Life-course socio-economic influences on mortality have different implications for intervention but have not been established in the New Zealand context. We will analyse mortality records linked with the historic longitudinal Census dataset (linked records across the 1981–2006 Censuses) to fulfil four research aims: 1) To test which life-course hypotheses best explain associations between socio-economic status and mortality: accumulation, sensitive period, social mobility or instability? 2) To test whether social and cultural capital protects against socio-economic risk. 3) To assess ethnic disparities in mortality and test whether these are explained by the greater experiences of long-term harsh and unstable environments among some ethnic groups (e.g. Maori, Pacific). 4) To test life-course hypotheses among siblings discordant on socio-economic risk, and among siblings discordant for social and cultural capital (thus controlling for family background factors). This research will be both methodologically innovative (e.g. discordant sib-pair analyses) and important for the understanding of ethnic inequalities in New Zealand.