This study examines two related issues concerning the subjective well-being of elderly adults: change over time and correlates of that change. Data come from a three-wave panel study of 401 elderly residents in St. Louis. Residualized change score regression analyses indicate: there is change in subjective well-being over 4-5 months and over 12 months; the 4-5 month and 12 month changes are remarkably similar; the effect of subjective well-being over time indicates regression to the mean; and, only socioeconomic status is a significant predictor of change in subjective well-being (net of the effects of subjective well-being itself). The implications of these results for our understanding of subjective well-being in the elderly are discussed, as are the policy implications of the positive effect of socioeconomic status on changes in the subjective well-being of the elderly.