We pursue an economic approach to analysing poverty. This requires a focus on the variables that individuals can influence, such as forming or dissolving a union or having children. We argue that this indirect approach to modelling poverty is the right way to bring economic tools to bear on the issue. In our implementation of this approach, we focus on endogenous demographic and employment transitions as the driving forces behind changes in poverty. We construct a dataset covering event histories over a long window and estimate five simultaneous hazards with unrestricted correlated heterogeneity. The model fits the demographic and poverty data reasonably well. We investigate the important parameters and processes for differences in individuals’ poverty likelihood. Employment, and particularly employment of disadvantaged women with children, is important.