People with lower income or lower educational level have worse health and higher mortality. This project addresses the two basic underlying questions for this finding: How does your socioeconomic status determine your health? And how does your health determine your socioeconomic status? The project aims at studying life courses of persons aged 50+ at the time of the interview who were surveyed prospectively, and also retrospectively for their entire life history starting at childhood. We will use data from the Survey of Health Aging and Retirement in Europe (SHARE) from 30,000 individuals in 14 European countries, and the English Longitudinal Study of Ageing (ELSA) that offers the same data for more than 7,000 persons. This data contains exact information on periods and events of ill health and periods and events of change in socioeconomic status. In addition, there is explicit information on the consequences of poor health in childhood in terms of schooling, or, later in life, in terms of working hours or career perspectives. Life histories from childhood to old age are the ideal basis for disentangling causality between SES and health because confounding of the key variables prior to measurement is minimal. By applying simultaneous equation models for hazards (survival analysis) the correlated processes of health deterioration (influenced by SES) and the development of SES (influenced by health) can be disentangled. This will produce unbiased estimates of the effect sizes for both causation directions and answer the question of causation and selection in a life course perspective taking into account endogeneity and confounders. We will also reveal if common background factors are influencing both SES and health and we will identify possible differences in the causation direction between life stages. This study will provide important new insight into the dual relationship between SES and health and will help to understand and to tackle social differences in health.