Despite a burgeoning literature on consumption and consumer society the consequences for consumerism of the UK’s ageing population have been relatively neglected by researchers. The trend to earlier retirement, greater income heterogeneity and the relative affluence of many retired people presents new challenges for the understanding of later life. The cohorts of people entering retirement over the next few years are among those who participated in the creation of the post war consumer culture. These consumers have grown older but have not stopped consuming; their choices and behaviour are products of the collective histories of both cohort and generation. Our project aims to chart the engagement of different age cohorts with consumer society and to show how retired people make up an increasingly important strand of consumer activity. In this paper we explore some of the variation in older people’s engagement with consumer society and ask the question whether retired households with less income have different spending patterns than those with higher levels of income and whether there is a general convergence over time?

Outline of the project
People approaching retirement, entering retirement or currently living in retirement experience later life differently from their predecessors. With increasing affluence and the growth of mass consumer society the social nature of ageing has become more differentiated. It is widely accepted that a single category of old age or old age pensioner can no longer encapsulate the range of experiences faced by those entering later life (Phillipson 1998). Paradoxically, as later life has become more internally differentiated, it has become less distinct from other parts of the adult lifecourse (Gilleard and Higgs 2005). Age specific (or appropriate) consumption and activities have diminished as retired people buy the same products and engage in many of the same activities as the rest of the adult population. This is happening in the context of progressively earlier retirement, improved health and greater longevity. These changing circumstances have combined with the cultural developments that were brought about by the emergence of consumer society and its emphases on lifestyle and leisure (Ransome 2005). Little empirical research however has been undertaken into the development of consumption patterns of retired people over the past few decades.

Consumption has become important in the study of social life because it is increasingly an arena for the construction of identity or lifestyle (Zukin and Maguire, 2004). It has been argued that as the cultural ascriptions of a production-based class society have less personal significance, what people consume has a symbolic value as well as a utility value (Slater 1997). This is as true of older people as it is for those at younger ages. The idea of consumer culture therefore is important in helping us understand how people make sense of their own lives even at older ages. Consumer culture at the beginning of the 21st century has also become generationally dependent. Not only are the current cohorts of retirees bringing into retirement their experiences of and attitudes toward consumption (pop music, more adventurous food and holidays), they are also bringing new expectations of post-working life. While affluent 20th century retirees once moved to British seaside towns to retire because that is where they had spent their holidays when younger, increasing numbers of British retirees today choose to retire to the Mediterranean coastline that they first experienced as the package holiday took off in the 1960s and 1970s (King, Warnes and Williams 2000). Expectations of later life are influenced by the growing engagement with consumer culture that people have experienced, growing up and growing older in the second half of the 20th century.

The cultures of consumption related to later life are ones that change to keep pace with rising affluence and which also change as different generations retire. What is central to these changes is the idea of lifestyles, because it is through lifestyles that consumer choices are crystallised providing an important basis for the more fluid identities of post-industrial society (Tomlinson, 2003). What people spend their money on is important in showing us how they live their lives and how they develop their lifestyles. This concern with lifestyles can be seen to have emerged out of post-war youth culture and its engagement with consumer culture. The importance of this for the study of later life experiences today is that these same cohorts who created the ‘generational field’ of the sixties are now either retired or retiring. Studies on generational patterns of consumption in modern society have tended to neglect the changing cultural context of consumption patterns of older people and the contrasts between them. This is a major weakness because in order to explain changing consumption patterns it is necessary to have an understanding of the extent to which current choices made by successive cohorts are rooted in their cultural experiences of consumption. Those who came to adulthood in the 1960s will have different expectations and experiences than those born ten years earlier.

Crucial to understanding ageing today is an awareness of the differences in consumption patterns between and within generations. To this end the project examined, through the use of quantitative data from the Family Expenditure Survey, variations in consumption, leisure and work among people aged fifty plus since 1963. We wanted to discover if participation in these fields has shown patterns of divergence between wealthier and poorer retired households. In previous work (Higgs et al 2006) we had established that there was increased convergence between retired and working age households in the ownership of key consumer goods. In our present analysis we examine patterns of consumption for different time points between the 1960s and the beginning of the 21st century using secondary data from the FES on expenditure patterns related to fuel, food, alcohol, clothing and household goods among the retired population. By comparing the expenditure patterns of retired households by income quintile we are able to examine whether there was greater or less divergence among retired households.

Methods
The data were taken from 8 years of the Family Expenditure Survey (FES)* of Great Britain at 5 year intervals over the period from 1968-2002. The FES is a voluntary survey of a random sample of private households in the United Kingdom carried out by the Office for National Statistics (ONS). The FES is primarily a survey of household expenditure on goods and services, as well as household income. The original purpose of the survey was to provide information on spending patterns for the purpose of calculating the Retail Price Index (RPI). In addition to expenditure and income data, the FES includes information on a range of socio-economic characteristics of the households, (e.g. composition, size, social class, occupation and age of the head of household).The basic unit of the survey is the household although data is collected on both the household and the individual level. On average about 7,000 households are surveyed each year representing an average response rate of 60 per cent although this has declined over the period.

In order to look at differences among the retired population we analysed data on trends in expenditure by retired households. Households were stratified according to equivalised income quintiles using the new OECD equivalised income scale.

[* The ONS also conduct the Northern Ireland Family Expenditure Survey which is very similar to the FES . However we did not include these data in our analyses.]

Findings
We focused on trends in expenditure in five key areas (food, fuel, alcohol, clothing and household goods).

Figure 1. Trends in expenditure on selected categories as a percentage of total expenditure, 1968-2001.

Figure 1 focuses on the proportion of total expenditure allocated to five broad categories of spending (corresponding to ONS categories). The figure shows that while the percentage of total expenditure accounted for by food has dropped from just under a third to just over a quarter, the percentage spent on household goods has risen from about 4 per cent to just under 10 per cent. Spending on fuel has declined while that on alcohol and clothing has stayed broadly similar. However, as before, these figures mask variation within retired households that become apparent when examined according to levels of income.

Figure 2. Proportion of total household expenditure spent on food by quintiles of equivalised household income.

Figure 2 shows that expenditure on food declines for all quintiles but still represents a quarter of all expenditure for the bottom 20 per cent of the income distribution while it only accounts for 15 per cent of those in the top quintile suggesting that income level is still an important determinant of household expenditure patterns.

Figure 3. Proportion of total household expenditure spent on fuel by quintiles of equivalised household income.

Figure 3 shows the same pattern for fuel and again the bottom quintile are spending a greater proportion of their total expenditure on fuel in 2001 than the top quintile were in 1968. It is however interesting to note the trend is not uniformly downward with some evidence of convergence occurring in the late 1970s and early 1980s.

Figure 4. Proportion of total household expenditure spent on alcohol by quintiles of equivalised household income.

Figure 5. Proportion of total household expenditure spent on clothing by quintiles of equivalised household income.

Figure 6. Proportion of total household expenditure spent on household goods by quintiles of equivalised household income.

Figures 4, 5 and 6 show that while there is differentiation for alcohol, clothing and household goods there is a greater tendency towards convergence by the end of the period under consideration. There is also a significant increase in the proportion of expenditure on household goods across all income groups. This suggests a shift in the pattern of expenditure across the time period that is common to all households. These data are likely to hide variations in the quality of goods being purchased and further research is required to untangle relative price effects.

We have also looked at percentages of expenditure by different income quintiles on categories of expenditure to see if there were marked disparities in expenditure patterns that might mask the ownership patterns of goods that we have described. Again, it would appear that members of different quintiles do not differ greatly in their expenditure, with all income groups seeing a decline in spending on food and fuel and an increase in spending on household goods. This would seem to suggest that most retired households are participating in the changes that mark out contemporary consumption.

Discussion
The data we have presented here demonstrate that there is evidence for the argument that retired households are converging in their consumption patterns. This is consistent with our earlier finding that over time the ownership of key consumer items by different sections of the retired population converges and becomes similar to that of working age households (Higgs et al 2006). This is a transformation from the situation that was prevalent in the recent past. During the 1950s social researchers such as Abrams (1951) regarded the circumstances of older people as to be so low as to necessitate their exclusion from any analysis of consumption. A decade later, Marsh (1965) saw ‘pensioners’ as constituting the lowest social class grouping because they were at ‘the lowest level of subsistence’. This began to change slowly as the relative prosperity that accompanied the post-war boom began to filter through to the older population. Progress was slow and the numbers of older people (particularly women) who remained in poverty should not be underestimated (Ginn and Arber 1999). However the circumstances of the different cohorts of people entering retirement during the second half of 20th century changed the experiences of retirement. This was to some degree inevitable as newer retirees had benefited more from the rise in standards of living than those who had retired before them (DWP 2003). Those reaching retirement age in the 1950s and 1960s were still marked and constrained by the effects of economic recession and mass unemployment that had dominated their earlier lives. This group had not had the opportunities to accumulate the resources that would become more common with later cohorts of retirees. Consequently, even in 1979, 45 per cent of the retired population were in the bottom quintile of income in the UK (DWP 2003). The transformation of the circumstances of the retired population can be gleaned from the fact that the same data sources show that the increase in the income of retired people outstripped that of the rest of the population and was even faster if the retiree had access to an index linked occupational pension based on final salary. These changes had such a significant impact that by 2001 the income distribution of the retired population as a whole became more closely matched to that of the working age population than it did to other groups such as the unemployed or lone parents with children. Paradoxically this also meant that there are wider disparities of income between the better-off and the worse-off retired as later life begins to more closely reflect the greater inequalities of society at large. Following on from our earlier argument, the UK may therefore be experiencing a transformation of the field of ageing as cohorts who grew up in relative prosperity become used to that relative affluence and bring into later life a generational habitus of consumption. Data from Casey and Yamada (2002:39) point to the fact that not only is total spending by retired couples close to that of working couples but that spending on recreation and leisure by this group is high. These trends are certainly being encouraged by the British Government as it seeks to reduce its commitments to funding later life. Mann (2006) argues that the UK Government is increasingly using the language of consumption in articulating its pensions and retirement policies. Rather than retirement being a Government responsibility it is now seen as the arena of the active citizen consumer and lifestyle manager. Those who wish to remain as ‘passive welfare recipients’ lose out all of the way whether this is in terms of income or in terms of being seen as ‘passive’. People in retirement find that they are incorporated into the processes of consumption and the imperatives of consumer society.

Conclusion
Our intention in presenting this paper is to show how that even at a simple descriptive level – the relative proportions of their expenditures spent on different categories of consumption - the retired population appear not only to be part of consumer society but reflect many of the patterns and divisions present in society as a whole. Further work is needed to investigate the inter-relationship between inter alia: different cohorts, gender, and retirement as well as the nature of retirement households on the patterns we have described. Only by doing so will we be able to deepen our understanding of how later life has been transformed by the generational habitus of the first generations of the consumer society as they themselves move into retirement.

Acknowledgements
Research supported by ESRC grant RES-154-25-0007 'From Passive to Active Consumers: Ageing and Consumption in Britain 1963-1998' as part of the AHRC/ESRC 'Cultures of Consumption' Programme

References
Abrams, M. (1951) Social Surveys and Social Action London, Heinemann.

Casey, B. and Yamada, A. (2002) Getting older, getting poorer? A study of the earnings, pensions assets and living arrangements of older people in nine countries. Labour Market and Social Policy Occasional Papers, no. 60. Paris, OECD.

Department of Work and Pensions (2003) Pensioner Income Series 2001/2, HMSO, London.

Gilleard, C. & Higgs, P. (2005) Contexts of ageing: Class, cohort and community, Cambridge, Polity Press.

Ginn, J. Arber, S. (1999) Women and Pension Poverty: Prospects and options for change. In S. Walby (Ed.) New Agendas for Women. London, Macmillan.

Higgs P., Hyde M., Gilleard C.,Victor C.,Wiggins R. and Jones I.R. (2006) From Passive To Active Consumers? Trends In Ownership Of Key Goods In Retired And Non-Retired Households In The UK From 1968-2001, Cultures of Consumption Working Paper Series, http://www.consume.bbk.ac.uk/working_papers/HiggsetalJune2006.doc

King, R. Warnes, A. and Williams, A. (2000) Sunset lives: British retirement migration to the Mediterranean, Oxford, Berg.

Mann, K. (2006) ‘Three Steps to Heaven? Tensions in the management of welfare: Retirement pensions and active consumers’ Journal of Social Policy 35:77-96.

Marsh, D. (1965) The changing social structure of England and Wales 1871-1961, London, Routledge and Kegan Paul.

Phillipson, C. (1998) Reconstructing old age, London, Sage.

Ransome, P. (2005) Work, consumption and culture: affluence and social change in the 21st Century, London, Sage.

Slater, D. (1997) Consumer culture and modernity, London, Sage.

Tomlinson, M. (2003) Lifestyle and social class, European Sociological Review, 19: 97-111.

Zukin, S. and Maguire, J.S. (2004) Consumers and consumption, Annual Review of Sociology, 30: 173-197.