Social Security Lifts More People Above the Poverty Line Than Any Other Program

Social Security benefits play a vital role in reducing poverty in every state, and they lift more people above the poverty line than any other program in the United States. Without Social Security, 22.5 million more adults and children would be poor, according to analysis using the March 2021 Current Population Survey. Although most of those whom Social Security keeps out of poverty are aged 65 or older, 6.4 million are under age 65, including almost 1 million children. (See Table 1.) Social Security is particularly important for older women and people of color, who have fewer retirement resources outside of Social Security. Depending on their design, reductions in Social Security benefits could significantly increase poverty, particularly among older adults.

Effect of Social Security on Poverty (Official Poverty Measure), 2020
  Percent in Poverty  
Age Group Excluding Social Security Including Social Security Number Lifted Above the Poverty Line by Social Security
Children Under 18 17.4% 16.1% 977,000
Adults Aged 18-64 13.2% 10.4% 5,383,000
Adults Aged 65 and Over 37.8% 9.0% 16,097,000
Total, All Ages 18.3% 11.4% 22,457,000

Source: CBPP analysis of data from the U.S. Census Bureau’s March 2021 Current Population Survey

Social Security Lifts 16 Million People Aged 65 or Older Out of Poverty

Most people aged 65 and older receive the majority of their income from Social Security.[1] Without Social Security benefits, 37.8 percent of older adults would have incomes below the official poverty line, all else being equal; with Social Security benefits, only 9.0 percent do. (See Figure 1.) The benefits lift 16.1 million older adults above the poverty line, these estimates show.

Comprehensive studies that match Census survey data to administrative records suggest that the official estimates overstate older adults’ reliance on Social Security but confirm that Social Security lifts millions of older adults above the poverty line and dramatically reduces the poverty rate among people aged 65 and older. (See Appendix for more information.)

figure 1

Social Security Lifts Roughly 1 Million Children Above the Poverty Line

Social Security is important for children and their families as well as for older adults. About 6.5 million children under age 18 (9 percent of all children in the U.S.) lived in families that received income from Social Security in 2020, according to Census data. This figure includes children who received their own benefits as dependents of retired, disabled, or deceased workers, as well as those who lived with parents or relatives who received Social Security. In all, Social Security lifts almost 1 million children above the poverty line.[2]

Social Security records show that 2.8 million children under age 18 qualified for Social Security payments in December 2020. (See Appendix Table 2.) Of these, 1.2 million were the survivor of a deceased worker. Another 1.2 million received payments because their parent had a severe disability. And 341,000 children under 18 received payments because their parent was retired.[3]

Social Security Protects Groups That Are Particularly Vulnerable to Poverty

Social Security is especially important for women and people of color. Women tend to earn less than men, take more time out of the paid workforce, live longer, accumulate less savings, and receive smaller pensions. Social Security brings 9.5 million older women above the poverty line, as Table 2 shows.

Black and Latino workers benefit substantially from Social Security because they have higher disability rates and lower lifetime earnings than white workers, on average. In addition, Black workers have higher rates of premature death than white workers, and so are more likely to be eligible for Social Security survivor benefits. Latino workers have longer average life expectancies than white workers, which means they have more years to collect retirement benefits. Without Social Security, the poverty rate among older Latino adults would be 45 percent, and the poverty rate among older Black adults would be 48 percent.[4]

Effect of Social Security on Age 65+ Poverty by Sex and Race, 2020
  Percent in Poverty  
Demographic Group Excluding Social Security Including Social Security Number Lifted Above the Poverty Line by Social Security
Men 33.8% 7.6% 6,594,000
Women 41.1% 10.1% 9,503,000
White 35.8% 6.8% 12,214,000
Black 48.0% 17.1% 1,617,000
Latino 45.4% 16.6% 1,441,000
Asian 32.5% 11.4% 563,000
Other 45.4% 12.7% 261,000
Total, Age 65+ 37.8% 9.0% 16,097,000

Source: CBPP analysis of data from the U.S. Census Bureau’s March 2021 Current Population Survey

Social Security Reduces Poverty in Every State

Social Security reduces poverty among older adults dramatically in every state, as Figure 2 and Appendix Table 1 show. Without Social Security, the poverty rate for those aged 65 and over would meet or exceed 40 percent in roughly one-fourth of states; with Social Security, it is less than 10 percent in roughly three-fourths of states. Social Security lifts more than 1 million older adults above the poverty line in California, Florida, and Texas, and over half a million in Illinois, Michigan, New York, North Carolina, Ohio, and Pennsylvania.

figure 2

Social Security’s Effect on Poverty Using the Supplemental Poverty Measure

Unlike the official poverty measure, the Census Bureau’s Supplemental Poverty Measure (SPM) counts government non-cash benefits (such as food or rental assistance) and tax-based benefits (such as the Earned Income Tax Credit) as income. Comparing across all programs, Social Security is the most important anti-poverty program, according to the Census Bureau.

The SPM also subtracts from a household’s income various taxes, work expenses, and out-of-pocket medical spending. The poverty rate among adults aged 65 and older including Social Security is slightly higher using the SPM compared to the official measure (9.5 percent and 9 percent, respectively), mostly because older adults spend significantly more of their incomes on health care than working-age adults or children, and the SPM takes health care spending into account. The poverty rate for adults aged 65 and older without Social Security would reach 43 percent.

Among younger beneficiaries, poverty under the SPM is lower — both with and without Social Security — because the supplemental measure takes benefits such as SNAP and the EITC into account. Even after accounting for those benefits, Social Security lifts over 1 million children and nearly 7 million adults aged 18-64 above the poverty line.

Effect of Social Security on Poverty (Supplemental Poverty Measure), 2020
  Percent in Poverty  
Age Group Excluding Social Security Including Social Security Number Lifted Above the Poverty Line by Social Security
Children Under 18 11.2% 9.7% 1,115,000
Adults Aged 18-64 12.3% 8.8% 6,923,000
Adults Aged 65 and Over 42.5% 9.5% 18,464,000
Total, All Ages 17.3% 9.1% 26,502,000

Source: U.S. Census Bureau, The Supplemental Poverty Measure: 2020, Tables 6 and 7,

Technical Note

This analysis uses the Census Bureau’s official definition of poverty (except for the box, “Social Security’s Effect on Poverty Using the Supplemental Poverty Measure”). In determining poverty status, the Census Bureau compares a family’s cash income before taxes with poverty thresholds that vary by the size and age of the family. The poverty thresholds in 2020 were $12,413 for an elderly individual, $15,644 for an elderly couple, and $26,496 for an average family of four.[5] To calculate the anti-poverty effects of Social Security, we determined each family’s poverty status twice — first excluding and then including the family’s Social Security benefits.

Our analysis considers the non-institutionalized population using data from the Census Bureau’s Current Population Survey (CPS), the survey that is used to produce official poverty estimates.[6] Each March the CPS collects information on personal income, health coverage, and other social and economic characteristics for the previous year. The national estimates reported here are for 2020. The state-by-state estimates are based on a three-year average (2018 to 2020) to improve their reliability.

This analysis does not consider other changes that would occur in the absence of Social Security. If Social Security did not exist, many individuals aged 65 or older likely would have saved somewhat more and worked somewhat longer, and many might live with their adult children rather than in their own households. Other studies confirm, however, that Social Security has made a very large contribution to reducing poverty and that cutting Social Security benefits could substantially increase poverty among older adults.[7]


A 2017 study by Adam Bee and Joshua Mitchell of the Census Bureau matched the CPS survey responses used for official poverty statistics to administrative data from the Social Security Administration, Internal Revenue Service, and other government sources.[8] They found that official estimates overstate older adults’ reliance on Social Security and poverty rates among older adults, but confirmed that Social Security lifts millions of adults aged 65 and older above the poverty line and dramatically reduces the poverty rate among these adults. They did not find significant underreporting of income among people of working age. Further study has confirmed their findings using more recent data.[9]

Bee and Mitchell found that survey respondents generally reported accurately whether they received Social Security or earnings, but not whether they received pension income. Roughly half of older adult respondents who received pension income (either income from traditional defined-benefit pensions or withdrawals from defined-contribution pensions like 401(k)s or individual retirement accounts) failed to report this, particularly respondents whose pension income was small or inconsistent. On the other hand, respondents who reported receiving Social Security, earnings, or pension income generally reported accurately the amount they received from those sources.


Most retirees have modest incomes, save for some at the top of the income spectrum. Bee and Mitchell show that most low-income older adult households have very little pension income, if any; the majority of older adult households in the bottom third of the income distribution receive no pension income at all (compared to more than 80 percent in the top two-thirds). The study shows older adult households had a median income of about $44,000 in 2012 (compared to about $34,000 using the CPS alone, about a 30 percent difference). Further, about 1 in 4 retiree households live on less than $20,000, and the majority live on $50,000 or less (see Appendix Figure 1). Meanwhile, the wealthiest tenth of older adult households had incomes of $230,000, on average, according to Bee and Mitchell.

Bee and Mitchell’s study confirms Social Security’s large effect on poverty among older adults, but the enhanced data reduce both the poverty rate for these adults and the number of older adults lifted above the poverty line, compared to the official measures. The study estimates an elderly poverty rate in 2012 of nearly 7 percent, rather than the official rate of 9 percent. It also estimates that Social Security lifted about 3 in 10 older Americans — 10 million — out of poverty, about one-third lower than official estimates.

The study also confirms that Social Security remains the foundation of retirement income. It is the largest single source of income for older adults, providing the majority of income for half of retirees, and at least 90 percent of income for 18 percent of retirees. These rates of reliance are similar to Health and Retirement Survey and Survey of Income and Program Participation estimates.[10] However, they indicate significantly less reliance on Social Security than the CPS alone, which estimated that about 65 percent of older adults received at least half of their income from Social Security, and that 36 percent received at least 90 percent. Bee and Mitchell’s study also finds that Supplemental Security Income (SSI) plays a more important role in older adults’ income than official figures suggest, as many older adults with low incomes confuse SSI with Social Security.

Bee and Mitchell’s data extend only through 2012. Their findings were replicated by SSA researchers using 2016 data but cannot easily be extrapolated to future retirees. Trends strongly indicate that the composition and distribution of retirement income will change significantly. Bee and Mitchell found that roughly two-thirds of non-Social Security retirement income for current retirees came from traditional defined-benefit pensions, which have largely been replaced by defined-contribution plans in the private sector for today’s workers. Future retirees will be much less likely to have these pension benefits, and more of their retirement income will come from defined-contribution plans and individual retirement accounts, in which balances are highly unequal.

Effect of Social Security on Poverty Among Adults Aged 65 and Older by State, 2018-2020
  Percent in Poverty  
  Excluding Social Security Including Social Security Number Lifted Above the Poverty Line by Social Security
Alabama 47.1% 11.9% 297,000
Alaska 29.4% 6.4% 23,000
Arizona 39.0% 9.1% 393,000
Arkansas 49.1% 11.0% 202,000
California 35.3% 9.9% 1,505,000
Colorado 30.4% 7.4% 201,000
Connecticut 31.8% 8.2% 156,000
Delaware 31.9% 5.6% 53,000
District of Columbia 36.6% 19.4% 16,000
Florida 40.2% 10.5% 1,337,000
Georgia 41.8% 11.4% 455,000
Hawai’i 27.1% 8.5% 52,000
Idaho 36.5% 6.0% 84,000
Illinois 34.4% 8.7% 515,000
Indiana 36.2% 8.2% 304,000
Iowa 28.8% 6.2% 116,000
Kansas 33.1.% 5.7% 135,000
Kentucky 47.5% 13.4% 265,000
Louisiana 44.3% 14.1% 223,000
Maine 36.9% 6.8% 89,000
Maryland 28.7% 8.0% 193,000
Massachusetts 30.8% 7.4% 274,000
Michigan 36.9% 7.5% 526,000
Minnesota 29.6% 5.3% 226,000
Mississippi 51.7% 13.3% 180,000
Missouri 37.6% 7.7% 325,000
Montana 37.9% 8.1% 65,000
Nebraska 35.6% 6.4% 83,000
Nevada 35.0% 10.1% 129,000
New Hampshire 33.3% 4.7% 77,000
New Jersey 34.9% 8.6% 399,000
New Mexico 42.1% 11.0% 123,000
New York 36.0% 10.3% 868,000
North Carolina 43.5% 9.8% 601,000
North Dakota 36.4% 8.3% 34,000
Ohio 40.1% 9.3% 636,000
Oklahoma 43.5% 9.9% 217,000
Oregon 32.0% 7.3% 199,000
Pennsylvania 35.7% 7.6% 653,000
Puerto Rico* 70.0% 39.1% 204,000
Rhode Island 34.2% 8.1% 50,000
South Carolina 39.0% 8.7% 293,000
South Dakota 30.8% 7.3% 34,000
Tennessee 43.8% 9.4% 394,000
Texas 39.3% 11.2% 1,071,000
Utah 29.2% 5.7% 94,000
Vermont 30.9% 6.6% 33,000
Virginia 31.8% 7.3% 333,000
Washington 28.8% 6.0% 253,000
West Virginia 45.9% 10.3% 131,000
Wisconsin 33.5% 6.6% 264,000
Wyoming 33.6% 8.0% 26,000
Total, Persons Aged 65+ 37.1% 9.2% 15,203,000

Notes: Income is family cash income. The poverty rate “including Social Security” is the official poverty rate.

* Totals do not include Social Security beneficiaries who live in Puerto Rico, other U.S. Territories, or abroad because the CPS does not collect data for them. Estimates for Puerto Rico are based on the Puerto Rico Community Survey.

Source: CBPP analysis of data from the U.S. Census Bureau’s 2019-2021 March Current Population Survey (CPS) and 2017-2019 Puerto Rico Community Survey

Social Security Beneficiaries by State or Other Area and Age, 2020
  Total Aged 65 and Older Aged 18-64 Children Under 18
Alabama 1,165,990 801,448 297,729 66,813
Alaska 107,982 82,072 18,892 7,018
American Samoa 6,099 2,886 1,941 1,272
Arizona 1,433,237 1,139,530 235,761 57,946
Arkansas 707,846 488,776 178,035 41,035
California 6,150,009 4,959,887 970,652 219,470
Colorado 915,854 740,504 140,909 34,441
Connecticut 695,402 558,585 112,353 24,464
Delaware 224,617 176,977 39,421 8,219
District of Columbia 83,647 63,228 16,820 3,599
Florida 4,840,275 3,829,655 828,068 182,552
Georgia 1,902,790 1,390,094 410,893 101,803
Guam 19,032 13,560 3,535 1,937
Hawai’i 282,623 236,593 35,977 10,053
Idaho 370,385 287,118 67,485 15,782
Illinois 2,274,372 1,717,910 408,787 87,675
Indiana 1,382,024 1,029,957 387,634 64,433
Iowa 663,803 525,145 115,632 23,026
Kansas 569,120 441,366 103,022 24,732
Kentucky 1,009,092 692,742 260,769 55,581
Louisiana 925,400 651,315 218,872 55,213
Maine 355,433 267,617 73,972 13,844
Maryland 1,032,078 811,143 176,765 44,170
Massachusetts 1,294,623 1,001,823 240,482 52,318
Michigan 2,250,141 1,668,167 485,905 96,069
Minnesota 1,069,913 854,766 177,735 37,412
Mississippi 681,219 463,268 174,724 43,227
Missouri 1,323,195 972,912 288,732 61,551
Montana 244,937 194,008 41,528 9,401
Nebraska 357,164 284,411 58,526 14,227
Nevada 565,671 444,107 97,997 23,567
New Hampshire 317,389 242,497 60,931 13,961
New Jersey 1,651,408 1,319,660 268,714 63,034
New Mexico 453,282 340,372 90,563 22,347
New York 3,680,264 2,854,053 684,366 141,845
North Carolina 2,183,353 1,633,792 455,369 94,192
North Dakota 138,461 111,002 22,106 5,353
N. Mariana Islands 3,310 2,031 797 482
Ohio 2,405,217 1,816,744 487,093 101,380
Oklahoma 811,064 592,421 175,832 42,811
Oregon 906,127 725,767 152,550 27,810
Pennsylvania 2,877,728 2,215,691 551,730 110,307
Puerto Rico 824,365 583,353 208,033 32,979
Rhode Island 230,018 174,474 46,140 9,404
South Carolina 1,197,138 892,440 250,214 54,484
South Dakota 185,752 149,831 28,983 6,938
Tennessee 1,496,750 1,084,371 336,236 76,143
Texas 4,421,803 3,366,564 828,453 226,786
U.S. Virgin Islands 21,922 17,874 3,055 993
Utah 430,247 335,726 70,879 23,642
Vermont 156,005 120,760 29,331 5,914
Virginia 1,585,194 1,224,124 295,136 65,934
Washington 1,401,525 1,117,941 235,604 47,980
West Virginia 479,303 338,918 116,401 23,984
Wisconsin 1,275,932 988,817 241,401 45,714
Wyoming 118,420 92,823 20,768 4,829
Total 64,850,867 49,826,655 12,269,505 2,754,707

Note: Totals include residents of territories and people in the U.S. who reside abroad.

Source: Social Security Administration, Annual Statistical Supplement to the Social Security Bulletin, 2021, Table 5.J5,

End Notes

[1] Center on Budget and Policy Priorities, “Policy Basics: Top Ten Facts About Social Security,” updated March 4, 2022,

[2] CBPP analysis of data from the U.S. Census Bureau’s March 2021 Current Population Survey.

[3] Social Security Administration, “Annual Statistical Supplement to the Social Security Bulletin,” 2021, Table 5.F4,

[4] CBPP analysis of data from the U.S. Census Bureau’s March 2021 Current Population Survey.

[5] Poverty thresholds depend on the size of the family and its members’ ages; the $26,496 figure is a weighted average for families of four. For more information, see

[6] U.S. Census Bureau, “Income, Poverty, and Health Insurance Coverage in the United States: 2020,” September 14, 2021,

[7] Eugene Smolensky, Sheldon Danziger, and Peter Gottschalk, “The Declining Significance of Age in the United States: Trends in the Well-Being of Children and the Elderly Since 1939,” in John L. Palmer, Timothy Smeeding, and Barbara Boyle Torrey, eds., The Vulnerable, Urban Institute, 1988; Gary V. Engelhardt and Jonathan Gruber, “Social Security and the Evolution of Elderly Poverty,” National Bureau of Economic Research Working Paper 10466, May 2004.

[8] Adam Bee and Joshua Mitchell, “Do Older Americans Have More Income Than We Think?” U.S. Census Bureau, SESHD Working Paper No. 2017-39, July 2017,

[9] Irena Dushi and Brad Trenkamp, “Improving the Measurement of Retirement Income of the Aged Population,” Social Security Administration, ORES Working Paper No. 116, January 2021,

[10] Irina Dushi, Howard M. Iams, and Brad Trenkamp, “The Importance of Social Security Benefits to the Income of the Aged Population,” Social Security Bulletin, 2017,