Estimating poverty persistence in Britain

Abstract

This article uses longitudinal data from the BHPS (1991-2006) to document low-income dynamics and persistence for individuals living in Britain. Poverty exit and re-entry rates are estimated, and the resulting distribution of time spent in poverty is calculated, both in single and in multiple-spell frameworks. Poverty persistence predictions are also produced for various subgroups of the populations. In order to do so, I estimate a multiple-spell model of transitions in and out of poverty, controlling for observed and correlated unobserved individual heterogeneity and for a potential initial condition problem. Components-of-variance models are also used to predict the number of years in poverty for the targeted groups. The two alternative modeling approaches are shown to produce a consistent picture of poverty persistence.