In or out? Poverty dynamics amongst pensioner households in the UK.

Presenter: Ricky Kanabar, ISER, University of Essex

Author: Ricky Kanabar

Individuals in the UK are living longer: The 2011 Census highlighted one in six individuals in the UK is aged over 65, however relatively little is known about the income dynamics of this group. Rising longevity is not a problem assuming (1) The number of healthy life years lived also increases and (2) Individuals are not in poverty. The focus of this paper is to investigate the latter. We use a first order Markov model to test for correlation between initial and conditional poverty status in Waves 2-4 of Understanding Society. We explicitly control for the fact that survey response differs, for example due to differences in health which may have implications for our results. The results suggest individuals who are initially poor are no more or less likely to remain poor than the initially non-poor. The model estimate indicates no evidence of a correlation between initial and conditional poverty status with survey retention. These results imply that future research investigating low income dynamics in pensioner households can be accomplished using relatively simple and well established methods.

Using the model estimates it is possible to demonstrate how differences in individual and household characteristics can heavily influence particular aspects of the poverty experience such as the time spent in poverty, the time spent out of poverty and probability of entering poverty. This provides a useful tool to policymakers in understanding how the introduction of particular policy might affect the poverty experience. The model also allows us to test whether there are scarring effects from experiencing poverty; we find no evidence of such effects. However we do find evidence of state dependence, that is to say the extent to which current poverty depends on past poverty is large (around 93%), with only the remaining 7% attributable to differences in individuals. This result is striking, the fact that an individual was poor last year largely determines their chances of being poor this year. The size of this effect highlights the need for policies which prevent pensioner households from falling into poverty.