The Enlistment, Enrollment and Employment Decisions of Youth: 1973-2001

Meredith A. Kleykamp, Princeton University

The Enlistment, Enrollment and Employment Decisions of Youth: 1973-2001 Meredith Kleykamp Princeton University Military service is one of many options available to youth, one that is disproportionately chosen by African Americans. To understand demographic representation in the military, one must examine the context of the decision to enter the service and its relation to the alternatives available. After high school, youth face the choice to enter the military, go on to college, enter the labor market or be jobless. Blacks, whites and Hispanics are making different decisions about which path to follow. In this paper I seek to understand how trends in educational attainment, employment, and enlistment in the military are related, and how these trends have affected the race and ethnic composition of the armed forces since the inception of the All-Volunteer force in 1973. The military is an institution that plays a pivotal role in the lives of many young men, black men in particular. In 1990, nearly 21% of black men between the ages of 18 and 35 had some experience serving in the armed forces, while only 12-13% of whites did (author's computation from 1990 Census data). In 2000, roughly 24% of the most junior enlisted soldiers were black (E1-2). This percentage increases steadily through the enlisted ranks, ending with 40% of the highest ranking enlisted soldiers (E9) being African American1. That the proportion of blacks increases with rank indicates that more blacks find the military to be a viable career alternative as well as a short term educational and training opportunity, than do whites. Military service by African American men is a relatively invisible equalizing force between blacks and whites at the general population level. The military population is not included in most national statistics, including calculations of employment and unemployment figures. Though we often tend to forget or ignore this population, it is a non-trivial part of the working-age population, especially for African Americans. In 2001, there were nearly 200,000 African American men, 93,000 Hispanic men and almost 630,000 white men in the military, the majority of whom are younger than thirty . Since proportionately more military personnel are black, this results in the undercounting of many fully employed individuals. If we included the active duty population in employment figures, we find that in 2001, traditional estimates overstate the number of young black men unemployed by roughly 2.5 percentage points, while doing little to change our estimates for whites. This is in a year with relatively high employment and low opportunities for military service, as the 80's and 90's drawdown reduced the number of military "jobs" available to youth. In other years, the effect is likely to be much larger. To examine the race/ethnic composition in the military, college population and in the labor market, I use data from the annual March Current Population Survey (CPS) supplement data from the years 1973 to 2001 and a census of the active duty military population. Males age 16-29 are selected for analysis from the CPS and DoD data. The data are individual level data including the following variables: employment status, level of education, age, race/ethnicity, veteran status, enrollment/years out of school, and survey year. The DoD data include age, race/ethnicity, highest grade of school completed, and year. My strategy is to employ a multinomial probit model to estimate the choice to enlist, enroll, work or remain out of the labor force among youth in the aforementioned period. I use these two model specifications because the independence of irrelevant alternatives (IIA) assumption of the multinomial logit model is inappropriate. The goal of this analysis is to understand the changing relationships between the choices to enlist, enroll or work, and the consequences for the demographic composition of the military. Imposing the restriction that these choices be independent clearly defeats this purpose, and is inappropriate when we have an a priori expectation that these choices are related. The multinomial probit specification has the advantage that it allows for the estimation of covariances among the four choices available to youth upon completion of high school. Thus, we can identify whether the choice to enlist in the military is more similar to the decision to got to college or to work in the civilian labor market, and whether this relationship changes over time. Previous researchers have made strong claims about the effect of the military and college "creaming" the most employable blacks from the labor market, resulting in convergence between black and whites in many social indicators, but in diverging employment trends2. This implies that enlistment and enrollment are much more similar to each other than to employment, or staying out of the labor force. Previous analyses did not consider enlistment and enrollment separately, but rather lumped them together, obscuring the independent effects of each on race-specific employment trends. This is problematic given the claim that both the military and college siphon the most employable minorities away form the labor market. It is impossible to determine from these previous analyses the relative effect of each. 1 Department of Defense. 2001. Population Representation in the Military 2002. 2 See Mare, R. D., and C. Winship. 1984. "The Paradox of Lessening Racial-Inequality and Joblessness among Black-Youth - Enrollment, Enlistment, and Employment, 1964-1981." American Sociological Review 49:39-55 and Mare, R. D., C. Winship, et al. (1984). "The Transition from Youth to Adult - Understanding the Age Pattern of Employment." American Journal of Sociology 90: 326-358.