Abstract:
Having a military that is representative of the population is a stated social goal by policy makers. Since the armed forces do not gather data on the family income of recruits, studies on the socioeconomic background have relied on potentially biased geographic data, reaching conflicting conclusions. We use the National Longitudinal Survey of Youth 1997 to study population representativeness in the years 19972011 based on individual level data. In contrast to previous periods, and contrary to popular perception, those who joined the military had higher than median family income and wealth. The lowest and highest parental income categories are both underrepresented in the military. Those who joined were more likely to pursue higher education and had higher self-reported life satisfaction. Recruits had 0.2 standard deviation higher cognitive tests scores than the civilian population. Higher cognitive test scores strongly increases the probability of joining for those from lower and middle income families while interestingly lowering the probability of joining for those from high-income homes. The over-representation of minorities in the military has declined in recent decades. In sharp contrast to the Vietnam War, Non-Hispanic Whites are significantly overrepresented as casualties in Iraq and Afghanistan.