The analysis used a panel data set collected by Delbert Elliot and David Huizinga at the Behavioral Research Institute, University of Colorado. The NYS used a probability sample of households in the continental United States. The 1,725 participating youths, aged 11 to 17 at the time of the initial interview in 1977, were selected to be representative of the total youth population of the same age in the U.S. Census Bureau. The youths were interviewed annually from 1976 to 1980, and then at 3-year intervals beginning in 1983. This study dealt with first arrest during the period from 1983 to 1986 for individuals who were out of school in both years. The NYS arrest data are from self-reports rather than official records. After excluding women and African-American men, there were 178 white men who were out of school and in the work force in both periods and not arrested by 1983. Two measures of employment stability were used: number of weeks worked at the major source of employment over the last year (job length) and a proxy for job stability. An individual was considered stable in his job if he worked only one job this year for 40 weeks or more, or worked more than one job but had a full year at one job. Findings suggest that arrest can lead to minor problems in the labor market above and beyond the impact of current or past criminal activity. 2 figures, 3 tables, 8 notes, and 45 references