Labour market variables may moderate the link between unemployment and mental health, as has been found in numerous research papers. The aim of this study was to test, in the context of a very favorable labor market, (1) the detrimental effect of unemployment on mental health and (2) the predictive validity of mental health on further employment status. The population of the study comprised 384 newly registered unemployed persons. Mental health was assessed through four variables: self-esteem, psychological distress, perceived stress, and depressive symptoms. The results did not confirm previous studies. Using analyses of variance for repeated measures and linear regression, a negative change of mental health was not found after 6 and 12 months' unemployment, and the four measurements of mental health at registration did not predict employment status 6 and 12 months later. Results are discussed in terms of moderating variables that capture Luxembourg labor market specificities. This research was supported by the Luxembourg National Research Fund in the framework of the VIVRE Program; grant no. FNR/02/05/07. We would like to thank Mary Devine for the linguistic revision of the text.