Order Code RL33387
CRS Report for Congress
Received through the CRS Web
Topics in Aging: Income of Americans
Age 65 and Older, 1969 to 2004
April 20, 2006
Patrick Purcell
Specialist in Social Legislation
Domestic Social Policy Division
Debra B. Whitman
Specialist in the Economics of Aging
Domestic Social Policy Division
Congressional Research Service ˜ The Library of Congress

Topics in Aging: Income of Americans
Age 65 and Older, 1969 to 2004
Summary
The aging of the American population and the retirement of the baby boom
generation will place financial strains on Social Security, public and private pensions,
and on retirees’ personal savings. Since the 1960s, birth rates have fallen and
average life expectancy has increased. Consequently, the number of workers relative
to the number of retirees is projected to decline, and retirees will have to stretch their
savings and other assets over longer periods of retirement than their parents and
grandparents experienced. This CRS report presents data collected by the Census
Bureau from 1969 through 2005 that describe how the demographic traits,
employment patterns, and the sources and amounts of income of people 65 and older
have changed over a period of nearly 40 years.
America’s elderly today are older, more racially and ethnically diverse, better
educated, and less likely to be widowed than the elderly population of the late 1960s.
The increasing number of Americans living to age 80 and older is of particular
significance because it is the very old who are most likely to need medical, social,
and long-term care services, and who are at the greatest risk of depleting their
financial resources and slipping into poverty.
Rates of employment among older persons have been rising in recent years.
Employment rates fell among men 55 and older from the late 1960s to the early
1990s. Since then, employment rates have risen for older men, but they remain
below the employment rates of the 1970s. Among older women, employment rates
have steadily increased, but older women’s employment rates today remain lower
than those of men the same age. Earnings, Social Security, pensions, and income
from assets comprise the majority of income among people 65 and older today, just
as they did 35 years ago. Wages and salaries today constitute a greater proportion of
the aggregate income of the elderly than at any time since the late 1970s. Since the
early 1990s, both the share of aggregate income received from assets and the
percentage of older persons with asset income have fallen. Social Security has
remained a fairly stable share (about 40%) of the total income of people 65 and older.
The percentage of men receiving pension income rose until the early 1990s and then
fell. More women today have pension income, but their average pension is half the
amount that men receive.
Median household income rose faster among elderly households than among
nonelderly households from 1969 through 2004, but in 2004 the median annual
income of households in which the householder or spouse was 65 or older ($25,210)
was just half of the median income of younger households ($50,466). The rise in the
income of elderly households coincided with a decline in the poverty rate among
people 65 and older from 25% in 1969 to just 10% in 2004. Also since the 1960s,
the rate of decline in median personal and household income that occurs as people
pass age 60 and begin to retire has slowed. Growth in real wages — and in the Social
Security and pensions based on these wages — has contributed greatly to the
economic well-being of today’s retirees. This report will be updated every two years.

Contents
Background: America’s Aging Population . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Americans Are Living Longer than Ever Before . . . . . . . . . . . . . . . . . . . . . . 1
The Income of Older Americans and the Federal Budget . . . . . . . . . . . . . . . 2
Trends in the Income of Older Americans . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
The Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Demographic Profile of the CPS Population Age 65 and Older . . . . . . . . . . . . . . 5
Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Sex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Race and Ethnicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Marital Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Employment Among People Age 65 and Older . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Rates of Employment, by Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Employment Rates among Men and Women . . . . . . . . . . . . . . . . . . . . . . . 11
Full-Time vs. Part-Time Employment among Older Workers . . . . . . . . . . 15
Sources and Amounts of Income Among People 65 and Older . . . . . . . . . . . . . . 15
Earnings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Social Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Asset Income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Pension Income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Welfare Income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Other Income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Sources of Income, by Quartile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Sources of Income Differ Between Men and Women . . . . . . . . . . . . . . . . . 25
Household Income of People 65 and Older . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Median Household Income, by Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Household Income at the 75th and 25th Percentiles . . . . . . . . . . . . . . . . . . . . 30
Poverty Status of the Elderly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Decline in Poverty Among the Elderly . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Poverty among Demographic Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Changes in Income as People Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Changes in Individual and Household Income . . . . . . . . . . . . . . . . . . . . . . 36
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Appendix: Historical Tables of Individual and Household Income . . . . . . . . . . 42

List of Figures
Figure 1. Number of Older Americans, 1969-2005, by Age Group . . . . . . . . . . . 2
Figure 2. Employment Rates of Men by Age, 1969-2005 . . . . . . . . . . . . . . . . . 13
Figure 3. Employment Rates of Women by Age, 1969-2005 . . . . . . . . . . . . . . . 14
Figure 4. Share of Income by Source among People 65+, 1969-2004 . . . . . . . . . 20
Figure 5. Sources of Income by Quartile, 1975 . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Figure 6. Sources of Income by Quartile, 2004 . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Figure 7. Percentage of Men and Women Age 65+ with Income from Each
Source, by Year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
List of Tables
Table 1. Projections of the Resident U.S. Population, by Age . . . . . . . . . . . . . . . 2
Table 2. Age Distribution of People Age 65 and Older, 1970-2005 . . . . . . . . . . . 6
Table 3. Race and Sex of Individuals Age 65 and Older, 1970-2005 . . . . . . . . . . 7
Table 4. Marital Status of Men and Women Age 65 and Older, 1970-2005 . . . . 8
Table 5. Educational Attainment of Individuals Age 65 and Older, 1970-2005 . 9
Table 6. Employment Rates, by Age, 1969 to 2004 . . . . . . . . . . . . . . . . . . . . . . 11
Table 7. Employment Rates Among Men and Women, by Age . . . . . . . . . . . . . 12
Table 8. Full-Time and Part-Time Employment, by Age of Worker . . . . . . . . . 15
Table 9. Percentage of People Age 65 and Older with Income from Each
Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Table 10. Income from Each Source as a Percentage of Total Income Among
People 65 and Older . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Table 11. Mean Annual Income of People 65 and Older from Each Source . . . 19
Table 12. Median Annual Income of Individuals 65 and Older from
Each Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Table 13. Median Income of Men and Women Age 65 and Older from
Each Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Table 14. Average Household Size by Year and Age of Householder . . . . . . . . 28
Table 15. Mean and Median Household Income, by Age of Householder . . . . . 29
Table 16. Percentage Change in Median Household Income, 1969 to 2004,
by Age of Householder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Table 17. Household Income at 75th and 25th Percentiles, by Age of
Householder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Table 18. Cumulative Percentage Change in Household Income at 75th
and 25th Percentiles, by Age of Householder . . . . . . . . . . . . . . . . . . . . . . . . 31
Table 19. Poverty Status of Individuals Age 18 and Older, by Year . . . . . . . . . . 33
Table 20. Poverty Status of Individuals Age 18 and Older in 2004,
by Demographic Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Table 21. Median Individual Income at Five-Year Intervals . . . . . . . . . . . . . . . 39
Table 22. Median Household Income at Five-Year Intervals . . . . . . . . . . . . . . . 40
Table A1. Percentage of People Age 65 and Older with Income from
Each Source, by Year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Table A2. Median Annual Income of Individuals Age 65 and Older,
by Income Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Table A3. Mean and Median Annual Household Income in 2004 Dollars . . . . . 45

Topics in Aging: Income of Americans
Age 65 and Older, 1969 to 2004
Background: America’s Aging Population
The aging of the American population and the impending retirement of the
“baby boom” generation will place significant strains over the next several decades
on Social Security and on retirees’ own financial resources. The decline in birth rates
since the 1960s and increases in life expectancy will result in fewer workers relative
to the number of retirees.1 Consequently, Social Security benefits will have to be
financed by a working population that is shrinking relative to the number of retirees
and retirees will have to stretch their savings and other assets over longer periods of
retirement than their parents and grandparents experienced.
Americans Are Living Longer than Ever Before
The average life expectancy of Americans born in 1969 was 70.5 years. It has
been estimated that those who were born in 2003 will live for an average of 77.5
years.2 Women continue to have a longer average life expectancy than men, but both
men and women have experienced gains in average life expectancy since the 1960s.
A man who reached age 65 in 1969 could expect to live another 13.0 years, whereas
a woman who turned 65 in 1969 had a remaining life expectancy of 16.5 years. A
man who reached age 65 in 2003 could expect to live another 16.8 years, whereas a
woman who turned 65 in 2003 had a remaining life expectancy of 19.8 years. As
more people live into old age, the age-profile of the population will shift. In 1969,
18.6 million people in the United States — 9.4% of the population — were age 65
or older. In 2005, there were 36.7 million Americans age 65 or older, representing
12.4% of the population. (See Figure 1.) By 2025, according to projections made
by the Bureau of the Census, there will be 63.5 million people age 65 or older,
comprising 18.2% of the U.S. population. (See Table 1.)
These demographic trends will strain the components of the traditional “three-
legged stool” of retirement income: Social Security, pensions, and personal saving.
The Social Security Board of Trustees has estimated that the Social Security trust
fund will be exhausted by 2041 unless actions are taken to preserve it.3 Pensions are
the second largest source of income among the elderly, after Social Security, but only
1 For more information, see CRS Report RL32981, Age Dependency Ratios and Social
Security Solvency
, by Laura Shrestha.
2 U.S. National Center for Health Statistics, Vital Statistics of the United States.
3 Social Security and Medicare Boards of Trustees, Status of the Social Security and
Medicare Programs: A Summary of the 2005 Annual Reports
, March 2005.


CRS-2
about half of all workers in the United States have pension coverage through their
jobs. Moreover, the traditional pension that provides a lifelong annuity is becoming
less common. Today, more workers participate in savings and thrift plans than in
traditional pension plans. A key characteristic of these savings plans is that the
worker must actively participate, deciding whether to contribute to the plan, how
much to contribute, and how to invest the funds. Workers who do not choose to
save, or who save too little, may face difficult financial circumstances in retirement.
Figure 1. Number of Older Americans,
1969-2005, by Age Group
(in thousands)
Source: CRS analysis of the March income supplements to Current Population Survey.
Table 1. Projections of the Resident U.S. Population, by Age
(in thousands, as of July 1 each year)
Age
2005
2015
2025
2035
2045
2050
Under 20
81,972
85,208
92,026
98,192
105,344
109,147
20 to 64
176,839
190,368
193,888
203,052
217,559
224,001
65 to 69
10,123
15,621
19,647
18,683
18,829
20,444
70 to 79
15,876
18,748
28,309
34,114
32,183
32,566
80 and up
10,696
12,422
15,568
23,844
31,947
33,696
Total
295,506
322,367
349,438
377,885
405,862
419,854
65 and older
36,695
46,791
63,524
76,641
82,959
86,706
% of Total
12.4%
14.5%
18.2%
20.3%
20.4%
20.7%
Source: U.S. Census Bureau [http://www.census.gov/ipc/www/usinterimproj/].

CRS-3
The Income of Older Americans and the Federal Budget
Congress has an interest in the sources and amounts of income among older
Americans, not only because they represent a large and growing proportion of the
U.S. population, but also because much of the income of older Americans is provided
through government-sponsored income transfers, or is subsidized through income tax
deductions and exemptions. In FY2006, for example, federal expenditures for Social
Security will exceed $550 billion, representing more than 20% of total federal
spending and nearly 38% of federal spending on entitlement programs.4 The federal
government also provides income directly to low-income older Americans through
the Supplemental Security Income (SSI) program and to retired federal employees
through the Civil Service Retirement System and the Federal Employees’ Retirement
System. Of the estimated $35.3 billion federal expenditure for SSI benefits in
FY2006, approximately one-sixth, or $5.9 billion, will be paid to recipients who are
65 or older.5 Federal outlays for retirement and disability pensions for former federal
employees will exceed $58 billion in FY2006, with most of this amount being paid
to individuals age 55 and older.
The public also subsidizes the income of older Americans through several
exemptions and deductions that Congress has included in the Internal Revenue Code.
These tax incentives promote sponsorship of pension plans by employers and
encourage workers to save for retirement. The tax deductions and deferrals granted
to qualified retirement plans are the second largest tax expenditure in the federal
budget, and will reduce federal tax revenues by an estimated $103.3 billion in
FY2006.6 Congress also allows taxpayers who are 65 or older to claim an additional
standard deduction on their income tax returns, which will reduce federal tax
revenues by an estimated $1.7 billion in FY2006.
Trends in the Income of Older Americans
Developing public policies that will promote income security among older
Americans while also meeting other important fiscal priorities will pose many
challenges for Congress. Congress has already begun to consider policy options to
meet these challenges. In its first session, the 109th Congress debated proposals to
incorporate individual accounts into Social Security. In the second session, Congress
is attempting to reconcile bills passed by the House and Senate that would
substantially reform the laws that govern the funding requirements for employer-
sponsored pensions in the private sector. Social Security, pensions, and retirement
savings will continue to occupy the attention of Congress as the baby boom
4 Budget of the United States Government, Fiscal Year 2007: Analytical Perspectives, p.
360. Not all Social Security spending is for persons 65 and older. Data collected by the
Census Bureau indicate that about 76% of Social Security recipients in 2004 were 65 or
older and these individuals received 78% of Social Security benefits paid that year.
5 Budget of the United States Government, Fiscal Year 2007, and Census Bureau data.
6 Budget of the United States Government, Fiscal Year 2007: Analytical Perspectives, p.289.
The largest tax expenditure is the exclusion for employer contributions for employee health
insurance, which will reduce federal tax revenues by an estimated $118.4 billion in FY2006.

CRS-4
generation retires, and more households begin to rely on Social Security, pensions,
and savings to provide the majority of their income.
As policymakers consider proposals that would affect the main income sources
of older Americans, it may be helpful for them to know how the amounts and sources
of income of people who have reached retirement age have changed over time. This
CRS report presents data collected by the Census Bureau from 1969 through 2005
that describe how the demographic traits, employment patterns, and the sources,
amounts, and distribution of income among people 65 and older have changed over
a period of nearly 40 years. The report begins with a demographic profile of older
Americans, which describes the population 65 and older in terms of age, race, sex,
educational attainment, and marital status and how these demographic traits have
changed since the late 1960s. This is followed by a section that focuses on trends in
employment among people 65 and older, including differences in rates of
employment among men and women. The third section of the report provides
detailed information on the sources and amounts of income received by people 65
and older, and how the proportion of total income from each source has changed over
time. The fourth section examines trends in the distribution of income among
households, and compares the distribution of income among the households of older
Americans with the income distribution of income among the households of people
under 65 years old. This is followed by a section that describes the decline in the
proportion of older Americans living in poverty since the 1960s. The final section
of the paper explains how individual and household incomes fall as people age and
gradually reduce their attachment to the paid labor force.
The Data
All of the data presented in this report were collected by the Census Bureau
through the Current Population Survey (CPS), a survey of the civilian,
noninstitutional population of the United States. The CPS was begun in the 1940s
as a way to measure unemployment each month on the basis of a random sample of
U.S. households. It is still the source of the official rate of unemployment reported
each month by the Bureau of Labor Statistics. The CPS also is used to collect annual
income data. Each year, in March, the survey includes a detailed set of questions on
the amounts and sources of income people received during the previous calendar
year. Over the years, the number of income questions has expanded, and today
information is gathered on more than 50 different sources of income, and noncash
benefits such as food stamps, employer-provided pension plans, employer-sponsored
health insurance plans, Medicaid, Medicare, and home energy assistance. The survey
also collects detailed information on the employment status, work experience,
occupation, and industry of employment of people 15 years old and over. The
number of households surveyed has increased over the years. In March 1969, the
earliest survey CRS analyzed for this report, the CPS sample included 47,000
households and contained records for 152,000 individuals. In March 2005, the most
recent survey studied for this report, the CPS sample included 100,000 households
and contained records for 216,000 individuals.

CRS-5
People 65 and Older in Nursing Homes and Long-term Care Facilities
The data presented in this report were collected by the Census Bureau through the
Current Population Survey (CPS). The CPS is conducted among the civilian,
noninstitutionalized population of the United States. It does not include residents of
prisons, nursing homes, or military personnel living on base. In 2000, an estimated
1.72 million persons resided in nursing homes. Of this number, 1.56 million (91%)
were age 65 or older, comprising 4.5% of the total number of persons age 65 or older.
Elderly residents of nursing homes differ demographically from older persons who
live in the community. Nursing home residents are older, more likely to be female,
and are less likely to be married than community residents. In 2005, among people
65 and older living in the community, 52% were 65 to 74 years old, 37% were 75 to
84 years old and 11% were 85 or older. The age distribution among nursing home
residents age 65 and older in 2005 was a mirror image of the community population:
15% were 65 to 74 years old, 37% were 75 to 84 years old, and 48% were 85 or
older. Among people 65 and older living in the community in 2005, 57% were
women. Among nursing home residents age 65 and older, 74% were women.
Among community residents age 65 and older in 2005, 55% were married. Among
nursing home residents, just 20% were married. The remainder were widowed,
divorced, or never married.7

Demographic Profile of the CPS Population
Age 65 and Older
The demographic traits of people 65 and older have changed over the past 35
years. Demographically, America’s elderly today are older, more racially and
ethnically diverse, better educated, and less likely to be widowed than the elderly
population of the late 1960s and early 1970s. The increasing number of Americans
living to age 80 and older is of particular importance to policymakers, because it is
the very old who are most likely to need medical, social, and long-term care services,
and who are at the greatest risk of depleting their financial resources and slipping into
poverty.
Age
The American population is not only growing larger, it is getting older. During
the past 35 years, the total U.S. population increased by 47%, rising from 198 million
in 1970 to 291 million in 2005. Also during that period, the number of Americans
age 65 and older has nearly doubled, growing from slightly less than 19 million to
more than 35 million. (See Table 2.) The proportion of the population who are 65
or older is projected to continue rising. In 1970, 9% of the population were age 65
7 The demographic characteristics of nursing home residents presented here were taken from
the Minimum Data Set Resident Reports prepared by the Centers for Medicare and Medicaid
Services, U.S. Department of Health and Human Services. These data are a representative
sample of 1.4 million residents of Medicare- and Medicaid-certified long-tem care facilities.

CRS-6
and older, and by 2005 the share had risen to 12%. The Census Bureau projects that
by 2035, 20% of the U.S. population will be age 65 and older.8
As the proportion of the total population age 65 and older has increased, the age
distribution among older persons also has changed. Compared to 35 years ago, a
greater share of the elderly today are age 80 or older. In 1970, 36% of Americans age
65 or older were 65 to 69 years old, 47% were 70 to 79 years old, and just 17% were
80 or older. In contrast, by 2005, the proportion of people age 65 or older who were
65 to 69 years of age had declined to 29% and the proportion who were age 80 or
older had risen to 26%. (See Table 2.)
Table 2. Age Distribution of People Age 65 and Older,
1970-2005
Percentage
Number
80 and
65-69
70-79
Total
of U.S.
(thousands)
older
populationa
1970
18,899
35.7%
47.3%
17.0%
100%
9.4%
1980
23,743
36.1
46.3
17.6
100
10.9
1990
29,566
34.3
46.1
19.6
100
12.0
2000
32,621
28.7
47.9
23.4
100
11.9
2005
35,213
28.8
45.0
26.2
100
12.1
Source: CRS analysis of the March income supplements to Current Population Survey.
a. People age 65 and older as a percentage of total U.S. population.
Sex
The proportion of Americans age 65 and older who are male or female has
remained relatively constant over the past 35 years. (See Table 3.) Roughly 43% of
older Americans are men, whereas women comprise roughly 57%. The difference
in the share of elderly by sex is primarily due to females having a longer life
expectancy than their male counterparts.9 On average, a woman age 65 in 2002 could
expect to live an additional 19.5 years (to age 84 and 6 months) and a man age 65
could expect to live an additional 16.6 years (to age 81 and 7 months).10 Although
the population age 65-69 is roughly equally divided between men and women —
with men comprising 48% and women comprising 52% in 2005 — women’s greater
longevity results in a dramatic difference for older age groups. Nearly two-thirds
(63% in 2005) of Americans age 80 and older are women.
8 U.S. Census Bureau, “U.S. Interim Projections by Age, Sex, Race, and Hispanic Origin,”
2004 at [http://www.census.gov/ipc/www/usinterimproj].
9 For an explanation of the differences in life expectancy by sex, see CRS Report RL32792,
Life Expectancy in the United States, by Laura Shrestha.
10 U.S. National Center for Health Statistics, Vital Statistics of the United States.

CRS-7
Race and Ethnicity
Reflecting changes in the general population, Americas’s older population has
become more racially and ethnically diverse over the past 35 years.11 Although they
remain a majority, the share of Americans age 65 and older who report their race as
white has declined. (See Table 3.) In 1970, 92% of those age 65 and older were
white, whereas in 2005, the share had fallen to 87%. The share of older Americans
who are black has risen slightly during the same period from 7% to 8%. The share
of older individuals who the Census Bureau categorizes as other races — primarily
persons of Asian and Native American heritage — has risen over the last 35 years
from less than 1% to more than 4% of the population. Persons of Hispanic ethnicity
may report themselves to be of any race.12 Between 1971 and 2001, the proportion
of Americans age 65 and older who identified themselves as being of Hispanic
ethnicity increased from 1.6% to 5.3%.
Table 3. Race and Sex of Individuals Age 65 and Older,
1970-2005
Male
Female
White
Black
Othera
1970
42.7% 57.3% 91.9% 7.3%
0.8%

1980
41.2
58.8
90.3
8.5
1.2
1990
41.7
58.3
89.6
8.4
2.0
2000
42.6
57.4
88.5
8.4
3.0
2005
43.0
57.0
87.2
8.4
4.4
Source: CRS analysis of the March income supplements to Current Population Survey.
a. Mainly Asian and Native American. Persons of Hispanic ethnicity may be of any race.
Marital Status
In 2005, women age 65 and older were almost equally likely to be widowed as
they were to be married. (See Table 4.) This a consequence of both women’s
longevity and the fact that many women marry older men. In 2005, roughly one-half
of married women age 65 and older had husbands who were at least two years older.
These factors help to explain how less than one-half (42%) of older women, but
nearly three-quarters (72%) of older men were married in 2005. The marital
composition of the elderly has changed over the last 35 years. The proportion of men
and women who are divorced has more than tripled, and the share of men and women
who are widowed has fallen. For both older men and women, the share who have
11 In 1970, 88% of the total U.S. population of all ages reported their race as white, 11% as
black, and 1% as other. By 2005, the share reporting their race as white had fallen to 81%
and the share reporting their race as black had risen to 13%; other rose to 7%.
12 For a discussion of the federal standards for presenting data on race and Hispanic origin,
see CRS Report RL3270, The Changing Demographic Profile of the United States, by Laura
Shrestha.

CRS-8
never been married has fallen. In 1970, roughly 7% of persons age 65 and older had
never been married; by 2005 it was roughly 4%.
Table 4. Marital Status of Men and Women Age 65 and Older,
1970-2005
Married
Separated
Divorced
Widowed
Single
Men
1970
71.0%
2.3%
2.3%
17.6%
6.8%
1980
75.5
2.0
3.7
13.6
5.1
1990
74.3
2.3
5.0
14.2
4.2
2000
72.6
2.7
6.1
14.4
4.3
2005
71.7
3.2
7.1
13.7
4.4
Women
1970
35.3%
1.3%
2.1%
54.1%
7.2%
1980
38.1
1.7
3.4
51.0
5.9
1990
39.7
1.7
5.1
48.6
4.9
2000
41.3
2.6
7.2
45.3
3.6
2005
42.0
2.7
8.5
42.9
4.0
Source: CRS analysis of the March income supplements to Current Population Survey.
Education
America’s older population is an increasingly well-educated group. In 1970, the
vast majority (71%) of Americans age 65 and older had not graduated from high
school and only 13% had either some college or were college graduates. (See Table
5
.) By 2005, only a quarter of older Americans had not graduated from high school
and 38% had either some college or were college graduates. The rise in the education
level of the elderly reflects the increasing educational attainment in the total
population as well as the positive correlation between education and life
expectancy.13,14
13 Nicole Stoops, Educational Attainment in the United States, 2003, Population
Characteristics
, U.S. Census Bureau, Current Population Reports, P20-550.
14 CRS Report RL32792, Life Expectancy in the United States, by Laura Shrestha.

CRS-9
Table 5. Educational Attainment of Individuals Age 65
and Older, 1970-2005
Less than
High School
Some
College
High School
Graduate
College
Graduate
1970
71.4%
16.0%
6.2%
6.3%
1980
59.4
24.0
8.1
8.5
1990
42.9
33.1
12.1
11.9
2000
29.0
37.3
18.0
15.6
2005
26.1
36.4
18.7
18.8
Source: CRS analysis of the March income supplements to Current Population Survey.
Employment Among People Age 65 and Older
Patterns of employment among older Americans have changed since the 1960s,
and the trends in employment among older workers have differed between men and
women. From the late 1960s to the early 1990s, employment rates fell steadily
among men 55 and older. Since then, employment rates have risen for older men, but
they remain below the employment rates that prevailed among men 55 and older as
recently as the late 1970s. Among older women, the trend in employment since the
late 1960s has been one of steady increase, as the larger numbers of women who
entered the labor force in the 1960s and 1970s have begun to pass age 55. Despite
this steady increase, however, older women’s employment rates today remain lower
than those of men the same age. Moreover, among both men and women
employment rates continue to drop significantly after age 70. Consequently, earnings
are likely to remain a fairly small contributor to the total income of people older than
age 70.
Although the reasons for the increased employment among older individuals —
particularly the upturn in employment among older men — cannot be determined
with complete certainty, two factors often cited by economists are long-term
developments in the design of employer-sponsored pensions and the decline in the
proportion of retirees with access to employer-sponsored group health insurance.
With respect to pensions, the trend for the past 25 years has been away from
traditional defined-benefit pensions that pay a guaranteed annuity for life — and
which often pay early-retirement subsidies to encourage workers to retire at 55 —
toward defined contribution plans, such as the 401(k). The typical 401(k) plan does
not include any type of subsidy for early retirement, and some workers may have
begun to delay retirement in order to retire with a larger account balance. In the area
of health insurance, rapidly rising health care costs have discouraged employers from
beginning or continuing to offer health insurance to retirees.15 In the absence of this
coverage, more workers may choose to continue working until age 65, when they
become eligible to enroll in Medicare.
15 For more information, see CRS Report RL33361, Employer-Sponsored Retiree Health
Insurance: An Endangered Benefit?
, by Neela Ranade.

CRS-10
Rates of Employment, by Age
As people grow older, both the proportion of those who work and of those who
work full-time decline. In 2004, 90% of men between the ages of 25 and 54 worked
for pay at some time during the year.16 Among men age 55 to 64, 73% worked in
2004, and of men 65 and older only 23% were employed at some time during the
year. Similarly, 77% of women between the ages of 25 and 54 were employed at
some time in 2004; the employment rate among women 55 to 64 years old was 60%.
Among women 65 or older, just 14% were employed at some time in 2004.
Older workers also tend to work fewer hours than younger workers. Among all
workers between the ages of 25 and 54, 87% worked full-time in calendar year 2004
and 13% worked part-time. Even among workers aged 55 to 59, 86% worked full-
time in 2004. However, among those 60 to 64 years old, only 78% who worked in
2004 were employed full-time. Beginning at age 62, individuals are eligible for
Social Security retired worker benefits, and this appears to have some effect on rates
of full-time employment. Among people who were 60 or 61 years old and who
worked in 2004, 82% were employed full-time. Among those who were 62 to 64
years old and who worked in 2004, just 74% worked full time. Rates of full time
employment fall further after age 65. Of those 65 and older who were employed in
2004, only 51% worked full-time.17
The trend in employment rates among all persons 55 and older since the late
1960s is described by a slightly “U-shaped” curve. The employment rate among
people 55 and older at first declined, reaching a low point in the mid- to late 1980s.
During the past 10 to 15 years, employment rates among older persons have risen,
and the rate of increase appears to have accelerated recently. In 1969, 70% of people
55 to 59 years old were employed at some time during the year, as were 60% of
people 60 to 64 years of age, 35% of those aged 65 to 69, and 13% of those age 70
and older. (See Table 6.) By the late 1980s, employment rates for all four of these
age groups had fallen, and for those age 60 or older employment rates had fallen
substantially. In 1989, the employment rate among people 55 to 59 years old was
67%, just 3 percentage points lower than in 1969. Among those 60 to 64 years old,
however, the employment rate in 1989 was 49%, or 11 percentage points lower than
20 years earlier. Likewise, among people ages 65 to 69, the employment rate of 25%
in 1989 was 10 percentage points lower than it had been in 1969. Among people age
70 and older, the 1989 employment rate was more than four percentage points lower
than it had been in 1969.
16 The statistics in this paragraph were derived from the March 2005 Current Population
Survey. For more on labor force participation among people 55 and older see, CRS Report
RL30629, Older Workers: Employment and Retirement Trends, by Patrick Purcell.
17 Although Social Security retired worked benefits are first available at age 62, beneficiaries
who are under the full retirement age (65 years and 8 months for those who reach age 65 in
2006) are subject to an earnings test that reduces the monthly benefit if their earnings exceed
a threshold amount. In 2006, Social Security beneficiaries who are under the full retirement
age have their benefit reduced by $1 for every $2 of annual earnings over $12,480.

CRS-11
Since the late 1980s, employment rates among older persons have risen,
although among people 60 and older they remain lower than they were in the late
1960s. Among people 55 to 59 years old, 74% were employed at some time in 2004,
an increase of 7 percentage points since 1989. The employment rate among
individuals age 60 to 64 was 57% in 2004, an increase of 8 percentage points since
1989. The employment rate among those 65 to 69 years old also rose by 8 percentage
points between 1989 and 2004, rising from 25% to 33%, whereas the employment
rate among those age 70 and older rose from 9% to 12% during this time.
Table 6. Employment Rates, by Age, 1969 to 2004
(persons employed full-time or part-time in any month of the year)

Age
Age
Age
Age 70
55-59
60-64
65-69
and Older
1969
70.4%
59.5%
34.9%
13.3%
1974
66.4
55.2
27.7
11.7
1979
68.5
54.2
28.4
11.2
1984
68.3
51.1
24.8
10.0
1989
67.4
48.5
24.8
8.5
1994
69.1
48.6
24.4
8.2
1999
70.3
52.7
27.0
10.0
2004
73.9
57.0
32.7
12.2
Source: CRS analysis of the March income supplements to Current Population Survey.
Employment Rates among Men and Women
Employment rates differ between men and women at every age, and the recent
trends in employment rates among older Americans also differ between men and
women. Although women have lower labor force participation rates than men, most
of the increase in employment among older Americans since the mid-1980s has been
the result of the increased labor force participation of women. Employment rates
among older men have begun to rise over the past 15 years, but they remain
substantially lower than the employment rates that prevailed among men up until
about 1980. Employment rates among older women, in contrast, were higher in 2004
than at any point in the previous 35 years.
In 1969, 91% of men between the ages of 55 and 59 were employed at some
time during the year. (See Table 7.) The employment rate among men in this age
group declined steadily over the next 30 years, falling to 78% by 1999. By 2004, the
employment rate among 55 to 59 year-old men had risen to 80%, but this was still
lower than the employment rate among men in this age group had been 20 years
earlier in 1984. Similar patterns can be seen in the employment rates of men in the
60-64, 65-69, and 70 and older age categories. In each of these groups, employment
rates among men fell from the late 1960s through the mid-1990s, and then rose a few
percentage points. Employment rates among men in these age groups, while higher
in 2004 than in 1999 or 1994 were still lower than they were 25 years earlier in 1979.

CRS-12
Table 7. Employment Rates Among Men and Women, by Age
(persons employed full-time or part-time in any month of the year)

Age
Age
Age
Age 70
55-59
60-64
65-69
and older
Men
Women
Men
Women
Men
Women
Men
Women
1969
90.7% 52.0% 80.3% 41.4% 50.2% 22.3% 21.3%
7.7%

1974
83.8
50.5
73.8
39.0
39.2
18.5
19.7
6.5
1979
84.6
53.9
70.5
39.9
38.4
20.3
17.9
6.8
1984
81.8
55.8
64.2
40.0
32.9
18.3
15.1
6.7
1989
79.6
56.3
59.9
38.5
30.6
20.0
12.9
5.7
1994
78.4
60.7
56.0
41.8
30.1
19.6
12.4
5.4
1999
77.7
63.7
61.6
44.6
33.2
21.4
14.4
6.9
2004
80.1
68.2
64.5
50.3
37.1
28.7
16.7
9.0
Source: CRS analysis of the March income supplements to Current Population Survey.
Employment rates among older women did not follow the U-shaped pattern of
men’s employment rates over the period from 1969 to 2004. From 1969 through the
mid-1980s, employment rates among women age 55 to 59 generally rose, whereas
employment rates among women 60 and older fell slightly. Since the late 1980s,
employment rates among women 55 to 69 years old have continued to rise as the
large numbers of women who entered the labor force in the 1960s and 1970s began
to pass age 55. In 2004, for example, 68% of women aged 55 to 59 were employed,
up from 56% in 1984. Employment rates also have risen among women 60 and
older. Fifty percent of women 60 to 64 years old were working in 2004, compared
with 40% in 1984, and 29% of women aged 65 to 69 were employed at some time
in 2004, up from 18% 20 years earlier.
Figures 2 and 3 show the percentage of men and women, respectively, who
were employed in March each year from 1969 to 2005.

CRS-13
Figure 2. Employment Rates of Men by Age, 1969-2005
90.0%
80.0%
70.0%
oyed 60.0%
Empl
55-59
50.0%
60-64
65-69
f Men 40.0%
70+
30.0%
Percent o 20.0%
10.0%
0.0%
1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005
Year
Source: CRS analysis of the March income supplements to Current Population Survey.

CRS-14
Figure 3. Employment Rates of Women by Age, 1969-2005
90.0%
80.0%
ed 70.0%
oy
pl 60.0%

m
n E
55-59
50.0%
60-64
65-69
40.0%
f Wome
70+
t o 30.0%
n
e
rc 20.0%

e
P

10.0%
0.0%
1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005
Year
Source: CRS analysis of the March income supplements to Current Population Survey.

CRS-15
Full-Time vs. Part-Time Employment among Older Workers
As noted above, individuals who continue to work past age 55 remain almost
as likely to work full-time as their younger counterparts until about age 62, the first
age of eligibility for Social Security retired worker benefits. For workers 55 to 59
years old, the distribution between full-time and part-time employment changed
relatively little between 1969 and 2004, whereas for those age 60 to 64, the
proportion of workers employed full-time fell from 86% in 1969 to 78% in 1989 and
has remained near that level since then. Among workers 65 to 69 years old and those
70 and older, the percentage who work full-time fell between 1969 and 1979, but has
since increased. (See Table 8.)
Table 8. Full-Time and Part-Time Employment,
by Age of Worker
Age
Age
Age
Age 70
55-59
60-64
65-69
and older
Full-
Part-
Full-
Part-
Full-
Part-
Full-
Part-
time
time
time
time
time
time
time
time
1969
89.1% 10.9% 85.7% 14.3% 65.2% 34.8% 45.2% 54.8%
1974
88.6
11.4
85.9
14.1
57.4
42.6
43.2
56.8
1979
86.2
13.8
81.9
18.1
51.1
48.9
37.8
62.2
1984
84.1
15.9
79.0
21.0
51.5
48.5
34.2
65.8
1989
85.1
14.9
77.7
22.3
53.5
46.5
38.9
61.1
1994
84.5
15.5
76.2
23.8
53.6
46.4
40.2
59.8
1999
86.9
13.1
78.8
21.2
54.9
45.1
46.0
54.0
2004
86.2
13.8
78.2
21.8
59.0
41.0
42.2
57.8
Source: CRS analysis of the March income supplements to Current Population Survey.
Sources and Amounts of Income Among
People 65 and Older
Older Americans receive income from a variety of sources such as earnings,
Social Security, assets, pensions, and welfare.18 Four sources of income — earnings,
Social Security, pensions, and income from assets — comprise the majority of
income among people 65 and older today, just as they did 35 years ago. In 2004,
these four income sources provided 97% of all income received by people 65 and
older. This was a slight increase from 1975 (the first year for which separate data on
pension income are available from the CPS) when these four sources provided 95%
of income received by people 65 and older.
18 The CPS does not collect information on income received from capital gains.

CRS-16
Wages and salaries today constitute a greater proportion of the aggregate income
of the elderly than at any time since the late 1970s, and while this is due in part to
increasing rates of employment, it is also partly the result of a decline over the past
15 years in the proportion of income received as interest and dividends. Since the
early 1990s, interest rates and dividend yields have fallen, and both the share of
aggregate income received from these sources and the percentage of older individuals
with income from assets have fallen since that time. Social Security has for most of
the past 35 years remained a fairly stable share (about 40%) of the income of people
65 and older. Trends in pension income have been more complex.
Trends in the percentage of people receiving pension income have differed
between men and women. The proportion of men 65 and older with pension income
rose between 1980 and the early 1990s. By that time, almost half of men 65 and
older received some pension income. Since about 1992, however, the proportion of
men receiving pension income has fallen to about 44%. Although the percentage of
women 65 and older with pension income has continued to rise — although very
slowly — women 65 and older are only half as likely as men to receive income from
this source. Moreover, women’s pension income is, on average, only half that of
male pension recipients. The share of income the elderly receive from pensions has
fallen since the early 1990s because the proportion of men with pension income has
fallen and the rise in the proportion of women with pension income has been offset
by the smaller amounts they typically receive. Because the proportion of the total
workforce who participate in employer-sponsored pensions has remained stable at
about half the workforce since the 1970s, it is unlikely that the share of total income
of the elderly received from pensions will rise substantially in the next 10 to 20 years.
Few individuals age 65 and older do not receive any income. Between 1969 and
2004, the percentage of people 65 and older receiving income from at least one
source increased from 90% to 97%.19 Among people who had income from any
source, the mean annual amount (in constant 2004 dollars) rose from $13,414 in 1969
to $23,762 in 2004, an increase of 77% or 1.6% per year. Median total income, also
in constant 2004 dollars, rose from $8,196 in 1969 to $15,189 in 2004, an increase
of 85% or 1.7% per year. (See Tables 9 through 12.)20
Earnings
In 1969, a quarter of all people 65 and older worked, and earnings provided a
third of the total income of the elderly. Both of these percentages declined sharply
over the next decade, and by 1980, the proportion of people 65 and older with earned
income had fallen to 17%, and earnings had fallen to 16% of the aggregate income
of the elderly. The proportion of Americans age 65 and older with earnings has
19 The majority of individuals without income were women. In 1969, 16% of women and
1% of men received no income. By 2004, the share of men who received no income had
risen slightly (to 2%) and the share of women who received no income had fallen to 4%.
20 The median lies at the middle of the income distribution. Half of the population have
higher incomes, and half have lower incomes. Mean income is generally higher than median
income because a relatively small percentage of people have very high incomes. The
median is therefore widely considered to be a more accurate measure of average income.

CRS-17
increased slightly in recent years, rising from 16% in the early 1990s to 18% in 2004.
As a share of aggregate income, however, earnings have risen more substantially,
growing from 15% of the total income of the elderly in the mid-1990s to 23% in
2004. (See Figure 4.) The increase in the proportion of income from earnings is the
result mainly of four factors: (1) growth in real wages; (2) increased employment
among people 65 and older; (3) an increase in the percentage of working elderly who
are employed full-time; and (4) a drop in interest and dividend income that resulted
from falling interest rates, falling dividend yields, and a decline in the percentage of
people receiving interest and dividend income.21 Among people who had earned
income, the mean total amount (in constant 2004 dollars) rose from $14,497 in 1969
to $29,459 in 2004, an increase of 103% or 2.0% per year. Median earned income,
also in constant 2004 dollars, rose from $6,856 in 1969 to $15,000 in 2004, an
increase of 119% or 2.2% per year.
Social Security
In 1969, 75% of people 65 and older received Social Security income, either as
retired workers, as spouses, or as survivors of deceased workers. By 1991, the
proportion of elderly receiving Social Security benefits had risen to 91%, and it has
remained within two percentage points of 90% since that time. Social Security
provided almost one-third of all income received by people 65 and older in 1969. By
1975, Social Security benefits accounted for 42% of the total income of people 65
and older. During the past 30 years, Social Security has consistently comprised
between 38% and 43% of the aggregate income of the elderly population. Among
people 65 and older who received Social Security income, the mean annual amount
rose from $5,171 in 1969 to $10,681 in 2004, an increase of 107% or 2.0% per year.
Median Social Security income rose from $4,866 in 1969 to $10,399 in 2004, an
increase of 114% or 2.1% per year. (All amounts are in constant 2004 dollars.)
Asset Income
Income from assets — primarily interest and dividends, but also including rents
and royalties — was received by 39% of people 65 and older in 1969. The
proportion of older persons receiving asset income rose steadily over the next decade,
and by 1980 two-thirds of the elderly received income from assets. This proportion
remained relatively stable until the late 1990s when it began to decline. By 2004, the
proportion of people 65 and older who received asset income had fallen to 56%. The
decline in the proportion of people receiving income from assets coincided with a
long period of falling interest rates and dividend yields. As the rate of return on these
assets fell, fewer people chose to hold such assets.
In 1969, interest, dividends, rent, and royalties provided 20% of the aggregate
income of people 65 and older. This proportion steadily increased over the next two
decades and peaked at 26% of total income in 1986. The share of the income of the
elderly composed of income from assets has progressively fallen since 1986, and in
21 Between 1984 and 1994, the yield on 10-year Treasury notes fell from 12.4% to 7.1% and
by 2004 had fallen to 4.3%. Dividend yields declined from 4.6% in 1984 to 2.8% in 1994
and 1.7% in 2004. (Source: Economic Report of the President, 2006.)

CRS-18
2004 only 12% of the income of people 65 and older came from this source. Among
people 65 and older who received income from assets, the mean and median annual
amounts received fluctuated from year to year. Mean annual income ranged from a
low of about $5,000 per year to a high of about $7,500 per year, in 2004 dollars.
Median income from assets was substantially lower than mean asset income because
a relatively small proportion of individuals had very large amounts of asset income.
Expressed in 2004 dollars, median income from assets was highest in 1984, reaching
$2,637 that year. Since the mid-1980s, median asset income has steadily fallen.
Among those who received income from interest, dividends, rent, and royalties in
2004, the median annual amount was just $817.
Table 9. Percentage of People Age 65 and Older
with Income from Each Source
Source of Income
1969
1970
1980
1990
2000
2004
Any income
90.1%
90.4%
98.7%
98.8%
97.8%
97.1%
Earnings
24.9
23.4
16.6
16.3
16.9
18.0
Social Security
74.7
75.9
90.5
91.0
89.8
88.2
Asset income
39.4
39.5
67.1
69.9
59.2
55.5
Pension income
n/a
n/a
26.6
37.0
34.9
35.5
Welfare
9.2
9.2
8.5
5.5
3.8
3.5
Other income
24.6
24.2
6.9
7.7
6.0
5.5
Source: CRS analysis of the March income supplements to Current Population Survey.
n/a = not available
Table 10. Income from Each Source as a Percentage of
Total Income Among People 65 and Older
Source of Income
1969
1970
1980
1990
2000
2004
Earnings
29.9%
27.1%
15.9%
15.3%
20.0%
23.0%
Social Security
32.0
35.3
42.8
37.9
40.3
40.8
Asset income
20.4
19.5
22.4
24.0
17.2
12.3
Pension income
n/a
n/a
15.3
19.8
19.7
21.2
Welfare
2.7
2.6
1.6
0.8
0.7
0.6
Other income
15.1
15.5
2.0
2.2
2.1
2.0
Source: CRS analysis of the March income supplements to Current Population Survey.
n/a = not available

CRS-19
Table 11. Mean Annual Income of People
65 and Older from Each Source
(in 2004 dollars)
Source of Income
1969
1970
1980
1990
2000
2004
Income from all sources
$13,414 $13,429 $15,662 $20,072 $23,257 $23,762
Earnings
14,497 14,039
14,789
18,602
27,001
29,459
Social Security
5,171
5,656
7,305
8,253
10,213
10,681
Asset income
6,256
5,986
5,161
6,803
6,622
5,122
Pension income
n/a
n/a
8,924
10,617
12,810
13,953
Welfare
3,556 3,408 2,818
3,035
4,149
4,245
Other income
7,405
7,788
4,445
5,743
7,911
8,161
Source: CRS analysis of the March income supplements to Current Population Survey.
n/a = not available
Table 12. Median Annual Income of Individuals
65 and Older from Each Source
(in 2004 dollars )
Source of Income
1969
1970
1980
1990
2000
2004
Income from all sources
$8,196 $8,271 $10,723 $13,523 $14,727 $15,189
Earnings
6,856
6,644
8,281
8,705 12,937 15,000
Social Security
4,866
5,267
7,212
8,036
9,903 10,399
Asset income
2,134
2,038
1,656
2,385
1,477
817
Pension income
n/a
n/a
5,995
6,991
8,625
9,600
Welfare
3,125
3,180
2,374 2,314 3,247 3,600
Other income
5,123
5,189
2,459
2,893
3,880
4,799

Source: CRS analysis of the March income supplements to Current Population Survey.
n/a = not available










CRS-20
Figure 4. Share of Income by Source among People 65+, 1969-2004
Source: CRS analysis of the March income supplements to Current Population Survey.
Pension Income
The CPS has collected data on pension income as a separate category only since
the March 1976 survey, which collected information on income received in 1975.
Before that, income from pensions was included in a broad category of “other
income.” In 1975, 22% of people 65 and older received income from pensions
earned through a job or jobs in either the public or private sectors or as the survivor
of a worker who was receiving a pension. The percentage of people 65 and older
receiving income from pensions rose until the early 1990s, peaking at 38% in 1992.
The proportion of the elderly population receiving pension income has since fallen
slightly. In 2004, 36% of people 65 and older received pension income.22
In 1975, income from pensions made up 14% of the total income of people 65
and older. This proportion increased over the next 15 years, along with the
percentage of older persons receiving income from pensions, reaching a peak of 22%
in 1993. Since then, pension income has comprised roughly 20% of the total income
of people 65 and older each year. Among people 65 and older who received pension
income, the mean annual amount (in 2004 dollars) rose from $8,848 in 1975 to
22 This category includes pensions from previous employers or unions, and other retirement
income, excluding Social Security and Veterans’ Administration payments. It includes
company or union pensions and profit sharing plans; annuities; military retirement; federal
government pensions; state or local government pensions; Railroad Retirement; payments
from annuities or paid-up insurance policies; and payments from IRA or Keogh accounts.

CRS-21
$13,953 in 2004, an increase of 58% or 1.5% per year. Median pension income rose
from $6,081 in 1975 to $9,600 in 2004, also increasing by 58% or 1.5% per year.
Welfare Income
In the late 1960s and early 1970s, about 10% of the elderly population in the
United States received some form of cash welfare assistance from the federal or state
governments. The largest source of cash welfare assistance for people 65 and older
is the Supplemental Security Income (SSI) program, which was authorized by
Congress in 1972 and began making payments to eligible low-income aged and
disabled persons in 1974.23, 24 The proportion of the elderly population receiving
cash welfare payments has steadily fallen since the mid-1970s, mainly because the
proportion of people 65 and older who receive Social Security benefits has increased
and the real value of Social Security benefits has risen due to the growth in real
wages. The maximum eligible income limit for SSI benefits, in contrast, grows at
the annual rate of price inflation.25 As a share of the aggregate income of people 65
and older, cash welfare payments fell from 3% in 1969 to 1% in 1986. Welfare
income has been less than 1% of the total income of older Americans for each of the
past 20 years. During the period from 1969 to 2004, the mean and median amounts
of welfare income received by older individuals fluctuated within relatively narrow
ranges, measured in constant 2004 dollars. Mean welfare income ranged from a low
of $2,818 in 1980 to a high of $4,426 in 2002. Median welfare income reached a low
of $2,261 in 1992 and a high of $3,629 in 2003.
Other Income
Other sources of income recorded on the CPS include workers’ compensation,
unemployment compensation, alimony, child support, and financial assistance from
friends or relatives not living in the same household. During the past 30 years, the
proportion of people 65 and older receiving any income from these sources has
consistently remained in a narrow range of 6% to 8% of the elderly population. As
a share of the aggregate income of the elderly, all of these sources combined have in
each of the past 30 years accounted for just 2% to 3% of total income. Among
people 65 and older who received income from one or more of these sources in 2004,
the mean amount received was $8,161 and the median amount was $4,799. This
compares to mean and median amounts received from these sources in 1975 of
$4,191 and $3,118, respectively. (All amounts are in 2004 dollars.)
23 For more information on SSI, see CRS Report 94-486, Supplemental Security Income
(SSI): A Fact Sheet
, by Scott Szymendera.
24 This category includes only cash welfare payments. It does not include the value of food
stamps, medical assistance, rental subsidies, heating assistance, or other in-kind transfers.
25 In any given year the rate of price inflation may exceed the rate of growth of average
wages, but over long periods of time wages grow faster than prices because average wages
grow at the average rate of inflation plus the rate of growth of labor productivity. Prices can
grow faster than wages in the long run only if labor productivity fails to increase.

CRS-22
Sources of Income, by Quartile
The sources of income received by people 65 and older whose incomes are at
the lower end of the income distribution differ from the sources of income of people
at the upper end of the income distribution. This was true of income throughout the
period from 1969 to 2004. In 1975, individuals 65 and older whose incomes (in 2004
dollars) exceeded $17,784 were in the top 25% (the top quartile) among people in
this age group. On average, they received 30% of their income from earnings, 22%
from Social Security, 19% from pensions, 27% from assets, less than 1% from
welfare, and 2% from other sources. That same year, individuals whose incomes
were less than $6,424 (in 2004 dollars) were in the bottom quartile among
individuals age 65 and older. Older individuals in the bottom income quartile
received less than 1% of their income from earnings, 80% from Social Security, 1%
from pensions, 3% from assets, 13% from welfare, and 2% from other sources. (See
Figure 5 and Figure 6.)
By 2004, the sources of income in the top and bottom quartiles of the income
distribution had changed relatively little. In the top quartile — people 65 and older
with 2004 income of $26,600 or more — earnings and pension income each had
increased as shares of total income, while asset income had declined somewhat as a
percentage of total income. In both 1975 and 2004, Social Security accounted for
about $1 out of every $5 of income among people 65 and older with incomes in the
top 25% of the income distribution. In the bottom quartile — people with 2004
income of less than $9,260 — Social Security accounted for about $4 out of every $5
of income. Earnings accounted for only 1% of income in both 1975 and 2004.
Although pensions and asset income increased slightly in relative significance among
people 65 and older in the bottom income quartile, in 2004 both of these sources of
income combined accounted for just 7% of their income. Cash welfare declined from
13% of the income received by persons in the bottom income quartile in 1975 to 5%
in 2004. This was due both to the increase in the proportion of people 65 and older
who received Social Security benefits and to growth in the real value of Social
Security benefits. In the bottom income quartile, Social Security increased from 80%
of total income in 1975 to 86% of total income in 2004.





















CRS-23
Figure 5. Sources of Income by Quartile, 1975
Source: CRS analysis of the Current Population Survey.


















CRS-24
Figure 6. Sources of Income by Quartile, 2004
Source: CRS analysis of the Current Population Survey.

CRS-25
Sources of Income Differ Between Men and Women
Men and women are not equally likely to receive income from each of the four
main income sources among individuals age 65 and older, and the importance of each
source of income has changed for both men and women over the past 35 years. (See
Figure 7.) Throughout this period, men 65 and older were more likely than women
to have income from wages and salaries, income from assets, and income from
pensions. In 1970, 25% of men 65 and older had wage and salary income, compared
with 12% of women. By 2004, just 18% of older men had wage and salary income,
whereas the proportion of women with wage and salary income had increased to
13%. Up until about 1980, men 65 and older also were more likely than women to
receive income from Social Security. Since then, however, the percentage of both
men and women 65 and older receiving Social Security has ranged between 88% and
92%. The percentage of both men and women 65 and older receiving income from
assets increased sharply through the 1970s and 1980s. By 1990, 72% of men and
68% of women had asset income. By 2004, these figures had fallen to 59% of men
and 53% of women. Although the percentage of both men and women receiving
pension income (including survivor benefits) increased substantially between 1975
and 1990, the upward trend ceased in the early 1990s. Moreover, throughout the 20
years from 1975 to 1995, men 65 and older were more than twice as likely as older
women to have received retired-worker or survivor benefits from a pension. Since
1995, the gap between the proportion of men and women receiving pension income
has narrowed slightly, but this has been due more to a decline in the percentage of
men receiving pension income rather than to an increase in the proportion of women
with income from pensions.
Older men and women also differ with respect to the amount of income that they
receive from each source, with men typically receiving larger amounts. Among men
age 65 and older who had wage and salary income in 1970, the median amount, in
2004 dollars, was $8,560. (See Table 13.) Among women age 65 and older who had
wage and salary income in 1970, the median amount was $6,163. By 2004, the
median wage and salary income of men 65 and older had increased to $20,800, a
143% increase. That same year, the median wage and salary income of women 65
and older was $12,000, or 95% higher than in 1970. Men also had higher median
income from Social Security through the period from 1970 to 2004, but the rate of
growth in the median benefit was slightly higher for women, as more women became
eligible for retired worker benefits that were larger than the spousal benefit they
would have received had they not worked or had worked less. Men had higher
pension income than women, and the rate of growth of pension income was faster
among men than women. Among men 65 and older who had pension income, the
median amount rose from $7,096 in 1980 to $12,000 in 2004, an increase of 69%.26
Median pension income among women 65 and older rose from $4,347 in 1980 to
$6,141 in 2004, an increase of 41%. Men also had higher median income from
assets, although for both men and women the median income from this source was
relatively small compared to income from earnings, pensions, and Social Security.
26 The CPS began to collect data on pension income as a separate category in 1976.








































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































CRS-26
Figure 7. Percentage of Men and Women Age 65+ with Income from Each Source, by Year
100.0%
e
90.0%
m
o
c

80.0%
n
70.0%
ith I
1970
e w
60.0%
1980
opl
50.0%
e
1990
2000
40.0%
of P
2004
ge
30.0%
ta
n
e

20.0%
rc
e

10.0%
P
0.0%
Men
Women
Men
Women
Men
Women
Men
Women
Wages & Salaries
Social Security
Asset Income
Pensions
Source: CRS analysis of the March income supplements to Current Population Survey.

CRS-27
Table 13. Median Income of Men and Women
Age 65 and Older from Each Source
(amounts in 2004 dollars)

Wages & Salaries
Social Security
Income from Assets
Pension Income
Men
Women
Men
Women
Men
Women
Men
Women
1970
$8,560
$6,163
$6,848
$4,353
$2,242
$1,895
n/a
n/a
1980
10,143 6,444 8,706 6,173 1,967 1,553 7,096 4,347
1990
11,783 7,713 9,889 6,726 2,410 2,345 8,886 4,820
2000
17,572 8,786
11,601 8,180 1,690 1,289
10,877 5,892
2004
20,800 12,000 12,583 8,799
964
750 12,000 6,141
Source: CRS analysis of the March income supplements to Current Population Survey.
n/a = not available
Household Income of People 65 and Older
The previous section of this report described the sources and amounts of income
received by people 65 and older at the level of the individual recipient. Although a
substantial — and growing — proportion of Americans of all ages live in single-
person households, a majority live with a spouse, other relatives, a roommate, or
others with whom they share household expenses.27 Consequently, gaining a
complete picture of the income of older Americans, how it has changed over time,
and how it compares with the income of the non-elderly population, requires an
examination of household income as well as individual income. This section of the
report describes income among all U.S. households and also among elderly
households — defined here as those in which either the householder28 or
householder’s spouse is 65 or older — and nonelderly households, defined here as
those in which neither the householder nor spouse is 65 or older.
Median household income rose faster among households in which the
householder or spouse was 65 or older than among households headed by younger
persons during the period from 1969 through 2004. Nevertheless, in 2004, the
median income of households in which the householder or spouse was 65 or older
($25,210) was just half of the median income of households in which neither the
householder nor spouse had reached age 65 ($50,466). Elderly households, however,
tend to be smaller (averaging 1.7 residents) than nonelderly households (with an
average of 2.8 residents) and they do not have all of the work-related and child-
rearing expenses that many nonelderly households must pay.
27 In 1969, 16.6% of all U.S. households were one-person households, as were 38.9% of
households in which the householder was 65 or older. In 2005, 26.4% of all households
were one-person households, as were 45.8% in which the householder was 65 or older.
28 The Census Bureau defines the householder as the person in whose name the housing unit
is owned or rented. In a married couple, the householder may be either the husband or wife.

CRS-28
When comparing the income of elderly households with the household income
of the nonelderly population, it is important to keep in mind that the households in
which people 65 and older live tend to be smaller than the households of younger
adults. During the past 35 years, average household size has shrunk among both
elderly and nonelderly households, and while the decline has been greater among
nonelderly households, they remain larger, on average, than elderly households. In
1970, there were 3.5 persons in the average nonelderly household: 2.1 adults and 1.4
children. (See Table 14.) The average elderly household had 1.9 people in 1970.
By 2005, the average number of people in nonelderly households had fallen to 2.8
people: 1.9 adults and 0.9 children. The average size of an elderly household had
fallen to 1.7 people. Because the average elderly household has 1.1 fewer people
than the average nonelderly household, it requires less income to maintain a
comparable standard of living.
Table 14. Average Household Size by
Year and Age of Householder
Householder under 65
Householder Age 65+
Year
Mean
Mean
Mean
Mean
household
number of
household
number of
size
adults
size
adults
1970
3.5
2.1
1.9
1.8
1980
3.0
2.0
1.7
1.7
1990
2.9
2.0
1.7
1.7
2000
2.8
1.9
1.7
1.7
2005
2.8
1.9
1.7
1.7
Source: CRS analysis of the March income supplements to Current Population Survey.
Median Household Income, by Age
Adjusted for inflation, median household income in the United States rose from
$35,539 in 1969 to $44,214 in 2004, an increase of 24% or 0.6% per year in real
terms. (See Tables 15 and 16.) Real median household income reached a peak of
$45,302 in 2000 and fell by $1,100 or 2.4% between 2000 and 2004. Among
households in which neither the householder nor the householder’s spouse was 65
or older, median income rose from $40,278 in 1969 to $50,466 in 2004, an increase
of 25% or 0.6% annually, on average. Among households in which either the
householder or the householder’s spouse was 65 or older, real median income rose
from $14,498 in 1969 to $25,210 in 2004, an increase of 74%, or 1.6% annually, on
average.

CRS-29
Table 15. Mean and Median Household Income,
by Age of Householder
(in 2004 dollars)
All
Householder and
Householder or Spouse
Income Year
Households
Spouse Under Age 65
65 or Older
Mean
Median
Mean
Median
Mean
Median
1969
$40,378 $35,539 $44,886 $40,278 $22,393
$14,498
1970
40,423
35,461
44,986
40,295
22,303
14,641
1980
42,847
36,639
47,239
41,853
26,424
18,425
1990
49,053
40,036
53,619
45,476
33,173
22,910
2000
61,460
45,302
67,742
52,680
38,466
25,570
2004
60,530
44,214
66,433
50,466
38,963
25,210
Source: CRS analysis of the March income supplements to Current Population Survey.
For most of the period from 1969 to 2004, the real household income of elderly
households — those in which either the householder or the householder’s spouse was
65 or older — grew faster than the real income of younger households. For example,
over the period from 1969 to 2004, the real median income of elderly households
grew by 74%, compared to 25% for nonelderly households. Since 1990, however,
the rates of income growth in elderly and nonelderly households have been much
closer together, each increasing by 10% to 11% over that period. Since 2000,
however, although the real income of nonelderly households has fallen by 4.2%, the
income of elderly households has fallen by just 1.4%.
In constant dollars, the total increases in median household income among
elderly and nonelderly households since 1969 have not differed greatly. Between
1969 and 2004, the real income of households in which neither the householder nor
spouse was 65 or older grew by $10,188, in 2004 dollars. During the same period,
the real median income of elderly households grew by $10,712. One reason the
percentage change in household income has been greater in elderly households is
simply that the income of these households was growing from a much smaller base.
In 1969, the median income of elderly households, expressed in 2004 dollars was just
$14,498. An increase of $1,000 from this amount would represent a percentage
change of 7%, whereas a $1,000 increase in the 1969 median income of nonelderly
households ($40,278, in 2004 dollars) would represent an increase of just 2.5%. In
percentage terms, the increase in median income among elderly households since the
late 1960s looks relatively large because their median household income was — and
is — much lower than the median household income of the nonelderly population.

CRS-30
Table 16. Percentage Change in Median Household Income,
1969 to 2004, by Age of Householder
(in 2004 dollars)
All
Householder and
Householder or
Income
Households
Spouse Under Age 65
Spouse 65 or Older
Years
Total
Annual
Total
Annual
Total
Annual
Change
Average
Change
Average
Change
Average
1969-2004
24.4%
0.6%
25.3%
0.6%
73.9%
1.6%
1970-2004
24.7
0.6
25.2
0.7
72.2
1.6
1980-2004
20.7
0.8
20.6
0.8
36.8
1.3
1990-2004
10.4
0.7
11.0
0.7
10.0
0.6
2000-2004
-2.4
-0.5
-4.2
-0.9
-1.4
-0.3
Source: CRS analysis of the March income supplements to Current Population Survey.
Household Income at the 75th and 25th Percentiles
Household incomes have changed not only at the median — the middle of the
distribution — but also at the upper and lower ends of the income distribution. At
the 75th percentile — marking the point that separates the highest quarter of
household incomes from the lowest three-quarters — incomes grew by 47% among
nonelderly households from 1969 to 2004 and by 64% among households in which
the householder or spouse was 65 or older. In constant 2004 dollars, this represented
an increase in household income of $27,234 for nonelderly households at the 75th
percentile and an increase of $18,186 for elderly households at the 75th percentile.
(See Tables 17 and 18.)
Table 17. Household Income at 75th and 25th Percentiles,
by Age of Householder
(in 2004 dollars)
All
Householder and
Householder or Spouse
Income
Households
Spouse Under Age 65
65 or Older
Year
75th
25th
75th
25th
75th
25th
percentile percentile percentile percentile percentile
percentile
1969
$53,789 $18,886
$57,632 $25,614
$28,414
$7,718
1970
53,909
18,648
57,940
24,864
28,124
7,830
1980
58,962
18,839
63,446
24,153
32,930
10,265
1990
66,319
20,301
71,570
25,441
41,214
12,463
2000
78,694
23,192
85,298
28,163
45,846
13,913
2004
78,050
22,200
84,866
26,880
46,600
14,199
Source: CRS analysis of the March income supplements to Current Population Survey.

CRS-31
At the lower end of the income distribution, rates of growth in household
income differed dramatically between elderly and nonelderly households over the
period from 1969 to 2004. Among nonelderly households, income at the 25th
percentile grew by just 5% in real terms from 1969 to 2004. During the four years
from 2000 to 2004, household incomes at the 25th percentile fell by 5%. Among
elderly households, however, incomes at the 25th percentile grew by 84% from 1969
through 2004, and by 2% from 2000 through 2004. The steady growth in income
among elderly households in the lowest quartile of the income distribution since the
late 1960s is largely due to the fact that the great preponderance of their income
comes from Social Security. Initial Social Security benefits are based on the real
value of average lifetime wages, and the benefit formula is designed to replace a
higher percentage of average lifetime earnings for low-wage workers. During the
period from 1969 through 2004, the Social Security national average wage index
grew from $5,894 to $35,649, an average rate of increase of 5.0% per year.29 During
the same period, real wages grew by about 1% per year, which was reflected in higher
real Social Security benefits for new retirees each year.30 Once Social Security
benefits begin, they are adjusted each year by the rate of change in the Consumer
Price Index, thus preventing the real value of the benefit from being eroded by
inflation.
Table 18. Cumulative Percentage Change in Household Income
at 75th and 25th Percentiles, by Age of Householder
(in 2004 dollars)
All
Householder and
Householder or
Households
Spouse under Age 65
Spouse 65 or Older
Income Years
75th
25th
75th
25th
75th
25th
percentile percentile percentile percentile percentile percentile
1969-2004
45.1%
17.5%
47.3%
4.9%
64.0%
84.0%
1970-2004
44.8
19.0
46.5
8.1
65.7
81.3
1980-2004
32.4
17.8
33.8
11.3
41.5
38.3
1990-2004
17.7
9.4
18.6
5.7
13.1
13.9
2000-2004
-0.8
-4.3
-0.5
-4.6
1.6
2.1
Source: CRS analysis of the March income supplements to Current Population Survey.
29 See [http://www.ssa.gov/OACT/COLA/AWI.html].
30 From 1969 to 2004, the price level as measured by the Personal Consumption Expenditure
Index of the National Income and Product Accounts rose from 25.3 to 108.2, an average
annual rate of increase of 4.1%.(Source: Economic Report of the President, February 2006,
Table B-7, page 290.)

CRS-32
Poverty Status of the Elderly
The steady rise in the median income of elderly households over the past 35
years coincided with a dramatic decline in the proportion of older Americans living
in poverty. In 1969, nearly a quarter of Americans 65 and older had family incomes
below the federal poverty threshold, and the poverty rate among the population 65
and older was more than double the poverty rate among adults 18 to 64 years old. By
the early 1990s, the poverty rate among people 65 and older had fallen below the
poverty rate among adults age 18 to 64 and it has remained lower since that time.
During the same period of time, the ratio of the poverty threshold to the median
income of the population has fallen as wage growth has outpaced the rate of inflation,
by which the poverty threshold is adjusted each year. As a consequence, those who
are officially poor have become progressively less well-off than households with
income at the median. Nevertheless, the substantial reduction in the proportion of
older Americans living in poverty must be regarded as one of the most important
economic developments of the past 35 years. Without the decline in elderly poverty,
the economic burden of supporting those who can no longer work in old age would
weigh that much more heavily on their adult children, and many millions of older
Americans would likely have to give up their own households to live with other
relatives.
Decline in Poverty Among the Elderly
The percentage of older Americans in poverty has fallen dramatically over the
past 35 years, from one-in-four people age 65 and older in 1969 to just one-in-ten in
2004. As the result of the dramatic decline in elderly poverty and a relatively
constant poverty rate among working-age Americans, by the early 1990s, the poverty
rate for people 65 and older fell below that of adults age 18 to 64. By 2004, the
poverty rate among Americans age 65 and older had fallen to 9.8%, or 1.5 percentage
points lower than the poverty rate among adults age 18 to 64. However, while the
share of people age 65 and older in poverty has fallen, the number of poor elderly has
remained relatively constant since the mid-1970s because of the growth in the total
number of elderly. (See Table 19.)

CRS-33
Table 19. Poverty Status of Individuals Age 18 and Older,
by Year
People 18 to 64 Years Old
People 65 and Older
Year
Number
Number in Percent in
Number
Number in Percent in
(000s)
poverty
poverty
(000s)
poverty
poverty
1970
113,554
10,187
9.0
19,470
4,793
24.6
1975
124,122
11,456
9.2
21,662
3,317
15.3
1980
137,428
13,858
10.1
24,686
3,871
15.7
1985
146,396
16,598
11.3
27,322
3,456
12.6
1990
153,502
16,496
10.7
30,093
3,658
12.2
1995
161,508
18,442
11.4
31,658
3,318
10.5
2000
173,638
16,671
9.6
33,566
3,323
9.9
2004
182,121
20,514
11.3
35,213
3,457
9.8
Source: See [http://www.census.gov/hhes/www/poverty/histpov/hstpov3.html].
The decline in the elderly poverty rate was due both to the growth of older
Americans’ incomes and also to the constant real value of the poverty threshold. The
growth in wages from year to year reflects both the rising general level of prices and
gains in labor productivity. Because labor becomes more productive over time —
as a result of better education and training, improved methods of production and
distribution, and new technologies — wages rise faster than prices over the long run.
As noted above, from 1969 through 2004 average wages in the United States grew
by about 1% more per year than prices. Because both initial Social Security benefits
and most pension benefits are based on measures of career-average wages, higher real
wages have caused the average Social Security and pension benefits of each
successive cohort of retirees to exceed those of earlier cohorts. The rising real values
of these sources of income have led to an increasing standard of living among older
Americans. Further, retirees’ Social Security benefits are increased each year by a
cost-of-living adjustment that is equal to the rate of inflation. This has kept the real
value of Social Security benefits from eroding over time.31
The increase in average real income of the elderly population, which is due
mainly to the higher career-average earnings of each successive generation of
retirees, is not reflected in the federal poverty thresholds. The poverty thresholds are
intended to measure the income necessary to maintain a minimally adequate standard
of living, and thus they are increased each year by the rate of inflation. In other
words, the poverty thresholds measure the amount of income necessary to maintain
a constant standard of living, not the amount of income needed to keep up with the
rising average standard of living of the working population.
31 Many public-sector pension plans make regular cost-of-living adjustments (COLAs) to
retirees’ pensions. Most private-sector pensions do not pay annual COLAs.

CRS-34
The federal poverty thresholds were developed by government economists in the
1960s to serve as a standard for measuring income adequacy.32 The thresholds are
adjusted annually by the rate of inflation as measured by the consumer price index
(CPI-U) so that the real (inflation-adjusted) value remains constant over time.
Because the real value of the poverty threshold has remained constant over time
while the real income of the elderly has been rising, the poverty threshold has fallen
as a percentage of median income. For example, in 1970 the poverty threshold
($1,861) for an individual 65 or older was equal to 89% of the median income
($1,959) of single people age 65 and older. Over the next 34 years, the median
income grew 666% while the poverty threshold only rose 387%. Thus, the 2004
poverty threshold for a single person age 65 and older ($9,060) was only 60% of the
median income for an individual in that age group ($15,000). In the future, other
things being equal, the disparity between rising real incomes and a fixed real poverty
threshold will lead to a decreasing proportion of the elderly being in poverty.33
Poverty among Demographic Groups
In 2004, poverty among the elderly was highest among women, minorities,
single persons, those with less education, and the very old. (See Table 20.) Twelve
percent of women over age 65 were in poverty compared with 7% of men. While
only 8% of non-Hispanic whites age 65 and older were in poverty, 24% of non-
Hispanic black seniors and 14% of Hispanic seniors were in poverty. Married
couples had significantly lower levels of poverty (5%) than singles (16%). Poverty
rates are higher than average among the less-educated elderly. In 2004, older
individuals with less than a high school education had a poverty rate (19%) that was
more than three times as high as the poverty rate among college graduates. Finally,
poverty rates are higher among individuals age 80 and older than among those 65 to
79 years old. Just 9% of people between age 65 and 79 were poor in 2004, compared
with 11% of individuals age 80 and older.
32 The poverty threshold has changed slightly over time. For a comprehensive history of the
poverty threshold see Gordon M. Fisher “The Development and History of the Poverty
T h r e s h o l d ” S o c i a l S e c u r i t y B u l l e t i n , vo . 5 5 n o . 4 , 1 9 9 2 a t
[http://www.ssa.gov/history/fisheronpoverty.html].
33 Barbara Butrica and Cori Uccello, “How Will Boomers Fare at Retirement?” The Urban
I n s t i t u t e R e t i r e m e n t P r o j e c t , n o . 2 , N o v e m b e r 2 0 0 5 a t
[http://www.urban.org/UploadedPDF/900892_boomers.pdf]

CRS-35
Table 20. Poverty Status of Individuals Age 18 and Older
in 2004, by Demographic Characteristics
Percent in
Percent in
Number in
Poverty,
Poverty,
Poverty, 65+
Age 18-64
Age 65+
(000s)
All
11.3%
9.8%
3,457
Gender
Male
9.6
7.0
1,058
Female
12.8
12.0
2,398
Race and Ethnicity
White, Not Hispanic
8.3
7.5
2,150
Black, Not Hispanic
20.3
23.7
687
Othera
11.1
18.7
410
Hispanic
18.3
14.1
210
Marital Status
Married
5.8
4.5
871
Single, divorced,
17.9
16.2
2,585
widowed
Education
Less than High
26.6
18.7
1,719
School
High School
12.8
7.9
1,008
Graduate
Some College
8.9
5.6
369
College Graduate
4.1
5.4
360

Source: CRS analysis of the 2005 March income supplements to Current Population Survey.
a. “Other” includes mainly individuals of Asian and Native American heritage.
Changes in Income as People Age
As workers pass age 60, they begin — gradually, in many cases — to withdraw
from the paid labor force and to rely on pensions, Social Security, and personal
savings to supply more of their income. A positive development since the 1960s has
been that the rate of decline in median personal and household income (measured in
constant dollars) that occurs as individuals reach their mid- to late 60s appears to
have slowed. This has occurred despite the fact that a smaller proportion of people
over age 60 work today than was the case 35 years ago. The long-term growth in real
wages — and in the real value of Social Security and pension benefits that are based
on these wages — has contributed greatly to the economic well-being of today’s
retirees. In addition, the growth — recently stalled — in the proportion of older
Americans with income from pensions has helped more people maintain a standard
of living in retirement that is comparable to that which they enjoyed while working.
More Americans 65 and older also receive asset income than did in the 1960s and

CRS-36
1970s, but the amounts received as income from this source (with a median value of
$817 in 2004) pale in comparison to the amounts that most people 65 and older
receive from earnings, pensions, and Social Security.
As was noted earlier in Table 6, employment rates are lower among people 60
to 64 years old than among those under age 60, and they fall even further after age
65. Among people 65 to 69 years old in 2004, for example, just 33% were employed
at any time during the year. Rates of full-time employment also decline as workers
age, and among older Americans who remain in the workforce after age 60, many
work part-time. In 2004, while 86% of people 55 to 59 years old worked full-time,
this proportion fell to 78% among workers 60 to 64 years old and to 59% among
those 65 to 69 years old. As older workers reduce their hours of work and gradually
withdraw from the labor force, their earnings fall and they begin to rely to a greater
extent on income from pensions, savings, and after age 62, Social Security.34
Earnings, often from part-time employment, remain an important source of income
for many older people, but both their average hours worked and average earnings per
hour are lower than in their peak earning years.
Changes in Individual and Household Income
The income data collected in the CPS allow us to see how the median incomes
differ between people of different ages, and how the median incomes of people in
different age groups change as they age. For example, the top row of Table 21
shows that individuals 55 to 59 years old had a median income in 1969 of $23,437,
measured in 2004 dollars. People who were 55 to 59 years old in1969 all were born
between 1910 and 1914. Demographers refer to people born in the same year or
group of years as a “birth cohort.”
Following this same birth cohort forward five years to 1974, Table 21 shows
that individuals who were 60 to 64 years old that year had a median income of
$18,562, or $4,875 lower than the median income of this same birth cohort five years
earlier. Five years later in 1979, people in the 1910-1914 birth cohort were 65 to 69
years old and these individuals had a median income of $12,039, or $11,398 less than
the median income of individuals in this birth cohort 10 years earlier. Following
each of the pairs of diagonal arrows in this table allows one to see how median
individual incomes of successive cohorts changed as they aged. The most recent
birth cohort for which income data are available over a 10-year period comprises
people who were 55 to 59 in 1994. In constant dollars, the median income of
individuals in this birth cohort declined from $24,068 in 1994 to $22,321 in 1999 (at
age 60-64) and to $18,199 in 2004 (at age 65 to 69).
The third column of Table 21 shows median incomes at age 60 to 64 and age
65 to 69 as a percentage of the median income of the same birth cohort at age 55 to
59. For example, the median income of individuals who were 60 to 64 years old in
1974 was 79.2% of the median income of this same cohort in 1969, when they were
34 Social Security retired worker benefits are first available at age 62. Workers who become
disabled may be eligible for Social Security Disability Insurance at any age. Rates of
disability are higher among people 55 and older than among younger individuals.

CRS-37
55 to 59 years old. Their median income in 1979, at age 65 to 69, was 51.4% of their
income in 1969 at age 55 to 59. Looking at the most recent cohort for which income
data are available for a 10-year period, the reader can see that the median individual
income of people who were 60 to 64 years old in 1999 was 92.7% of the median
income of the same cohort in 1994 at age 55 to 59. In 2004, when the people in this
cohort had reached age 65 to 69, their median income was 75.6% of the median
income of people who were age 55 to 59 ten years earlier in 1994.
Table 22 shows the median household income in selected years of people in
these same three age cohorts as Table 21: 55 to 59, 60 to 64, and 65 to 69. Here a
similar pattern is apparent, with median household income falling as people age.
Among those who were 55 to 59 in 1969, median household income was $39,151,
measured in 2004 dollars. Five years later in 1974, the median household income of
this age group, who were then 60 to 64 years old, was $33,789, or $5,362 lower than
the household income of this cohort in 1969. By 1979, at age 65 to 69, the median
household income of this cohort was $25,357, or $13,794 lower than the median
household income of this cohort 10 years earlier. For the most recent 10-year period,
median household income among those who were 55 to 59 in 1994 was $50,400 in
that year. Five years later in 1999 when they were 60 to 64 years old, the median
household income of this cohort was $48,211. By 2004, at age 65 to 69, the median
household income of this cohort was $37,989.35
The third column of Table 22 shows median household incomes at age 60 to 64
and age 65 to 69 as a percentage of the median income of the same birth cohort at age
55 to 59. The median household income of people age 60 to 64 in 1974 was 86.3%
of the median household income of this same cohort five years earlier when they
were 55 to 59 years old. By 1979, the median household income of people age 65
to 69 was 64.8% of the median household income of this same age cohort 10 years
earlier when they were 55 to 59 years old. For the most recent cohort for which ten
years of income data are available, the median household income of people who were
60 to 64 in 1994 was 95.7% of the median income of the same cohort five years
earlier at age 55 to 59. By 2004, when they had reached age 65 to 69, the median
household income of this cohort was 75.4% of the median income of the same cohort
ten years earlier when they were 55 to 59 years old.
The data displayed in Tables 21 and 22 show that real median incomes have
risen at both the individual and household level among all three age groups since
1969. Real median individual income was more variable than median household
35 Average household size declines as people age, but the decline in the median household
incomes shown in Table 22 appears to be due mainly to the declining income of the people
in households rather than to a decline in the number of people in the household. When we
looked at households of fixed sizes, we still found that median household incomes fell
between ages 55 to 59 and 65 to 69. For example, the median household income in 1979
of persons 65 to 69 years old who lived in two-person households was 73% of the median
household income in 1969 of persons 55 to 59 years old who lived in two-person
households. The equivalent ratio for one-person households was 67%. Likewise, the
median household income in 2004 of persons 65 to 69 years old in two-person households
was 81% of the median household income in 1994 of persons 55 to 59 years old in two-
person households. The equivalent ratio for one-person households was 76%.

CRS-38
income, but the trend was upward for both. Some of the increase in real median
incomes in the three age categories was due to increased employment among older
persons, particularly women. Since the mid-1990s, employment among men 60 and
older also has risen. Over time, the ratio of median individual and household
incomes at ages 65 to 69 to income at age 55 to 59 also has risen, although the
changes in the household income ratios were variable rather than increasing steadily.
Although, at the median, the trend in incomes shown in Tables 21 and 22 is
positive, it is important to keep in mind that, by definition, half of all individuals and
households have incomes less than their respective medians. For some people, their
income in old age may be less than the amount needed to maintain their desired
standard of living. Many financial analysts recommend that workers aim for an
income replacement rate of 70% or more in their first year of retirement if they want
to maintain the standard of living they enjoyed while working.36 The median
household income of people who were 65 to 69 years old in 2004 was 75% of the
household income of people in this age cohort 10 years earlier when they were 55 to
59 years old. Thus, a household that was at exactly the median income for its birth
cohort in both 1994 and 2004, and that had retired in the interim, would probably
find itself able to maintain a standard of living in retirement that was comparable to
the lifestyle they had enjoyed while working. Those whose income from pensions,
savings, and Social Security do not reach this threshold — whose income in
retirement falls below 70% of their pre-retirement income — may have to make
difficult adjustments in order to keep their spending in retirement within their more
limited incomes. For most individuals, the most direct way to boost the income they
will have in retirement, both in absolute terms and relative to pre-retirement income,
would be to increase the percentage of earnings that they save in employer-sponsored
retirement plans, individual retirement accounts, or other savings plans.
36 The ratios in Table 22 are not true “replacement rates,” which measure income in the first
year or (or n years) of retirement to income in the last year (or n years) of employment. The
ratios in the tables are based on the median income of everyone in each age cohort, whether
working full-time, part-time, or fully retired. Nevertheless, they demonstrate the effect on
the median income of each cohort of the gradual retirement of the members of the cohort.


CRS-39
Table 21. Median Individual Income at Five-Year Intervals
(in 2004 dollars)
Source: CRS analysis of the March income supplements to Current Population Survey. Incomes have
been adjusted to 2004 dollars based on the NIPA PCE Index.


CRS-40
Table 22. Median Household Income at Five-Year Intervals
Source: CRS analysis of the March income supplements to Current Population Survey. Incomes have
been adjusted to 2004 dollars based on the NIPA PCE Index.

CRS-41
Conclusion
The past 35 years have seen a substantial increase in the real incomes of older
Americans and a dramatic decline in the percentage of people 65 and older living in
poverty. These developments can be attributed both to long-term economic growth
and increases in real wages and also to public policies that have contributed to
improving pension security and encouraged retirement saving. Perhaps the single
greatest contributor to the improved economic situation of the elderly in the United
States has been the growth in real wages that they experienced during their working
lives. In real terms, measured in 2004 dollars, the median annual earnings of workers
between the ages of 18 and 64 rose from $21,777 in 1969 to $28,251 in 2004.
Among households in which neither the householder nor spouse was 65 or older, real
median income rose from $40,278 in 1969 to $50,446 in 2004. Growth in real wages
— and in the Social Security and pensions based on these wages — has contributed
greatly to the economic well-being of today’s retirees.
Actions taken by Congress over the past 35 years also have affected the incomes
of older Americans. In 1974, Congress passed the Employee Retirement Income
Security Act (ERISA), which set funding standards and vesting requirements for
pension plans sponsored by employers in the private sector. This law also authorized
individual retirement accounts (IRAs) which encourage workers to save for
retirement by deferring income taxes on some contributions and on investment
earnings. In the Revenue Act of 1978, Congress added section 401(k) to the Internal
Revenue Code, which authorized employers to establish retirement savings plans for
employees in which contributions and investment earnings grow on a tax-deferred
basis until retirement.
Legislation affecting Social Security has also contributed to increasing the
income of older Americans. The Social Security Amendments of 1972 provided for
automatic cost-of-living adjustments (COLAs), which prevent the real value of Social
Security benefits from being eroded by inflation. The 1977 Social Security
amendments established the current benefit formula, in which initial benefits are
based on each worker’s career-average wages, indexed to current values based on a
national average wage index. This assures that each successive cohort of workers
receives a Social Security benefit that reflects the growth in real wages that occurred
during their working lives. In 2000, Congress repealed the “earnings test” for
workers who have reached the Social Security full retirement age. This test reduces
benefits for Social Security beneficiaries whose earnings exceed a threshold set in
law. The earnings test now applies only to beneficiaries under the full retirement age.
As the 78 million members of the “baby boom” — people born between 1946
and 1964 — approach retirement, Congress is likely to continue to debate changes
to the federal laws affecting pensions, retirement savings, and Social Security.
Future retirees are likely to live longer and will need to accumulate relatively greater
retirement assets than past retirees if they are to maintain their standard of living
through these longer periods of retirement. With Social Security facing a financial
shortfall and the number of private-sector pensions continuing to decline, it is likely
that a relatively greater share of current workers’ future retirement income will have
to be financed from their own personal savings.

CRS-42
Appendix: Historical Tables of Individual and
Household Income

CRS-43
Table A1. Percentage of People Age 65 and Older with Income from Each Source, by Year
Number of
Number with
Percentage
Percentage
Percentage
Percentage
Percentage
Percentage
Percentage
Income
People 65+
Any Income
with Any
With
Receiving
Receiving
Receiving
Receiving
Receiving
Year
(thousands)
(thousands)
Income
Earnings
Social Security Asset Income
Pension Income
Welfare
Other Incomea
1969
18,899
17,026
90.1%
24.9%
74.7%
39.4%
a
9.2%
24.6%
1970
19,231
17,393
90.4
23.4
75.9
39.5
a
9.2
24.2
1975
21,662
20,311
93.8
19.5
83.5
45.4
22.3%
10.0
8.2
1980
24,686
24,353
98.7
16.6
90.5
67.1
26.6
8.5
6.9
1981
25,231
24,947
98.9
16.1
91.2
67.3
27.2
7.5
6.1
1982
25,738
25,432
98.8
15.9
90.9
68.5
27.5
6.8
6.3
1983
26,291
25,938
98.7
15.1
91.2
69.7
29.2
7.0
6.3
1984
26,818
26,522
98.9
15.1
91.6
68.6
29.8
7.0
6.7
1985
27,322
26,976
98.7
15.0
92.0
67.6
30.3
6.6
6.2
1986
27,975
27,617
98.7
14.7
91.7
67.9
31.3
6.0
6.6
1987
28,487
28,196
99.0
15.6
91.1
69.6
34.5
5.5
7.9
1988
29,022
28,747
99.1
16.3
90.9
68.4
35.2
5.6
7.4
1989
29,566
29,320
99.2
16.3
90.8
69.6
35.8
6.0
8.1
1990
30,093
29,734
98.8
16.3
91.0
69.9
37.0
5.5
7.7
1991
30,590
30,256
98.9
15.5
90.2
69.2
37.7
5.9
7.9
1992
30,870
30,538
98.9
14.7
91.7
68.0
38.0
5.9
7.5
1993
30,779
30,223
98.2
15.8
90.5
67.8
37.5
5.0
8.1
1994
31,267
30,676
98.1
15.4
90.8
68.0
36.2
4.9
7.4
1995
31,658
31,081
98.2
15.7
90.8
67.2
35.4
4.3
7.1
1996
31,877
31,199
97.9
15.6
90.1
64.1
35.2
4.7
6.9
1997
32,082
31,401
97.9
15.3
89.9
62.9
35.5
4.4
6.8
1998
32,394
31,694
97.8
15.9
89.4
63.8
35.9
3.9
6.7
1999
32,621
31,978
98.0
16.9
89.5
62.7
36.3
4.1
6.2
2000
32,979
32,258
97.8
16.9
89.8
59.2
34.9
3.8
6.0
2001
33,770
32,911
97.5
16.2
90.1
58.4
34.5
3.7
6.3
2002
34,234
33,334
97.4
16.6
88.8
55.8
34.4
3.6
6.1
2003
34,659
33,779
97.5
17.5
89.1
56.4
35.3
3.7
5.6
2004
35,213
34,187
97.1
18.0
88.2
55.5
35.1
3.5
5.5
Source: CRS analysis of the March income supplements to Current Population Survey. Data represent the civilian, noninstitutional population.
a. Prior to the March 1976 survey, the CPS included income from pensions with “other income.”

CRS-44
Table A2. Median Annual Income of Individuals Age 65 and Older, by Income Source
(in 2004 dollars)
Income
Number
Median
Median
Median
Median
Median
Median
Median All
Year
With Income
Total Income
Earnings
Social Security
Asset Income
Pension Income Welfare Income
Other Incomea
a
1969
17,026
$8,196
$6,856
$4,866
$2,134
$3,125
$5,123
a
1970
17,393
8,271
6,644
5,267
2,038
3,180
5,189
1975
20,311
10,147
6,896
7,166
2,159
6,081
2,549
3,118
1980
24,353
10,723
8,281
7,212
1,656
5,995
2,374
2,459
1981
24,947
11,161
7,698
7,367
1,901
5,603
2,395
2,441
1982
25,432
11,788
8,321
7,817
2,071
5,641
2,338
2,318
1983
25,938
12,008
8,288
7,874
2,072
6,050
2,504
2,590
1984
26,522
12,394
7,487
7,987
2,637
5,990
2,453
2,496
1985
26,976
12,626
7,979
8,002
2,553
5,889
2,513
2,416
1986
27,617
12,725
9,394
8,094
2,431
6,289
2,608
2,377
1987
28,196
13,125
9,118
7,861
2,369
6,838
2,486
2,443
1988
28,747
13,027
8,887
7,964
2,462
6,771
2,508
2,456
1989
29,320
13,306
9,309
8,026
2,349
6,723
2,381
2,656
1990
29,734
13,523
8,705
8,036
2,385
6,991
2,314
2,893
1991
30,256
13,242
9,008
8,187
1,939
6,963
2,683
3,102
1992
30,538
12,936
9,400
8,032
1,507
6,783
2,261
3,166
1993
30,223
13,124
9,577
8,472
1,228
7,352
2,824
2,728
1994
30,676
13,420
9,620
8,962
1,203
7,215
2,872
2,525
1995
31,081
13,902
10,713
9,123
1,510
7,177
2,684
3,644
1996
31,199
14,014
10,430
9,217
1,465
7,697
2,849
3,678
1997
31,401
14,294
11,012
9,518
1,700
8,024
2,897
4,012
1998
31,694
14,743
11,233
9,661
1,685
8,088
3,087
4,044
1999
31,978
15,178
11,491
9,884
1,657
8,309
3,354
3,804
2000
32,258
14,727
12,937
9,903
1,477
8,625
3,247
3,880
2001
32,911
14,715
13,091
10,137
1,584
8,794
3,168
3,800
2002
33,334
14,668
15,617
10,270
1,070
8,746
3,536
3,997
2003
33,779
14,796
16,349
10,529
896
9,196
3,629
4,793
2004
34,187
15,189
15,000
10,399
817
9,600
3,600
4,799
Source: CRS analysis of the March income supplements to Current Population Survey. Data represent the civilian, noninstitutional population.
a. Prior to the March 1976 survey, the CPS included income from pensions with “other income.” Incomes have been adjusted to 2004 dollars based in the Personal Consumption
Expenditure Index of the National Income and Product Accounts.

CRS-45
Table A3. Mean and Median Annual Household Income in 2004 Dollars
All Households
Householder and Spouse Under Age 65a
Householder or Spouse 65 or Olderb
Number of
Mean
Median
Number of
Mean
Median
Number of
Mean
Median
Income
Households
Household
Household
Households
Household
Household
Households
Household
Household
Year
(thousands)
Income
Income
(thousands)
Income
Income
(thousands)
Income
Income
1969
63,365
$40,378
$35,539
50,667
$44,886
$40,278
12,698
$22,393
$14,498
1970
64,648
40,423
35,461
51,644
44,986
40,295
13,005
22,303
14,641
1975
72,867
40,945
35,200
57,715
45,288
40,317
15,152
24,406
16,963
1980
82,368
42,847
36,639
64,988
47,239
41,853
17,380
26,424
18,425
1981
83,527
43,058
36,119
65,737
47,268
41,442
17,790
27,501
19,115
1982
83,918
43,476
36,045
65,771
47,503
41,099
18,147
28,878
20,024
1983
85,407
43,540
35,887
67,008
47,605
41,172
18,399
28,733
20,455
1984
86,789
45,700
37,224
68,085
49,776
42,096
18,704
30,859
21,689
1985
88,458
46,407
38,020
69,319
50,742
43,175
19,138
30,707
21,539
1986
89,479
47,767
39,064
69,903
52,320
44,488
19,576
31,509
22,134
1987
91,124
48,567
39,520
71,102
53,375
45,600
20,022
31,490
22,265
1988
92,830
48,918
39,664
72,515
53,726
45,427
20,315
31,760
22,222
1989
93,347
50,009
40,523
72,607
54,851
46,236
20,741
33,061
22,325
1990
94,312
49,053
40,036
73,251
53,619
45,476
21,061
33,173
22,910
1991
95,669
48,156
38,760
74,268
52,861
45,065
21,401
31,827
22,197
1992
96,391
47,929
38,308
75,011
52,614
44,462
21,380
31,493
21,872
1993
97,262
48,250
38,068
75,823
52,856
43,913
21,439
31,961
22,104
1994
99,087
49,138
38,664
77,054
54,013
44,691
22,033
32,088
22,118
1995
99,683
52,870
40,018
77,546
58,228
46,027
22,137
34,101
22,780
1996
101,081
54,270
40,518
78,979
59,847
46,923
22,102
34,342
22,810
1997
102,584
56,289
41,839
80,234
61,763
47,953
22,350
36,635
23,918
1998
103,991
58,187
43,590
81,600
63,745
50,233
22,391
37,934
24,810
1999
104,780
59,446
44,813
82,201
65,064
51,825
22,579
38,994
25,592
2000
106,509
61,460
45,302
83,653
67,742
52,680
22,856
38,466
25,570
2001
109,387
61,433
44,484
85,839
67,812
52,004
23,548
38,178
25,013
2002
111,381
60,190
44,119
87,694
66,435
51,425
23,687
37,068
24,725
2003
112,015
60,381
44,150
87,860
66,378
51,100
24,154
38,568
25,031
2004
113,155
60,530
44,214
88,842
66,433
50,466
24,313
38,963
25,210
Source: CRS analysis of the March income supplements to Current Population Survey. Data represent the civilian, noninstitutional population.
a. Neither the householder nor the householder’s spouse were 65 or older.
b. Either the householder or householder’s spouse was 65 or older. Incomes have been adjusted to 2004 dollars based in the Personal Consumption Expenditure Index
of the National Income and Product Accounts