GO TO: Medicare HOME : Back to Summary

Income and Assets of the Elderly and Near Elderly

I. Income

Several sources of data on the past and current income of elderly and near elderly households are available. A commonly used source is the Current Population Survey (CPS) conducted annually by the Bureau of the Census. CPS data show that over the 1967-1996 period mean incomes for households in all age groups increased substantially in current dollar terms but modestly in constant 1996 dollars (adjusted to remove the effect of inflation over the years) (see Figure 1). Except for the youngest households, mean incomes tended to rise during the first 20 years but to be relatively flat during the last 10 years of the period.

Source: Graphic presentation of Bureau of the Census CPS data.

Figure 1 illustrates that income is not evenly distributed across age groups. The youngest age group is just beginning their working years and tend to have entry level jobs with low wages. Income increases as people gain experience and some of those in the age group enter mid- and upper-level positions. In the oldest age groups, income declines as people retire. Income is also not evenly distributed within age groups. For this reason, median income is often used when discussing incomes. Figure 2 shows household median income in constant dollars by age group from 1967 to 1996. For most age groups, median incomes increased over the 30 year period.

Source: Graphic presentation of Bureau of the Census CPS data.

The CPS data show that mean income for elderly households ($29,280 in 1996) was significantly higher than median income ($19,448 in 1996). Figure 3 shows the distribution of income in 1996 across elderly households.

Source: Graphic presentation of Bureau of the Census CPS data.

The growth in real income for the elderly during the 1970s and early 1980s helped lift many of them out of poverty. Table 1 compares the percentage of the elderly living at various multiples of the poverty level in 1968 and 1997 and illustrates the dramatic improvement as far as the elderly and poverty are concerned.

Table 1: Percent Distribution of the Noninstitutionalized Elderly Who Receive Social Security Retirement or Survivors Benefits by Family Income as a Percentage of the Poverty Level^a

Source: Presentation by Mark V. Pauly to Commission on August 10, 1998, from his tabulation of CPS data.

The Social Security Administration’s Office of Research, Education, and Statistics tabulates CPS data on the income of the elderly and near elderly and publishes them every two years with the most recent publication covering 1996. The tabulations show that as people grow older the portion of household income derived from working decreases and the portion from retirement benefits (mainly Social Security) increases (see Figure 4). Also, as people age, the percentage of income from assets increases.

Source: Graphic presentation of data from Social Security Administration’s tabulations of CPS data for households age 55 and older.

II. Assets

Information has also been gathered about the asset holdings of the elderly and near elderly. One source is the Survey of Consumer Finances conducted every three years by the Federal Reserve Board. Figure 5 presents information from the last three surveys conducted in 1989, 1992, and 1995 on median family net worth by age of the family head. Mean family net worth was much higher in all cases than median net worth; for example, mean net worth was $205,900 for all families in 1995 and median net worth was $56,400, and the same figures for the 65 to 74 age group were $331,600 and $104,100. Both mean and median net worth were higher for elderly households than younger households.

Source: Graphic presentation of data from the Federal Reserve Board’s Survey of Consumer Finances.

Another source of data on income and assets is the Survey of Income and Program Participation conducted by the Bureau of the Census, Department of Commerce. Census has reported data from this survey on estimated household net worth by age of the household head. Figure 6 presents these estimates in graphic form showing 1993 net worth by age of the household head for those falling into the income quintiles. As the head of household gets older, net worth tends to increase. As net worth increases, the portion represented by home equity tends to decrease. However, home equity represents the majority of net worth for all except for those households that fall into highest quintile in the age 65 and over groups and for a few quintiles in the under age 45 groups.

The income quintiles in Figure 6 are for the population as a whole and do not mean that a fifth of each age group is in each quintile because incomes are not evenly distributed across ages. Older households tend to have lower incomes than other households. Table 1 shows the distribution of households across the income quintiles by age of the household head. If income were evenly distributed across age groups, all values in Table 2 would be 20 percent.

Table 2: Percentage of Households Falling Into Income Quintiles by Age of Household Head, 1993

Source: Department of Commerce data from the Survey of Income and Program Participation.

An ongoing, extensive data collection effort in the area of income and assets of the elderly and near elderly is the Health and Retirement Survey (HRS) of people who were between the ages of 51 and 61 in 1992 and its sister survey, the Asset and Health Dynamics Among the Oldest Old (AHEAD), that interviewed people 70 years of age or older in 1993. Researchers have completed three waves of HRS and two waves of AHEAD interviews. Both surveys are conducted by the Survey Research Center at the University of Michigan under funding from the Administration on Aging and several other agencies.

Unpublished data from these surveys show that in 1996 the HRS group, who were then 55 to 65 years of age, had average household total assets (in 1991 dollars) of $273,000 with $70,000 of this amount represented by housing assets. The median household had total assets of $124,000. The household at the 5^th percentile had no assets while that at the 95^th percentile had assets of $925,000, of which $210,000 was housing. Married couples had much higher assets than single people, roughly three times higher, and single men in general had more than 50 percent more assets than single women did.

In 1995, the AHEAD group, who were then 72 or older, had average household total assets (in 1996 dollars) of $128,000 with average housing assets of $65,000. Median household assets totaled $62,000. The household at the 5^th percentile had no assets while that at the 95^th percentile had assets of $482,000, of which $158,000 was housing. Assets tended to decrease with age, to be higher for men than women, and to be lower for Blacks and Hispanics.

III. Projections

Relatively little is available in the literature regarding projections of future income for the elderly, especially recent projections. The projections generally show a low percentage of the elderly living below the poverty level ($8,290 for an individual and $11,090 for a couple in 1998); for example, an unpublished 1990 Lewin/ICF estimate projected that only 7 percent of the elderly residing in the community in 2020 will have incomes below the poverty level while 75 percent will have incomes of at least twice the poverty level. Comparable figures cited for 1990 were 17 percent below the poverty level and 52 percent at least twice the poverty level.

In a 1994 report for AARP^a, Lewin-VHI estimated that baby boomers would have median family income in 2010, the year before the first boomers turn 65, of $47,198 (1990 dollars). This report also estimated that the ratio of income of the lowest 20 percent of baby boomer households to the highest 20 percent would increase from 3.8 in 1990 to 4.4 in 2010. Lewin-VHI also projected that in 2030, 74 percent of boomers will have income from Social Security, pensions, and assets while 23 percent will have income from two of these sources and only 3 percent will have income from only one source. Median income in 2030 for these three groups (in 1990 dollars) was estimated at $32,935, $18,400, and $10,150, respectively.

Another Lewin-VHI report for AARP^b made projections of private pension plan income for the elderly. According to this report, 50 percent of the people age 66 to 84 received employer-provided pension benefits in 1990 with median pension income for those with pensions equal to $6,600 or 27 percent of income. For the total population age 65 or older, 14 percent received pensions of less than $3,000, 10 percent from $3,000 to $6,000, 10 percent from $6,000 to $10,000, and 16 percent $10,000 or more. For 2030, the report estimated that 83 percent of baby boomers will receive private pension benefits and that the median benefit for those covered by plans will be $6,918 in 1990 dollars. Pension benefits were estimated to represent 16 percent of income at the median. An estimated 25 percent of the total aged population will receive pensions of less than $3,000, 28 percent from $3,000 to $10,000, and 30 percent $10,000 or more (all in 1990 dollars).

Source: Income Security and Health Care: Economic Implications 1991-2020, 1991 Advisory Council on Social Security.

A report prepared for the 1991 Advisory Council on Social Security by an expert panel, entitled Income Security and Health Care: Economic Implications 1991-2020, included projections of both income and assets for the elderly. Figure 7 shows the projection of median total annual income by age group and Figure 8 shows the percentage of elderly households projected to have financial assets in selected ranges.

Source: Income Security and Health Care: Economic Implications 1991-2020, 1991 Advisory Council on Social Security.

The Urban Institute has an ongoing project, funded by the Social Security Administration, to build a model to project income and assets for the elderly well into the next century. The results of this project are expected to be available in the Fall of 1998.

     ^bLewin-VHI, Inc. Baby Boomer Pension Benefits: Will They be Adequate for the Future?, Washington, DC, 1994. The PRISM model was also used in this report for projections.

Back to Top