Data are presented from the seventh wave of a longitudinal study of school-leavers that commenced in 1980. Four groups were compared: satisfied employed, dissatisfied employed, unemployed and full-time tertiary students. The groups differed initially with respect to only one background variable, teacher-rated academic potential, and did not differ on any of the psychological measures of well-being. In general, the unemployed and dissatisfied employed groups displayed poorer psychological well-being than the satisfied employed and student groups on a range of measures. The longitudinal data suggested that this was because the two disadvantaged groups showed smaller improvements than the others, rather than any deterioration. The results also suggested that for the males unemployment was worse than unsatisfactory employment, but that for the females unsatisfactory employment was worse than unemployment.