Introduction

For over 60 years, significant research activity has addressed the extent to which the effects of social class over the life course have determined or contributed to an individual's economic and social fate in old age. This has led to the elaboration and discussion of a whole host of conceptual and measurement issues among a growing body of epidemiological and social researchers. In light of the social changes and accompanying theoretical developments over the same period, to these issues we must add questions about the viability of class as a means of understanding social relations and social inequality in contemporary society. In this chapter, we will interrogate these issues as they relate to the role of class in later life using the prism of health inequalities. We will seek to show how the relationship between class and later life, in terms of health inequalities, is more complex than would be expected. Furthermore, we will argue that this complexity demands that we need to think more creatively about how such relationships can work and whether the nature of contemporary retirement creates a more ‘individualised’ context for the operation of social class in all its different manifestations.

A major driver of our argument is that when examined in the round, much research on health inequalities in later life in North America and Western Europe has produced remarkably ambiguous results. While some studies have demonstrated a convergence in the health of those from different socio-economic positions as they enter old age, other studies have shown that class-related inequalities in mortality, morbidity and health behaviours continue deep into later life (Breeze et al, 2001). Researchers have focused on two competing hypotheses to explain these patterns: the cumulated disadvantage thesis, which suggests that the level of health inequality related to socio-economic status (SES) in a cohort increases as a cohort ages (Dannefer, 2003; Prus, 2007); and the ‘divergence/convergence’, or ‘age as leveller’, hypothesis, which suggests a widening of inequalities up to early old age but with a decrease in inequalities thereafter (Beckett, 2000). To date, much socio-epidemiological research interrogating these hypotheses has treated social class as a variable within standard log linear models.