It is believed that the length of a person's life depends on a mixture of economic and social factors. Yet the relative importance of these is still debated. We provide recent British evidence that marriage has a strong positive effect on longevity. Economics matters less. After controlling for health at the start of the 1990s, we cannot find reliable evidence that income affects the probability of death in the subsequent decade. Although marriage keeps people alive, it does not appear to work through a reduction of stress levels. Greater levels of psychological distress (as measured by General Health Questionnaire (GHQ) stress scores) cannot explain why unmarried people die younger. For women, however, we do find that mental strain itself is dangerous. High GHQ stress scores help to predict the probability of an early death.