The Effect of Arrests on the Earnings of Young Men: Evidence from the National Youth Survey

This paper seeks to test the hypothesis that criminal background has a negative impact on earnings, using data on arrests from a national longitudinal survey on youth and delinquency. I estimate that having been arrested causes a reduction in annual earnings of 18 to 26 percent. I find some evidence that continued involvement in criminal activity is an important factor that influences the size of the earnings gap. I also find evidence that human capital factors -education and work experience -influence the association between arrests and earnings. My analysis suggests that other factors to which this earnings gap has been attributed -such as individual background and attitude and social control factors -are far less important than hypothesized. Using a wage decomposition, I find that there is a substantial unexplained portion of the earnings gap, over 55 percent, which therefore does not rule out stigma as an important cause of the earnings gap.