April 8, 1999

Social Security and Poverty Among the Elderly:
A National and State Perspective

Executive Summary*

by Kathryn H. Porter, Kathy Larin and Wendell Primus

View the report.  View PDF file of complete report

View press release

State Fact Sheets:

Census data show that without Social Security, nearly half — 47.6 percent— of the U.S. population age 65 and older would have been poor in 1997.(1) Social Security reduced the poverty rate among elderly people in 1997 by three-quarters, to 11.9 percent. Social Security lifted 11.4 million elderly people out of poverty in that year.

State-by-State Effects

Census data also demonstrate that Social Security substantially reduces poverty among the elderly in every state. (See Table 1.)

Nationally, Social Security lowered the number of elderly poor from 15.3 million to 3.8 million in 1997, lifting from poverty nearly three of every four elderly people who would have been poor without it. Similarly, in a substantial majority of states, Social Security lifted from poverty approximately three-quarters of the elderly who would be poor in its absence. (See Table 2.)

Effects on Women

The majority of elderly people whom Social Security lifts from poverty are women. In 1997, women accounted for more than three of every five elderly people lifted from poverty by Social Security; the program lifted 7 million elderly women and 4.4 million men out of poverty that year. It lowered the number of poor elderly women from 9.8 million to 2.7 million and shrank the number of poor elderly men from 5.5 million to 1.1 million.

Dispelling Possible Confusion About the State Numbers

At first blush, a few of the state figures presented here may seem a bit confusing. In California, for example, Social Security lifted 30.7 percent of all elderly people from poverty. Also in the Golden State, 71 percent of the elderly people who would have been poor in the absence of Social Security benefits were lifted from poverty by these benefits. What is the difference between these two percentages?

During the period we examined, the Census data show there were an average of 3.35 million elderly people in California. Some 1.45 million of these 3.35 million people — 43.2 percent of them — were poor before receipt of Social Security benefits. After Social Security benefits are considered, 421,000 — or 12.5 percent of the elderly people in the state — remained poor. Social Security thus lifted 1.03 million elderly people from poverty.

These 1.03 million people whom Social Security lifted from poverty represented 30.7 percent of the 3.35 million elderly people in California. These 1.03 million people also represented 71 percent of the 1.45 million elderly people in the state who would have been poor in the absence of Social Security benefits.

Without Social Security benefits, 52.6 percent of elderly women would have had incomes below the poverty line in 1997. Social Security reduced the poverty rate for elderly women to 14.7 percent. Here, too, the state figures track the national figures. (See Tables 3 and 4.)

Table 1

Impact of Social Security on Elderly People

Percentage of Elderly
Who are Poor Before
Social Security
Percentage of Elderly
Who are Poor After
Social Security
Percentage of Elderly
Lifted from Poverty
by Social Security
Alabama 56.0% 17.6% 38.4%
Arizona 43.4% 10.5% 32.9%
Arkansas 60.4% 19.9% 40.5%
California 43.2% 12.5% 30.7%
Colorado 37.2% 8.5% 28.8%
Connecticut 40.2% 6.2% 34.0%
Florida 48.7% 11.9% 36.9%
Georgia 49.7% 15.6% 34.1%
Hawaii 30.9% 9.2% 21.7%
Idaho 49.3% 9.4% 40.0%
Illinois 48.4% 10.1% 38.3%
Indiana 54.0% 9.7% 44.3%
Iowa 49.5% 9.8% 39.7%
Kansas 49.4% 11.6% 37.8%
Kentucky 52.6% 13.8% 38.8%
Louisiana 54.9% 19.0% 35.9%
Maine 54.6% 13.0% 41.5%
Maryland 41.3% 11.0% 30.3%
Massachusetts 47.4% 11.4% 36.0%
Michigan 47.9% 10.0% 37.9%
Minnesota 51.7% 11.5% 40.2%
Mississippi 60.3% 21.3% 39.0%
Missouri 48.1% 11.4% 36.7%
Nebraska 52.7% 10.9% 41.8%
Nevada 45.1% 9.3% 35.8%
New Hampshire 48.3% 8.6% 39.8%
New Jersey 43.8% 9.7% 34.0%
New Mexico 48.4% 17.0% 31.3%
New York 50.0% 15.1% 34.9%
North Carolina 54.0% 15.7% 38.2%
Ohio 48.9% 10.6% 38.2%
Oklahoma 54.7% 14.9% 39.9%
Oregon 46.4% 6.7% 39.7%
Pennsylvania 53.0% 10.6% 42.4%
Rhode Island 54.6% 12.7% 42.0%
South Carolina 57.0% 18.2% 38.8%
Tennessee 53.6% 19.1% 34.6%
Texas 50.3% 16.5% 33.8%
Utah 40.0% 6.4% 33.6%
Virginia 44.7% 13.8% 30.8%
Washington 42.3% 9.4% 32.9%
West Virginia 58.5% 15.5% 43.0%
Wisconsin 47.7% 9.0% 38.6%
U.S. Total* 48.7% 12.6% 36.1%
Note: The following states have been omitted because the sample sizes in these states are too small to obtain reliable estimates: Alaska, Delaware, District of Columbia, Montana, North Dakota, South Dakota, Vermont, and Wyoming.
* The US total figures in this table differ from the national figures in the text of the report because the US total figures in this table, like the state figures, are based on pooled Census data for the five years from 1993 to 1997.  The national figures in the text are based on data for 1997 only.

 

Table 2
Impact of Social Security on the Elderly Poor
Number of Elderly Who Are Poor Before Social Security
(in Thousands)
Number of Elderly Who are Poor After Social Security
( in Thousands)
Number of Elderly Lifted from Poverty by Social Security
(in Thousands)
Percentage of Elderly Who Would be Poor Without Social Security Who are Lifted from Poverty by Social Security
Alabama 324 101 222 68.6%
Arizona 235 57 177 75.8%
Arkansas 200 66 134 67.1%
California 1,450 421 1,029 71.0%
Colorado 128 30 98 77.1%
Connecticut 180 28 152 84.6%
Florida 1,166 282 883 75.6%
Georgia 369 116 253 68.5%
Hawaii 46 14 33 69.8%
Idaho 65 12 52 81.1%
Illinois 654 137 517 79.1%
Kansas 161 37 124 76.5%
Kentucky 253 66 186 73.8%
Louisiana 257 88 169 65.3%
Maine 88 21 67 75.9%
Maryland 251 67 183 73.3%
Massachusetts 357 86 271 76.0%
Michigan 561 116 445 79.3%
Minnesota 242 54 188 77.7%
Mississippi 186 66 120 64.5%
Missouri 348 83 265 76.6%
Nebraska 106 22 84 79.3%
Nevada 85 17 67 79.4%
New Hampshire 65 12 54 82.0%
New Jersey 433 96 337 77.7%
New Mexico 90 32 58 64.5%
New York 1,165 353 812 69.6%
North Carolina 479 140 339 70.9%
Ohio 686 149 537 78.3%
Oklahoma 226 61 165 72.8%
Oregon 174 25 149 85.8%
Pennsylvania 914 182 732 80.0%
Rhode Island 83 20 64 76.8%
South Carolina 223 71 152 68.2%
Tennessee 305 108 198 65.0%
Texas 904 298 607 67.1%
Utah 72 11 61 83.8%
Virginia 312 96 216 69.1%
Washington 237 54 183 77.8%
West Virginia 170 45 125 73.5%
Wisconsin 271 51 221 80.8%
U.S. Total* 15,373 3,968 11,405 74.2%
Note: The following states have been omitted because the sample sizes in these states are too small to obtain reliable estimates: Alaska, Delaware, District of Columbia, Montana, North Dakota, South Dakota, Vermont, and Wyoming.
* The US total figures in this table differ from the national figures in the text of the report because the US total figures in this table, like the state figures, are based on pooled Census data for the five years from 1993 to 1997.  The national figures in the text are based on data for 1997 only.

 

Overall Effects on Elderly Poverty

Social Security has a much larger impact in lifting elderly people from poverty than all other government programs combined. Nine of every 10 elderly people lifted from poverty by government benefit programs, including state and local cash assistance programs, are lifted out by Social Security.

Social Security also reduces the depth, or severity, of poverty among elderly people who remain poor. Researchers use a measure known as the "poverty gap" to examine the depth of poverty; the poverty gap is the total amount by which the incomes of all poor people fall below the poverty line. In 1997, the poverty gap for the elderly was $70 billion before Social Security. Social Security reduced the poverty gap among the elderly from $70 billion to $10 billion.

Figure ASocial Security also is important to millions of elderly people who are not poor. It constitutes at least 50 percent of the total income of more than half of all elderly people. It constitutes at least 90 percent of income for one-quarter of the elderly. For 15 percent of the elderly, Social Security is their sole source of income.

Social Security is the principal source of income for many middle-income elderly people as well as for those of lesser means. It makes up approximately 80 percent of the income of both the poorest fifth of elderly people and the next-to-the-poorest fifth. It provides 62 percent of the income of those in the middle fifth of the elderly population. It even provides 41.5 percent of the income of the elderly in the next-to-highest income fifth. By contrast, it makes up only 17 percent of the income of those in the top fifth of the elderly population.

 

Effects on Elderly Women

Social Security is particularly beneficial to women. Women receive 53 percent of Social Security retirement and survivor benefits, while paying 38 percent of Social Security payroll taxes.

Social Security is highly favorable to women for several reasons. It provides lifetime retirement benefits that are fully indexed for inflation. Since women tend to live longer than men, this feature of Social Security is especially important to them. In addition, Social Security has a progressive benefit structure — it provides benefits that replace a higher percentage of the earnings of low-wage workers than of highly paid workers. Since women have lower average wages than men, this aspect of the program is beneficial to them as well. Finally, women are the primary beneficiaries of Social Security's special benefits for spouses (including divorced spouses who were married at least 10 years and have not remarried) and for widows and widowers.

Women outnumber men in the elderly population, and elderly women are more likely than elderly men to be poor. This is in large part because elderly women receive less income than elderly men from earnings, pensions, and investments. Older women tend to have fewer financial assets than older men.

In 1997, some 52.6 percent of women age 65 and older were poor before receipt of Social Security benefits, compared to 40.8 percent of elderly men. After Social Security, 14.7 percent of elderly women and 8.2 percent of elderly men remained poor. Social Security narrows the gap in poverty rates between elderly women and elderly men.

Figure BIn 1997, nearly two-thirds of elderly women received a majority of their income from Social Security. For almost one-third of elderly women, Social Security provided at least 90 percent of income. It was the sole source of income for nearly one in five elderly women. For elderly women overall, Social Security benefits provided more than three-fifths of total income.

Effects by Marital Status

Married women and widows are entitled to special Social Security benefits.(2) Social Security lifts from poverty a larger proportion of elderly women who are married than of those who are widowed and a larger proportion of those who are widowed than of those otherwise unmarried.

Table 3
Impact of Social Security on Elderly Women
Percentage of
Elderly Women Who are Poor Before Social Security
Percentage of
Elderly Women Who are Poor After Social Security
Percentage of
Elderly Women Lifted from Poverty by Social Security
Alabama 60.9% 21.4% 39.5%
Arizona 46.5% 11.8% 34.7%
Arkansas 66.4% 25.1% 41.2%
California 48.7% 15.3% 33.4%
Colorado 41.8% 10.5% 31.3%
Connecticut 45.9% 8.4% 37.4%
Florida 52.8% 14.4% 38.4%
Georgia 54.8% 20.6% 34.2%
Hawaii 34.0% 11.3% 22.7%
Idaho 53.1% 11.8% 41.3%
Illinois 54.9% 13.7% 41.2%
Indiana 58.3% 11.9% 46.4%
Iowa 55.3% 12.6% 42.6%
Kansas 52.2% 12.7% 39.5%
Kentucky 56.9% 18.3% 38.6%
Louisiana 57.4% 21.3% 36.1%
Maine 57.0% 14.6% 42.4%
Maryland 44.5% 13.2% 31.3%
Massachusetts 53.3% 13.9% 39.4%
Michigan 52.6% 12.7% 39.9%
Minnesota 55.9% 14.6% 41.3%
Mississippi 63.8% 26.1% 37.7%
Missouri 53.5% 14.8% 38.7%
Nebraska 57.5% 14.4% 43.0%
Nevada 50.1% 12.1% 38.0%
New Hampshire 53.9% 12.0% 41.8%
New Jersey 50.0% 11.8% 38.2%
New Mexico 52.5% 20.2% 32.3%
New York 55.4% 18.9% 36.5%
North Carolina 58.5% 20.0% 38.4%
Ohio 54.2% 14.1% 40.1%
Oklahoma 59.2% 18.9% 40.3%
Oregon 51.6% 7.3% 44.3%
Pennsylvania 58.1% 13.5% 44.6%
Rhode Island 60.1% 16.3% 43.9%
South Carolina 58.8% 22.2% 36.6%
Tennessee 60.6% 25.6% 35.0%
Texas 55.0% 19.3% 35.6%
Utah 43.6% 7.7% 35.9%
Virginia 49.7% 17.6% 32.1%
Washington 45.4% 11.4% 34.0%
West Virginia 64.6% 19.4% 45.3%
Wisconsin 51.8% 12.0% 39.9%
Note: The following states have been omitted because the sample sizes in these states are too small to obtain reliable estimates: Alaska, Delaware, District of Columbia, Montana, North Dakota, South Dakota, Vermont, and Wyoming.

 

Table 4
Impact of Social Security on Poor Elderly Women
Number of Elderly Women Who are Poor Before
Social Security
(in Thousands)
Number of Elderly Women Who are Poor After
Social Security

(in Thousands)
Number of Elderly Women Lifted from Poverty by
Social Security
(in Thousands)
Percentage of Elderly Women Who Would be Poor Without Social Security Who are Lifted from Poverty by
Social Security
Alabama 216 75 140 64.7%
Arizona 138 35 103 74.3%
Arkansas 127 48 79 62.3%
California 927 292 635 68.6%
Colorado 80 21 59 75.0%
Connecticut 124 23 101 81.5%
Florida 720 196 524 72.7%
Georgia 243 91 152 62.4%
Hawaii 29 10 19 66.8%
Idaho 40 9 31 77.7%
Illinois 432 108 324 75.0%
Indiana 244 49 195 79.8%
Iowa 120 28 92 77.3%
Kansas 101 24 77 75.2%
Kentucky 153 50 104 67.8%
Louisiana 153 56 97 63.0%
Maine 52 13 39 74.5%
Michigan 367 88 279 76.0%
Minnesota 149 39 110 73.5%
Mississippi 114 47 67 58.9%
Missouri 225 62 163 72.8%
Nebraska 67 17 50 74.8%
Nevada 53 13 40 75.7%
New Hampshire 39 9 30 77.4%
New Jersey 292 69 224 76.5%
New Mexico 55 21 34 61.7%
New York 777 265 512 65.8%
North Carolina 301 103 197 65.8%
Ohio 453 118 335 74.0%
Oklahoma 143 45 97 68.3%
Oregon 104 15 89 86.0%
Pennsylvania 594 139 456 76.7%
Rhode Island 58 16 42 72.9%
South Carolina 139 52 87 62.3%
Tennessee 203 86 118 58.2%
Texas 564 199 366 64.8%
Utah 43 7 36 82.9%
Virginia 208 74 134 64.6%
Washington 151 39 112 74.9%
West Virginia 114 34 80 70.2%
Wisconsin 168 39 129 76.9%
Note: The following states have been omitted because the sample sizes in these states are too small to obtain reliable estimates: Alaska, Delaware, District of Columbia, Montana, North Dakota, South Dakota, Vermont, and Wyoming.

Although Social Security lifts more elderly married women from poverty, widows are the group that relies on Social Security for the greatest portion of their income. Social Security made 67.6 percent of the total income of elderly widows in 1995-97. It constituted 55.5 percent of the income of elderly married women and 56.3 percent of the income of other unmarried elderly women.

Indeed, nearly three of every four elderly widows rely on Social Security for a majority of their income. Two-fifths depend on it for at least 90 percent of their income. For one-quarter, it is their only income source.

Effects by Age

Elderly women are more likely to fall into poverty as they grow older. Among women age 65 to 75, nearly half — 46 percent — were poor in 1995-97 before counting Social Security. By contrast, among those 85 and older, 65.3 percent were poor before Social Security.

Social Security benefits narrow the disparity in poverty rates between the younger and older elderly. After receiving Social Security benefits, 12.7 percent of women 65 to 75 remain in poverty. Among women 75 to 85, some 16.6 percent remain poor after receipt of Social Security. Among those 85 and older, 20.6 percent are poor after Social Security.

Just as women become poorer as they get older, so do they rely more heavily on Social Security. The proportion of women who rely on Social Security for at least half of their income rises from 58 percent for women 65 to 75 to 75 percent for those 85 and older. Similarly, 15 percent of women age 65 to 75 rely on Social Security as their only source of income. Some 26 percent of those 85 and older do.

 

Effects by Racial and Ethnic Group

Poverty is higher among elderly people in minority racial and ethnic groups than among the white elderly. In 1997, some 46.3 percent of the white elderly were poor before receipt of Social Security.(3) Some 54.9 percent of Hispanic elderly people were poor before Social Security, as were 59.9 percent of the black elderly.

Social Security cuts poverty rates by about half or more among elderly people in all three groups. In 1997, Social Security benefits reduced the poverty rate for the Hispanic elderly from 54.9 percent to 28.1 percent. These benefits shrank the poverty rate for the black elderly from 59.9 percent to 29.1 percent. The poverty rate among the white elderly plummeted from 46.3 percent to 9.1 percent.

In numerical terms, Social Security reduced the number of white elderly poor from 12.5 million to 2.4 million in 1997. These benefits cut the number of black elderly poor from 1.6 million to 800,000 and the number of Hispanic elderly poor from 900,000 to 500,000.

Much of the difference among racial and ethnic groups in the anti-poverty effects of Social Security is due to differences in wages. Social Security benefit levels are tied to beneficiaries' wages during their working years. Many of those who are old today worked for years before the passage of civil rights laws, when wage differentials between whites and minorities were even larger than they are today. During these periods, black and Hispanic workers often were limited to low-paying occupations. The smaller impact of Social Security on poverty among the Hispanic elderly also reflects the fact that many Hispanic people who are elderly today emigrated to the United States too late in life to amass a sufficient number of years of employment to qualify for substantial Social Security benefits.

For elderly people of all races, Social Security provides a larger share of income than does any other income source. Some 43 percent of the total income of the black elderly, 41 percent of the total income of the Hispanic elderly, and 36 percent of the total income of the white elderly comes from Social Security. The black and Hispanic elderly populations rely on Social Security for somewhat larger shares of their income than the white elderly do because they tend to have less income from other sources.


End Notes:

*  The entire report can be viewed as a PDF file at /archiveSite/4-8-99socsec.pdf

1. This report examines the effect of Social Security and other government programs on poverty by comparing the number of people who would be poor if government benefits were not counted as part of their incomes to the number who are poor when certain government benefits are counted. If no government benefits were counted as part of their incomes, 49.1 percent of elderly people would be poor. Counting social insurance benefits, such as federal pensions and unemployment insurance benefits, but not Social Security, brings the elderly poverty rate down slightly to 47.6 percent. Counting Social Security brings the poverty rate down to 11.9 percent. Thus, the impact of Social Security is to reduce the poverty rate among elderly people from 47.6 percent to 11.9 percent. In this report, the phrase "before (or without) Social Security" means after counting social insurance benefits other than Social Security, but without counting any other government benefits.

2. Social Security benefits are structured so one spouse can receive Social Security benefits based on the earnings of the other spouse. This is particularly important to women because their earnings tend to be lower than the earnings of their husbands. The lower-earning spouse is entitled to a Social Security benefit based on her own earnings or a benefit equal to half of the benefit her spouse receives, whichever is greater. Similarly, widows and widowers can receive Social Security benefits based on the earnings of their deceased spouses; a widow(er) is entitled to a Social Security benefit based on her own earnings or a benefit equal to 100 percent of her deceased spouse's benefit, whichever is greater.

3. In this report, "white" means non-Hispanic white and "black" means non-Hispanic black.


View the report  Click here
for PDF file
of
complete report

getacro.gif (712 bytes)
Download the Adobe Acrobat Reader
if you do not already have one.