The fourth dimension – time – has only recently come on to the agenda in Britain in the study of low income and poverty. Yet as Robert Walker states, ‘without taking time into account ... it is impossible to develop policies that successfully tackle the multiple causes of [poverty]’ (Walker, 1994). In the US studies specifically looking at ‘welfare dynamics’ – following ‘welfare’ recipients in and out of benefit – have a much longer history dating back to the early 1980s (Bane and Ellwood, 1983). Tracking those on low income over time requires access to longitudinal studies which contain appropriate income data. In the US, the Panel Study on Income Dynamics (PSID), in particular, goes back to the 1960s. In Britain there was no comparable survey generating longitudinal data until the advent of the British Household Panel Survey with a first wave in 1991.