The Value of High School Employment

"The Changing Benefits of Early Work Experience." Charles L. Baum and Christopher J. Ruhm. National Bureau of Economic Research Working Paper No. 20413, August 2014.

Millions of American teens have gone through the coming-of-age ritual of running the fry station at a fast-food place, ringing up clothes at a store, or watching for trouble from the lifeguard’s chair at the local pool. In return, they’ve gotten a modest wage, a chance for socializing, and work experience.

But the measured long-term economic benefits of that work experience have been going down sharply, according to a recent paper by Charles Baum of Middle Tennessee State University and Christopher Ruhm of the University of Virginia. While they don’t come to firm conclusions about the cause of the trend, it may reflect broader changes in the labor market.

Most previous research has found that high school employment has a positive effect on future employment and wages. But in theory, it could equally have the effect of hurting a student's prospects by interfering with academic achievement. Baum and Ruhm seek to determine whether the net benefits of high school jobs changed over time by looking at data from the National Longitudinal Surveys of Youth, a set of long-term surveys taken by the Bureau of Labor Statistics. They analyze data on two sets of high school students approximately two decades apart: the survey's 1979 cohort, a sample of Americans who were high school students in the late 1970s and early 1980s, and its 1997 cohort, a sample of high school students in the late 1990s and early 2000s.

The researchers look at the wages of respondents at ages 23 to 29, or roughly five to 11 years out of high school, along with other measures of employment success. They take into account various family characteristics (such as parents' education) and characteristics of the individual's high school program. They control for individual ability using scores from armed-services tests administered to many high school students; for the 1997 cohort, they also use grades from eighth grade.

Their main finding is that over the period of the study, the predicted financial benefit from working 20 hours or more per week in the senior academic year of high school dropped almost by half. (They find no statistically significant effect from sophomore- or junior-year jobs.) For those who were students in the late 1970s and early 1980s, working in the senior year yielded an average long-term wage increase of 8.3 percent; for the students of the late 1990s and early 2000s, it was 4.4 percent. The change was felt most by women. "Work experience during the high school senior year continues to predict positive effects on labor market outcomes 5-11 years after the expected date of high school graduation," they conclude, "but these beneficial consequences have attenuated fairly dramatically over time."

Why the drop? Baum and Ruhm note that for the first group, high school work reduced the likelihood of later holding service jobs, which are typically lower-paid — while for the second group, it increased the likelihood of a service job. In addition, high school work was associated with less of an increase in adult work experience for the second group than for the first group. They estimate that at least five-eighths of the decline in the benefit of high school work is from these two effects, in roughly equal proportions.

They find that the returns to high school work were higher for the non-college-bound and declined the most for those students. For the non-college-bound, the students in the late 1970s and early 1980s who worked 20 hours or more per week in the senior year of high school saw an average increase in their future wages of around 13 percent, compared to 7 percent for those in the late 1990s and early 2000s. For the college bound, those in the first cohort saw an increase from high school work of 3 percent, while those in the second saw an increase of 2.2 percent.

The decline in the long-term payoff of high school employment has coincided with a decline in high school employment itself. Overall, the labor force participation rate of teens aged 16 to 19 dropped from 57.7 percent in January 1980 to 52.2 percent in January 2000. Beyond the study period, the rate continued to fall, reaching 34.6 percent in December 2014. The authors posit a number of possible reasons for the drop during the 2000s, including "increased competition for jobs from immigrants, former welfare recipients and other adults, as well as an increased emphasis on education and in the availability of financial aid for college."

With regard to the share of the declining benefit of high school work that their findings do not explain, the explanation could lie in the dramatic labor-market changes of the past several decades. These changes may have created forces influencing the benefits of high school work differently in the more recent period. Among the changes are the hollowing out of the labor market (that is, the decline that many believe has taken place in demand for middle-skilled labor) and the rising value of college degrees in the labor market. The authors suggest that the causes behind their findings, especially the concentrated effects on women, deserve closer study.