This study investigates the association between self-esteem assessed either in adolescence or in adulthood with adult academic and psychosocial outcomes. One hundred and thirty-one junior high school students were selected based on their experience of academic difficulty or success, and both groups were selected equally from regular or low-performing schools. Ten years later, 100 of these individuals participated in a follow-up assessment of academic, socioprofessional, and health-related outcomes. Logistic and linear regression models were performed to estimate the association of self-esteem (measured by the Rosenberg Self-Esteem Scale) at baseline and at follow-up, as well as change in self-esteem categories (high vs. low) on all outcomes. Fifty-four percent of adolescents were consistently high in self-esteem over the 10-year period, 17% experienced an improvement, 21% a decrease, and 8% stayed at a low level. Outcomes at follow-up were more consistently associated with young adult self-esteem than adolescent self-esteem. Self-esteem was associated with several outcomes (i.e., personal goals, life satisfaction, alcohol, and health outcomes) and its effects were different depending on the period considered and categories of self-esteem change. The self-esteem difficulties which are common in adolescence raise concerns about their potential impact on important outcomes later in life. However, self-esteem difficulties in young adulthood were more associated with negative outcomes at this phase of life (whether low self-esteem persisted into adulthood, or emerged only during adulthood). Relative to outcomes in young adulthood, the period of greatest interest for improving well-being may be concentrated in the years immediately surrounding the studied outcomes. Self-esteem change from adolescent to adulthood is associated with a large diversity of outcomes. Less is known concerning the extent to which adolescent self-esteem may independently determine outcomes in young adulthood as compared to more temporally proximal self-esteem levels. Only self-esteem difficulties in young adulthood were associated with negative outcomes at this phase of life (whether they persisted from adolescence into adulthood, or emerged for the first time during adulthood). Targeting self-esteem difficulties that are temporally proximal to the period of interest may improve mental health and well-being.