The effect of unemployment on mental health was examined with meta-analytic methods across 237 cross-sectional and 87 longitudinal studies. The average overall effect size was d = 0.51 with unemployed persons showing more distress than employed persons. A significant difference was found for several indicator variables of mental health (mixed symptoms of distress, depression, anxiety, psychosomatic symptoms, subjective well-being, and self esteem). The average number of persons with psychological problems among the unemployed was 34%, compared to 16% among employed individuals. Moderator analyses demonstrated that men and people with blue-collar-jobs were more distressed by unemployment than women and people with white-collar jobs. Linear and curvilinear moderating effects of the duration of unemployment were also identified. Furthermore, the negative effect of unemployment on mental health was stronger in countries with a weak level of economic development, unequal income distributions, or weak unemployment protection systems compared to other countries. Meta-analyses of longitudinal studies and natural experiments endorsed the assumption that unemployment is not only correlated to distress but also causes it. Seemingly inconsistent longitudinal results of older meta-analyses can be explained by retest artifacts. We also identified mental-health related selection effects during job loss and job search, but they are weak. With an effect size of d = -0.35 intervention programs for unemployed people were found to be moderately effective in ameliorating unemployment-related distress among continuously unemployed persons. (Contains 1 figure and 8 tables.)