April 1999

Income of the Elderly, 1997

 

• The average income of an elderly individual (age 65 or older) in the United States in 1997 was $19,104, up from $12,239 in 1974, based on Employee Benefit Research Institute (EBRI) tabulations of data from the March 1997 Current Population Survey (CPS).

• The percentage of the elderly's income derived from Social Security declined from 42.0 percent in 1974 to 38.6 percent in 1989. It increased to 44.4 percent in 1994 and declined to 41.4 percent in 1997. In 1997, the average income received by the elderly from Social Security was $7,906.

• Income from pensions and annuities accounted for a steadily increasing share of the elderly's income from 1974 to 1994. In 1974, these sources accounted for 14.0 percent of the elderly's income; by 1994, that percentage had increased to 20.0 percent. In 1997, pensions and annuities accounted for 19.6 percent of the elderly's income, and the average amount an elderly person received in income from pensions and annuities was $3,737.

• Income from assets increased between 1974, when it accounted for 18.2 percent of the elderly's income, and 1984, when it accounted for 28.2 percent. It declined to 25.2 percent in 1989 and to 20.4 percent in 1997. In 1997, the average amount an elderly person received in income from assets was $3,904. Assets here are defined as stocks and bonds not held in a retirement account, income from rents, royalties, and trusts.

• Income from earnings declined as a percentage of the elderly's income from 21.3 percent in 1974 to 14.9 percent to 1994. In 1997, income from earnings increased to 16.6 percent of the elderly's income. In 1997, the average amount an elderly person received in income from earnings was $3,178.

• The lower an individual's total income, the greater the percentage of it that comes from Social Security. In 1997, Social Security accounted for 87.9 percent of the total income of elderly individuals in the lowest income quintile, compared with 21.0 percent for those in the highest income quintile.

• The income quintiles are derived by assigning all of the individuals age 65 and older to one of five groups based on income so that there is an equal number of individuals in each group. The income ranges of the quintiles vary year to year. In 1997, the lowest income quintile consisted of individuals with a 1997 income of $6,323 or less. The highest income quintile consisted of individuals with a 1997 income of $24,799 or higher.

• There was a significant difference between the average income of elderly men ($25,669) and that of elderly women ($14,320) in 1997. This is in part attributable to greater lifetime attachment to the work force among the generation of men that currently makes up the elderly. Given the increasing role of women in the work force, the income gap between the sexes will narrow.

• Elderly women derived a greater share of their income from Social Security and assets than men in 1997. Social Security accounted for 48.5 percent of elderly women's income, compared with 35.9 percent of elderly men's income. Income from assets accounted for 25.3 percent of elderly women's income, compared with 16.7 percent of elderly men's.

• Elderly men derived a larger share of their income from employment-based sources, including pensions and annuities and earnings, than elderly women. In 1997, pensions and annuities accounted for 23.8 percent of elderly men's income, compared with 14.0 percent of elderly women's. Income from earnings accounted for 21.3 percent of the elderly men's income, compared with 10.6 percent of elderly women's.

• These data are from the March supplement to the CPS. Some research has shown that the March CPS underestimates the percentage of the elderly's income that comes from employment-based pension plans. This research also uses data from the Bureau of Economic Analysis' National Income and Product Accounts, and from the Internal Revenue Service.(1)

For more information, contact Ken McDonnell (202) 775-6342 or mcdonnell@ebri.org.

Source: Employee Benefit Research Institute tabulations of data from the March 1998 Current Population Survey.

(1) For a more detailed discussion of this point, see Sylvester J. Schieber, Why Do Pension Benefits Seem so Small?, Watson Wyatt Worldwide (Washington, DC, 1995).

4/99