non-white and female sailors were more likely to reenlist and extend than otherwise similar white, male sailors, although the extension effect was larger than the reenlistment effect. Quality and education effects varied across reenlistment zones. Finally, the model did not seem to fit the retirement-eligible groups well; alternative tests suggested that taste and demand factors overwhelm pay and economic effects.

Subject Area

Labor economics|Labor relations

Recommended Citation

Mackin, Patrick C, "A model of occupational choice in the U.S. Navy" (1992). ETD Collection for Fordham University. AAI9300241.