Consumption and Age

views updated


Simply defined, consumption is household spending on consumption goods and services. However, consumption can also be defined as the satisfaction obtained by consumers from the use of goods and services. A person's life-cycle stage is usually regarded as the most important predictor of consumption. Households headed by young people usually spend less than average on products and services because their households are small and their incomes are low. In middle age, spending reaches a maximum as family size increases and incomes peak. Spending then declines in older age as household size and income decline.

The changing demographics of the American population are likely to affect consumption in the future. There were 34.4 million persons age sixty-five or older in the United States in 1998, and by 2030 there will be about 70 million persons age sixty-five or older. In 1998 this age group represented 12.7 percent of the U.S. population; in 2030 the comparable figure will be 20 percent. Diversity within the older population, and how it changes, will also influence consumption of goods and services.

Theoretical framework

Theories that attempt to explain how income and consumption vary over the life cycle include the life-cycle hypothesis, the permanent-income hypothesis, and precautionary savings. The life-cycle hypothesis (Ando and Modigliani, 1963) suggests that consumers try to maintain a relatively stable level of consumption over their lifetime. In practice, this means that younger people borrow to meet consumption needs, middle-aged people save as large a proportion of income as possible, and older people spend down their assets when their income is reduced in retirement. The permanent-income hypothesis (Friedman, 1957) suggests that consumers adjust their spending level to their perceived level of future income and that they dissave in retirement. The precautionary-saving model (Deaton, 1992) suggests that older people are extremely cautious about spending down their assets because they are uncertain about how long they will live, about the cost of health care in the future, and about the possibility of becoming impoverished.

Diversity among older persons

The diversity of the older population contributes to a broad range of consumption needs and preferences. The sex ratio is an example: in 1998, there were 20.2 million older (65+) women and 14.2 million older men, or a sex ratio of 143 women for every 100 men. This ratio increases in older age groups: in 1998, it ranged from 118 women per 100 men for the 65 to 69 age group, to 241 women per 100 men for persons aged 85 and over. Marital status presents another example of the diversity of the older populationolder men are much more likely to be married than older women. In 1998, 75 percent of older men and 43 percent of older women were married. There were four times as many widows as widowers in 1998.

Living arrangements of older persons are also diverse. In 1998 the majority of older noninstitutionalized persons lived in a family setting; that is, with a spouse or relatives. Approximately 80 percent of older men and 58 percent of older women lived in a family setting. Further, living alone was found to increase with advanced age. Three of every five women age 85 and older lived outside of a family setting in 1998. The percentage of persons 65 and older who lived in nursing homes in 1996 increased dramatically with age; only 1.1 percent of persons age 65 to 74 lived in nursing homes while 19.8 percent of persons 85 and over did so.

Participation of older persons in the labor force shows diversity as well. Twelve percent of older Americans were in the labor force in 1998, and older men were twice as likely to be employed as older women. Just over half (54 percent) of the workers over 65 were employed part-time: 48 percent of men and 62 percent of women. Twenty-three percent of older workers were self-employed, compared to 7 percent of younger workers. Over two-thirds (71 percent) of the self-employed were men.

The amount of income and the sources of income of older persons are also diverse. The median income of older persons in 1998 was $18,166 for males and $10,504 for females. The major sources of income in 1996, as reported by the Social Security Administration, were Social Security benefits (reported by 91 percent of older persons), income from assets (reported by 63 percent), public and private pensions (reported by 43 percent), earnings (reported by 21 percent), and public assistance (reported by 6 percent). In 1996, Social Security accounted for 40 percent of the aggregate income of the older population. The remainder consisted of earnings (20 percent), pensions (18 percent), assets (18 percent), and 4 percent from other sources. In summary, with advancing age, the population of older persons in the United States is more likely to be female, living alone, and to have less income than middle-aged people.

Overview of household spending

The Consumer Expenditure Survey (CEX), sponsored by the U.S. Bureau of Labor Statistics, is the primary source of information on U.S. consumption at the household level. Through indepth interviews with a large representative sample of U.S. households, the survey gathers spending data on nearly one thousand different products and services, focusing on consumer units, which are defined as all members of a particular housing unit related by blood, marriage, adoption, or other legal arrangement. The survey collects data only from consumer units that have independent living status. Residents of retirement communities are included in the survey, but long-term-care facility residents are excluded. There are thirteen major expenditure categories in the CEX: food at home, food away from home, alcohol and tobacco, housing, apparel and apparel services, transportation, health care, entertainment, personal care, reading materials and education, cash contributions, personal insurance, and miscellaneous.

Geoffrey Paulin used CEX data from 1984 to 1996 to analyze expenditure trends for different age groups and to test whether tastes and preferences changed over time for older consumers. He found that the change in average annual expenditures between 1984 and 1997 (in 1997 dollars) was as follows: for families under age 65 there was a 1.5 percent increase; for families 65 to 74 there was a 13.6 percent increase; and for families 75 and older there was a 18.0 percent increase. In other words, younger families had relatively stable expenditure levels, but real expenditures rose substantially for those over 65. Paulin also analyzed the change in income before taxes between 1984 and 1997 (in 1997 dollars) and found that families under age 65 had a 10.9 percent increase, families 65 to 74 had a 13.2 percent increase, and families 75 and older had a 7.4 percent increase.

Paulin investigated whether older consumers had different tastes, preferences, or physical needs than younger consumers by analyzing trends for several of the thirteen major expenditure categories. He found that older consumers purchased different amounts than younger consumers, but in most cases the trend of expenditures was similar for older and younger consumers. There was one interesting exception: recreation. All age groups exhibited a real decrease in spending for recreation during the 19901991 recession. In 1997, however, the recreation expenditures for younger consumers were down about 1 percent from their 1991 value, but they had risen substantially for older consumers. In fact, they had risen 19 percent for those age 65 to 74 and 28 percent for those at least 75 years old.

In another study, three of the thirteen CEX categories (food away from home, entertainment, and health care) were selected by Cheryl Russell to highlight changes in average spending by the various age categories between 1987 and 1997 for all households. The data show that most households, especially younger households, reduced spending on food away from home in 1997, compared to 1987. Households 65 to 74 and over 75 had increases of 21.6 percent and 24.2 percent, respectively, for health care expenditures when 1987 and 1997 were compared. Households 75 and over showed greatly increased spending on entertainmentalmost double the 1987 level (up 97.9 percent). It is important to note that Russell compared spending in two different years, 1987 and 1997, while Paulin analyzed the period from 1984 to 1997.

Differences in consumption among older persons

Some researchers have focused more specifically on older households. For example, Mohamed Abdel-Ghany and Deanna Sharpe compared expenditures of households with 65 to 74 year olds and households with members 75 and over using the 1990 CEX data. They found that expenditures for food at home, housing, transportation, and health care made up the largest share of expenses for both groups. Housing was the largest expenditure for each group. Transportation was the second highest expenditure for the 65 to 74 age group, and health care was the second highest expenditure for households 75 and over. In general, urban residents spent relatively more than rural households for food at home, housing, entertainment, and personal care, and relatively less for transportation and health care. The investigators suggested that rural residents may grow some of their own food, reducing expenditure for food at home, and urban residents may use public transportation and have a shorter distance to travel, reducing transportation costs.

Abdel-Ghany and Sharpe also found that households headed by college graduates spent relatively more than households headed by those who did not complete high school on items associated with an active social life: food away from home, alcohol and tobacco, apparel and apparel services, entertainment, and personal care. The results suggest that education influences consumers' tastes and preferences. Racial differences were observed as well. Compared to older white households, older African-American households spent significantly more on personal insurance and significantly less on food away from home and entertainment.

Compared to married-couple households, households headed by unmarried females spent significantly more for apparel and apparel services and significantly less on food at home, food away from home, alcohol and tobacco, health care, and personal care. Spending by households headed by unmarried males varied by age. The 65 to 74 year old unmarried-male households spent relatively more on food away from home, entertainment, and personal insurance; and significantly less on food at home, health care, and personal care, than did married couples age 65 to 74. Households headed by unmarried males age 75 and over spent significantly more on alcohol and tobacco and personal insurance than did married-couple households of the same age.

Based on their results, Abdel-Ghany and Sharpe concluded that households of 65 to 74 year olds had higher marginal propensities to spend for food at home, food away from home, alcohol and tobacco, transportation, entertainment, and personal insurance, compared to the 75 and over group. The 65 to 74 age group had a lower propensity to spend for housing, apparel and apparel services, health care, and personal care, compared to the 75 and over group.

In his study of CEX data from 1984 and 1997, Paulin observed that health care expenditures rose substantially for all groups during this period. His findings showed that those 65 and older, who made up about 20 percent of the sample being studied, accounted for nearly one-third of total health expenditures. He noted that health expenditure shares were most volatile for those age 75 and older. All groups spent a larger share of their health care dollars on health insurance in 1997, compared to 1984. Expenditures for drugs appeared to trend upward slightly as a share of the health care budget, at least for those 65 and older, but shares were most volatile for the 75 and older group.

Differences in consumption by the vulnerable elderly

Using 1990 CEX data, Rose Rubin and Michael Nieswiadomy compared consumption of older households that received cash assistance (food stamps or Supplement Security Income [SSI]) with those that received no cash assistance. The latter group was further subdivided into three income categories: poor (income less than 125 percent of the poverty level), low income (between 125 and 200 percent of the poverty level), and higher income (more than 200 percent of the poverty level). The results suggest that poor and low-income recipients were less well-off than cash assistance recipients. Overall, as income increased, the percentage of the budget spent on food decreased for all income categories. A 1 percent increase in income generated approximately a 1 percent increase in spending on housing for all three income groups. For health care, a 1 percent increase in income generated about a 1 percent increase in spending for the poor, no significant change for the low-income group, and a 0.37 percent increase for those with higher income. Overall, Rubin and Nieswiadomy concluded that poor and low-income older households that did not receive cash assistance were even more financially distressed than those receiving financial assistance: "Their reported current income is substantially lower than their expenditures, so they dissave at unsustainable rates. This annual dissaving, combined with their relatively low levels of financial assets, indicates continuing financial exigency" (Rubin and Nieswiadomy, 1997, pp. 9697).

Differences in consumption by work status

The expenditure patterns of workers and nonworkers were also studied by Thomas Moehrle, who drew his sample of people age 62 to 74 from the 19861987 CEX. Moehrle found that two-thirds of the sample was not working while one-third was still employed. The sample was further subdivided into low-, middle-, and high-income groups. Moehrle found that the head of household of the working households was younger and had attained higher levels of education than their nonworking counterparts. Across all income groups, those not working were more likely to own their homes without mortgages and to spend more on food prepared at home and on health care. Transportation expenditures were higher for the working households, and the working households had notably higher expenditure shares for retirement, pension, and Social Security contributions across all income levels.

Rubin and Nieswiadomy (1994) studied retired and nonretired households headed by those age 50 and older using 19861987 CEX data. Their sample was further subdivided into married couples, single men, and single women. They found that nonretired married couples spent 45 percent more than retired couples, while nonretired single men spent 65 percent more than retired men, and nonretired women spent 50 percent more than retired women. The retired households spent a significantly greater share on food at home, housing, rent, utilities, household operations, and health care. The nonretired households allocated a greater share to food away from home, alcoholic beverages, owned and other dwellings, home furnishings, apparel and services, all transportation categories (except public), entertainment, education, miscellaneous gifts, and insurance. Rubin and Nieswiadomy found that spending on health care was positively correlated with education levels. Retired single women, nonwhites, and those over 75 were given larger shares of cash gifts and contributions.

Volunteer service: time use and consumption

Consumption may also take the form of time used for services that provide for well-being and personal satisfaction. Time-use studies are based on the premise that time outside of work can be used either to earn more money to purchase goods or for volunteer services that provide services to others. Volunteering can be done outside of the home (e.g., Red Cross, church). Caregiving is another example of using time to provide services.

Volunteer service by older persons is a valuable resource. Data from the 1991 Commonwealth Fund Productive Aging Survey (see Caro and Bass, 1995) showed that about one-quarter of persons over 55 were currently doing volunteer work. Volunteering was found to be more common among women, those with education beyond high school, those with professional or technical skills, those in good health, and those active in religion. The proportion of volunteers was highest for older adults age 55 to 59 among whom 31 percent were volunteers. Interestingly, almost one-tenth (9.4 percent) of persons age 85 and over were volunteers, according to the survey.

When the characteristics related to volunteer assignments were analyzed, the following associations were observed by Caro and Bass: Younger respondents were more likely to be engaged in fund-raising. Older respondents were more likely to drive a vehicle. Women were more likely than men to work in an office. Higher-income people were more likely to serve on a board or a committee. Higher levels of education were associated with service on a board or committee, office work, and fund-raising. Better health was associated with direct-service assignments.

Further analysis of the 1991 Commonwealth Fund Productive Aging Survey revealed that those active in religious organizations were more often female, younger, well-educated, and active in religion than the sample as a whole. Older people were more likely to volunteer for a health organization or a senior citizen center. Those with higher levels of education were more likely to volunteer for religious and health organizations, and being in good health was linked to volunteering in the health sector.

Caregiving: time use and consumption

Another important use of time by older persons is the time spent providing care to others. Although studies report somewhat different statistics on caregiving, this is likely to be a reflection of the manner in which caregiving is defined. For example, the Senate Select Committee on Aging (1988) reported that about 80 to 90 percent of elder care is informally provided by the family. The average caregiver is a 57-year-old female, but 36 percent of caregivers are over 65.

Slightly more than one-fourth of those 65 and over who participated in the 1991 Commonwealth Fund Survey reported providing informal assistance to a sick or disabled relative, friend, or neighbor during the previous week (Doty, 1995). About 15 percent of Commonwealth Fund Survey respondents age 65 and older who reported providing care to sick or disabled persons during the previous week reported providing more than twenty hours of care. Forty percent of persons age 65 and older with children reported that they had provided informal assistance of a nonfinancial nature to children, grandchildren, or great-grandchildren during the previous week, and 11 percent reported providing twenty or more hours of help.

According to the 1989 National Long-Term Care Survey (NLTCS), 53 percent of the primary caregivers of the disabled elderly were themselves 65 or older, and 18.7 percent of the care-givers of the disabled elderly were 75 or older. Primary caregivers are defined as individuals who bear most of the responsibility for providing long-term care for a disabled elder.

Data collected for the Assets and Health Dynamics Among the Oldest Old (AHEAD) study on respondents age 70 and older revealed the amount of help provided for people age 70 years and older with activity limitations. Some 51 percent of paid and unpaid caregivers provide help every day, 21 percent provide help several times a week, 14 percent provide help once a week, and 14 percent provide help less than once a week. When help given to people aged 70 and over was measured by hours-per-day, 34 percent received 1 hour of help, 46 percent received 2 to 5 hours, 9 percent received 6 to 10 hours, 4 percent received 11 to 23 hours, and 7 percent received care 24 hours per day. The pool of family caregivers is dwindling, however. In 1990 there were eleven potential caregivers for each person needing care. In 2050 the ratio will be four to one.

The demography of the U.S. population is changing in many ways. The increasing average age of the population is important. As the proportion of older consumers continues to increase, they are likely to account for an increased share of total expenditures. An examination of trends revealed that older consumers were similar to younger consumers in what was purchased. However, Paulin points out that the population of older consumers in 2000 were not members of the baby boom generation. He speculates that there may be more diversity in tastes and preferences as the baby boomers age.

Sharon A. DeVaney

See also Consumer Protection; Economic Well-Being; Life Cycle Theories of Savings and Consumption; Poverty; Retail and Older Adults.


Abdel-Ghany, M., and Sharpe, D. L. "Consumption Patterns Among the Young-Old and Old-Old." Journal of Consumer Affairs 31, no. 1 (1997): 90112.

Ando, A., and Modigliani, F. "The Life Cycle Hypothesis of Saving: Aggregate Implications and Tests." The American Economic Review 53, no. 1 (1963): 5584.

Caro, F. G., and Bass, S. A. "Increasing Volunteering Among Older People." In Older and Active: How Americans Over 55 Are Contributing to Society. Edited by S. A. Bass. New Haven, Conn.: Yale University Press, 1995. Pages 7196.

Deaton, A. Understanding Consumption. Oxford: Oxford University Press, 1992.

DeVaney, S. A. "Economic Status of Older Adults in the United States: Diversity, Women's Disadvantage, and Policy Implications." In Gerontology: Perspectives and Issues, 2d ed. Edited by K. Ferraro. New York: Springer Publishing Company, 1997. Pages 285304.

Doty, P. "Older Caregivers and the Future of Informal Caregiving." In Older and Active: How Americans Over 55 are Contributing to Society. Edited by S. A. Bass. New Haven, Conn.: Yale University Press, 1995. Pages 97102.

Friedman, M. A Theory of Consumption Function. Princeton, N.J.: Princeton University Press, 1957.

Harrison, B. "Spending Patterns of Older Persons Revealed in Expenditure Survey." Monthly Labor Review, 109, no. 10 (1986): 1517.

Hitschler, P. B. "Spending by Older Consumers: 1980 and 1990 Compared." Monthly Labor Review, 1116, no. 5 (1993): 313.

Institute for Health and Aging. Chronic Care in America. University of California-San Francisco for the Robert Wood Johnson Foundation, 1996.

Jackson, M. E. Informal Care of the Disabled Elderly: A Research and Policy Initiative. Final report to the Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human Services. Lexington, Mass.: SysteMetrics, 1992.

Moehrle, T. "Expenditure Patterns of the Elderly: Workers and Nonworkers." Monthly Labor Review 113, no. 5 (1990): 3441.

National Academy on an Aging Society. "Caregiving: Helping the Elderly with Activity Limitations." Washington, D.C.. May, no. 7 (2000): 16.

Paulin, G. D. "Expenditure Patterns of Older Americans, 19841997." Monthly Labor Review 123, no. 5 (2000): 328.

Rubin, R., and Nieswiadomy, M. L. "Expenditure Patterns of Retired and Nonretired Persons." Monthly Labor Review (1994): 1021.

Rubin, R., and Nieswiadomy, M. L. Expenditure of Older Americans. Westport, Conn.: Praeger Publishers, 1997. Volume 117, no. 4.

Russell, C. "The New Consumer Paradigm." American Demographics April (1999): 5058.

Select Committee on Aging, U.S. House of Representatives. Exploding the Myths: Caregiving in America. Washington, D.C.: U.S. Government Printing Office, 1988.

Stone, R. "Defining Family Caregivers of the Elderly: Implications for Research and Public Policy." The Gerontologist 27 (1991): 616626.