Printer Friendly

Expenditure patterns of the elderly: workers and nonworkers.

Spending patterns of older households differ not only by income, but according to work status; older workers allocate more to retirement, pension, and Social Security funds, while the nonworking elderly spend more on food prepared at home and health care

The Nation's population continues to grow older. Recent information from the U.S. Bureau of the Census estimates that people aged 65 and over will make up more than 23 percent of the population in 2030, up from 12 percent in 1985.' Like other cohorts, the elderly have different spending patterns depending upon level of income and status as employed or retired. According to data from the 1986-87 Bureau of Labor Statistics Consumer Expenditure Survey, in terms of share of expenditure and separated into three income levels (low income = less than 15,000; medium income = $15,000 to $29,999; high income = $30,000 and over):

Nonworking elderly households spend more on food prepared at home than do working elderly households, regardless of income level. High-income nonworking elderly households spend more on housing-particularly for utilities, fuels, and public services and for housefurnishings and equipment-than do their low-income counterparts.

Low-income working elderly households spend more on transportation than do low-income nonworking elderly households.

Regardless of income level, nonworking elderly households spend more on health care than do working elderly households.

Working elderly households spend more on retirement, pension, and Social Security contributions than do nonworking elderly households, across all income levels.

Coupled with the well-known fall in the U.S. birth rate, the aging population will cause a decline in the growth, as well as the age distribution, of the labor force. An examination of the demographic characteristics of the labor force reveals, for example, that the century-long decline in the participation rate of older people has been moderating. In fact, new labor force projections to the year 2000 show that the participation rate of women in the 55-to-74 age group will increase. Also, although participation by men in the 55-and-over age group is likely to continue to decline, the labor force separation of these men will not be as significant as it has been in years past.(1)

Along with the aging of the population and the changes in labor force participation rates of older people, changes are expected in the spending habits of the U.S. population. This article examines the differences in expenditures between the working and nonworking elderly households.


Consumer Expenditure Survey data were used to calculate mean annual expenditures, income, and demographic characteristics for selected elderly U.S. consumer units.1 The survey is a household survey in which family expenditures are collected. The analysis presented here is of participants from the 1986 - 87 survey years.

An earlier study by Beth Harrison (4) examined the spending patterns of consumer units with reference persons aged 65 and over.5 Breaking these up into two age groups, 65 to 74 and 75 and over, Harrison found that, although persons aged 65 and over are commonly viewed as a single homogeneous group, there are identifiable differences in expenditures, incomes, and characteristics between the two age groups. Nonetheless, in the study presented in this article, consumer units with reference persons over age 74 are not included, because 93 percent of these reference persons are retired and because their spending patterns are greatly different from those of the 74-and-under group. Of course, the older group could be separated into working and nonworking subgroups, but then the working subgroup would have too few observations to be of sound statistical use.

Information from the Social Security Administration shows that the average age of retired persons awarded Social Security from 1967 to 1987 has been declining.6 More important, the percentage distribution of retired persons who are between the ages of 62 and 64 has increased noticeably compared to all other retirees. For instance, in 1967, men aged 62 to 64 who were Social Security retirement recipients made up 35.7 percent of the number of men who had retired that year. In 1987, this same group made up 67.1 percent of the number of men who had retired in that year. (See table 1.) Similar changes have occurred for women as well. From these results, it seems evident that more workers have decided to take advantage of early retirement. Families with reference persons aged 62 to 74, rather than the customary 65 to 74, are therefore included in this analysis.

Separation of consumer units into working and nonworking subgroups was based on the work status of the reference person. If the reference person received earnings from part-or full-time employment in the 12 months prior to his or her interview, the consumer unit was classified as a working household. Otherwise, the unit was classified as a nonworking household, even if one or more members other than the reference person were employed. Consumer units with reference persons who considered themselves involuntarily unemployed or who were working without pay were excluded from the study. These units were a small percentage of the units examined; thus, excluding them should not affect the results of the analysis.

Two-thirds of the consumer units examined fell into the nonworking group. Of these, 79 percent had reference persons who classified themselves as retired. The remainder of the nonworking group consisted of reference persons who considered themselves disabled, taking care of family or home, going to school, or doing something else. Many of these persons are elderly single women who never have worked and, therefore, do not consider themselves retired. Often, they are widows receiving Social Security payments from their deceased husbands' employment.

Because income has such an important influence on spending habits, households were further divided into low-, medium-, and high-income groups. Low income was defined as an annual after-tax household income of less than $15,000. Medium household income was defined as income between $15,000 and $29,999 annually, and high household income was defined as $30,000 or more per year.


Examining household characteristics affords insight into how households spend their income. Table 2 presents selected characteristics of elderly households, including weighted U. S. averages where appropriate.(7) Computed means of expenditures, also weighted averages, are shown in table 3. Two-thirds of the households with reference person aged 62 to 74 are classified as nonworking.

The reference persons of the working group were younger and had attained higher levels of education than their nonworking counterparts. Higher education is usually associated with higher labor earnings, and with higher labor earnings, the opportunity costs of retirement are greater. Hence, the greater concentration of more educated reference persons within the working group may indicate that retirement is being postponed among these younger individuals.

Among all three income divisions of the participants, the working groups had higher incomes. Within the low- and middle-income groups, the income differences between the nonworking and working were not notable. However, substantial differences existed within the high-income group, chiefly because that group has no upper income limit in its definition. Accordingly, the within-group variation for the high-income group will be higher compared to that for the other two groups, and caution is suggested when evaluating the expenditure differences between the high-income working and nonworking households.

The working groups were more likely than the nonworking groups to have multiple vehicle ownership and more likely to own at least one vehicle. Because there often were more earners than just the reference person in these households, and because workers need daily transportation to and from their workplaces, frequent demands for transportation are expected in these households. Within the lower income group, 85 percent of the working households owned at least one vehicle, compared to 72 percent of the nonworking households.

In general, older households have higher rates of home ownership as compared to younger households. For instance, data published from the 1986-87 Consumer Expenditure Survey show that 78 percent of households with reference person over the age of 65 own their homes, compared to 57 percent of all other households. Across all the income groups studied in this article, the nonworking were more likely to own their homes without mortgages.


The mean annual expenditures by income and work status from table 3 were used to compute expenditure shares in table 4. Then, from these shares, T-statistics were computed to determine whether there were any differences between expenditure shares within each of the income and work status groUpS.8 A positive T-statistic greater than two indicates that the nonworking group spent a larger share on the expenditure item at the 5-percent level of significance.

From table 4, it is plain that, across all income groups, nonworking households spent more than working households on food prepared at home. Also, high-income nonworking households allocated a larger share than did high-income working households for total housing expenditures, even though they were more likely (61 percent versus 41 percent) to own their homes mortgage free. In the low- and middle-income groups, working households spent a larger share on total housing expenditures than did nonworking households.) This is perhaps accounted for by the significantly higher share the high-income nonworking households spent on utilities, fuels, and public services, and on housefurnishings and equipment. Because the latter category includes infrequently purchased and large-ticket-price items, expenditure share differences in this category are likely related to factors not examined here, such as household inventory. By contrast, utilities and the like are usually frequently purchased items that would not be expected to vary significantly within an income group. Because the nonworking elderly households are probably made up of one or more retirees, these households contain people who are home more often and use more utilities, fuels, and public services.

Among lower income elderly households, transportation expenditures were significantly higher for the working, as opposed to the nonworking, group. This is to be expected because of the greater incidence of vehicle ownership within the working group. A decision must be made concerning the mode of primary transportation, and, among low-income working elderly, a higher percentage of households chose ownership of a vehicle because of the frequent demand for transportation. Higher vehicle ownership, coupled with frequent use, increased the transportation outlays for working households.

Health care expenditures reported in the Consumer Expenditure Survey are out-of-pocket expenditures. Hence, those who are uninsured or who must pay their own insurance premiums will necessarily spend more on health services and products. Across all three income groups, the expenditure shares for health insurance were significantly higher for the nonworking group. Because these households generally are not covered by an employer-paid health care package, they must provide coverage themselves or obtain coverage under public plans such as medicare. Medicare and other public health insurance programs, however, are not a complete substitute for employer-paid coverage; consequently, we observe higher shares by the nonworking households. (Of course, payments for physicians' services under medicare are the same for all participants and, therefore, would constitute a lower share for the higher income group.) Also, nonworking households are more likely to purchase supplemental insurance, and persons who retire between ages 62 and 65 are not eligible for medicare and so may pay more for private insurance during those years. Finally, related to health care, middle-income nonworking elderly households spent a higher share for prescription drugs and medical supplies than did their working counterparts, probably because medicare does not cover these drugs and supplies.

Across all income levels, the working group had notably higher expenditure shares for retirement, pension, and Social Security contributions. The reason for this is obviously that older nonworking individuals will not be contributing these expenditures anymore. Note, however, that expenditures for these items are not expected to be zero for the nonworking elderly because some members of the household may be employed and contribute through their work to Social Security and individual retirement funds.


Shares of sources of income are shown in table 5. Regardless of income level, the working group received most of its income from wages and salaries, while the nonworking received the bulk of its income from Social Security and retirement benefits. In the lower income class, the working group also depended heavily on Social Security and retirement benefits, accounting for 51 percent of income, as against 41 percent for wages and salaries. This heavy dependence on retirement benefits probably means that many low-income households were semiretired.

The share of income from wages and salaries was greatest for the high-income group. Just under 60 percent came from this source for those who worked. Even high-in-come nonworking households received 33 percent of their income from wages and salaries. It is readily apparent that the high-income group retained substantial earning power.

Unlike the lower income groups, the high-income nonworking relied heavily on interest, dividends, and rental income. This source accounted for 25 percent of the group's total income. Absent other considerations, individuals in these households may have been able to retire earlier than their low-income counterparts because of their substantially higher unearned income. In a similar manner, self-employment income was a small fraction of total income for all households except the high-income working, accounting for 15 percent of their income. Though obviously not contributory to early, complete retirement, self-employment does promote a degree of independence, both personal and financial.


This study has identified differences in expenditure patterns between working and nonworking older consumer units. These differences cannot be explained solely by income differences, particularly for the lower income households. For instance, low-income working households spent 25 percent more than their nonworking counterparts, even though their incomes were only 12 percent higher. Of course, some of the expenditure variation between working and nonworking households may be due to income, especially in the high-income group, where a significant difference in income already exists between working and nonworking households. For all three income levels, however, some expenditure variations can be explained by the difference in work status. Among these variations are expenditures for food prepared at home, health insurance, and Social Security and retirement contributions.


1 Population Estimates and Projections, Series P-25, No. 952, U. S. Department of Commerce, Bureau of the Census.

2 For an overview of labor force projections by demographic characteristics, see Howard N Fullerton, Jr. "New labor force projections, spanning 1988 to 2000," Monthly Labor Review, November 1989, pp 3-12.

3 A consumer unit is either (1) all members of a household who are related by blood, marriage, adoption, or other legal arrangements; (2) two or more persons living together who pool their incomes to make joint expenditure decisions; or (3) a person living alone or sharing a household with others, or living as a roomer in a private home or lodging house or in permanent living quarters in a hotel or motel, but who is financially independent. A person is considered financially independent if he or she provides the income for at least two of the three major expense categories of housing, food, and all other living expenses. The terms consumer unit and family are used interchangeably throughout.

4 Beth Harrison, "Spending patterns of older persons revealed in expenditure survey," Monthly Labor Resview, October 1986, pp. 15-17.

5 In the Consumer Expenditure Survey, the reference person of a consumer unit is the first member named by the respondent when asked to Start with the name of the person or one of the persons who owns or rents the home.". The relationships of the other members of the consumer unit are then determined with respect to the reference person.

6 Social Security Bulletin, Annual Statistical Supplement (Social Security Administration, 1987), table 53.

7 For information on the weighting procedure used in the Consumer Expenditure Survey, consult the Handbook of Methods, Bulletin 2285 (Bureau of Labor Statistics, 1988), Chapter 18.

8 Standard errors used in T-statistic computations have been replaced by the coefficients of variation (cv's). The variances used in computing the cv's are calculated from mean expenditures using a pseudoreplication technique discussed in Philip J. McCarthy, Data from Complex Surveys, Series 2, No. 14 (Washington, DC, Government Printing Office, 1966) ; and Philip J. McCarthy, Pseudoreplication, Further Evaluation and Application of the Balanced HalfSample Technique, Series 2, No. 31 (Washington, DC, Government Printing Office, 1969).

9 Because we are testing differences between shares and not means, the cv's have been adjusted for shares and pooled. If a is the mean expenditure for an item for the nonworking group, k is the total expenditures for that group, b is the mean expenditure for the item for the working group, and 1 is the total expenditures for the same group, then the pooled cv's for the shares are (a/k).sup.2[(cv(a)).sup.2 + cv(k)).sup.2 -2(a/k)((cv(a)).sup.2)] +(b/l).sup.2[(cv(b)).sup.2 +(cv/l)).sup.2 -2(b/l)((cv(b)).sup.2)].

The role of public policy

. . the United States continues to make labor market policy as though worker is a masculine noun. Employed mothers in the United States are expected to "make it" under present rules and conditions, coping as best they can. American fathers are expected to sustain their primary investment in work. But the practice of both parents taking on the traditional male work role-each working full-time and not taking time off to meet child care needs-would be patently unacceptable in Sweden. Although most Swedes value work and occupational achievement as highly as most Americans, they value home and family to an even greater extent. From the Swedish viewpoint, women cannot merely be assimilated into the traditional male world of work. Rather, this world must be recast in ways that permit fathers as well as mothers to participate equitably in the productions" of human beings and mothers no less than fathers to participate equitably in the production of goods and services.
COPYRIGHT 1990 U.S. Bureau of Labor Statistics
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Moehrle, Thomas
Publication:Monthly Labor Review
Date:May 1, 1990
Previous Article:Worker displacement in a period of rapid job expansion: 1983-87.
Next Article:The 1987-88 surge in exports and the rise in factory jobs.

Related Articles
Expenditure patterns of retired and nonretired persons.
Experimental price index for elderly consumers.
Imputing income in the Consumer Expenditure Survey.
Elderly and nonelderly expenditures on necessities in the 1980s.
Health care and prescription drug spending by seniors: spending for health care and for prescription drugs among seniors has increased over the...
Economic inequality through the prisms of income and consumption: between 1981 and 2001, economic inequality among groups in the general population...
Low-wage labor markets.
How do older Americans spend their time? Older Americans' time use changes dramatically with age, but it is the lower employment rates at older...
Fiscal impacts of an aging population in Montana.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters