Recent meta-analytic research suggests that the absence of pessimism is a stronger predictor of health than the presence of optimism (Scheier et al., 2021). Based on these findings, the present study examined the role of subjective well-being in the health effects of optimism and pessimism in romantic couples. It was expected that pessimism would be more strongly associated with both well-being and health than optimism, intra- and inter-personally. In addition, subjective well-being was hypothesized to mediate the health effects of optimism and pessimism. The study included two waves of data obtained from 153 opposite-sex couples. Self-report indicators of optimism, pessimism, subjective well-being, and health were assessed across waves. Cross-sectional and longitudinal actor-partner interdependence models showed that only the absence of pessimism, but not the presence of optimism, was associated with high levels and improvements in health (e.g., subjective health, sleep-efficiency, cold symptoms, chronic disease). In addition, pessimism was a stronger predictor than optimism of levels in some indicators of subjective well-being (e.g., negative affect, depressive symptoms). These effects were obtained intra- and inter-personally. Finally, levels of subjective well-being statistically mediated the effects of pessimism on levels and changes in health. The study’s results suggest that subjective well-being could represent a pathway that explains differential associations between pessimism and optimism with health, both intra- and inter-personally.