African Americans: Description of Social Security and Supplemental Security Income Participation and Benefit Levels Using the American Community Survey

by Patricia P. Martin and John L. Murphy
Research and Statistics Note No. 2014-01 (released January 2014)

Patricia Martin and John Murphy are with the Office of Retirement Policy, Office of Retirement and Disability Policy, Social Security Administration.

Acknowledgments: The authors thank Dave Shoffner, Barbara Smith, Christopher Tamborini, and Kevin Whitman for their valuable suggestions and comments.

The findings and conclusions presented in this note are those of the authors and do not necessarily represent the views of the Social Security Administration.


Selected Abbreviations
ACS American Community Survey
DI Disability Insurance
OASDI Old-Age, Survivors, and Disability Insurance
PUMS Public Use Microdata Sample
SSA Social Security Administration
SSI Supplemental Security Income

African Americans encounter significant economic disadvantages, making them a critical focus for social insurance programs. Examining how the African American population uses Old-Age, Survivors, and Disability Insurance (OASDI, or Social Security) benefits and Supplemental Security Income (SSI) payments clarifies the role these programs play in supporting at-risk populations.

Earlier research has explored various facets of the relationship between Social Security and African Americans. For instance, many studies investigate African Americans' low retirement benefit receipt rates relative to whites (Abbott 1977, 1980; Thompson 1975; Huntley 1979; Parsons 1980; Gibson 1987, 1991, 1994; Farley 1988; Hayward, Friedman, and Chen 1996; O'Rand 1996; Gendell and Siegel 1996; Choi 1997; Hendley and Bilimoria 1999; Gustman and Steinmeier 2004; Bridges and Choudhury 2007, 2009; Favreault 2010). Others examine the prominent role of children's benefits for African Americans (Newcomb 2003/2004; Tamborini, Cupito, and Shoffner 2011). This analysis contributes to that body of research by using a relatively new, publicly available, and comprehensive data source, the American Community Survey (ACS), to document the demographic and economic characteristics of African American OASDI beneficiaries and SSI recipients. It is designed to lay the groundwork for future detailed analyses of how African Americans interact with Social Security and related programs.

In this note, we first discuss the strengths of the ACS and the methodology of this analysis. Next, we present the demographic and economic characteristics of the African American population in the 2009 ACS. Then, we present ACS data on OASDI and SSI participation and benefit levels, comparing African American participants with overall participants in three age distributions: the full age range for which benefit statistics are available in the ACS (15 or older), working age (18–61), and retirement age (62 or older).

Data and Methodology

The Census Bureau developed the ACS to provide data previously collected on the long form of the decennial census.1 Researchers can acquire detailed ACS data on income, race and ethnicity, and other economic and demographic variables through the survey's Public Use Microdata Sample (PUMS), a subset of the full ACS sample. For 2009, the ACS contains data for about 3 million households (Census Bureau 2013).2

The strength of the ACS is its breadth of coverage, which allows for more detailed and statistically significant analyses of population subsets such as minority groups. The ACS sample is substantially larger than those of other large household surveys, even such high-quality surveys as the Current Population Survey and the Survey of Income and Program Participation.3 Furthermore, because ACS response is mandatory, its response rate is higher than that for other surveys.4 Additionally, starting in 2006, the ACS sampling frame included both institutional and noninstitutional group quarters such as college dormitories, prisons, barracks, shelters, and nursing homes. Although less than 3 percent of the total population currently resides in group quarters, that population subset may be particularly relevant to the current analysis. One-half of group quarters residents live in institutional settings such as correctional facilities and nursing homes that provide formally supervised custody or care to inmates or patients. For example, the share of working-age people with disabilities residing in institutional group quarters is disproportionately African American (Stapleton, Honeycutt, and Schechter 2012). By including institutionalized disabled persons, the ACS captures more data for black OASDI beneficiaries and SSI recipients than the other household surveys. Additionally, African Americans are overrepresented among the population in correctional facilities, and incarceration is particularly relevant to certain subgroups in the African American population.5

Identifying new sources of data on OASDI and SSI receipt among minority groups is also important because the Social Security Administration (SSA) no longer provides detail by race for program data from administrative records. In SSA's Annual Statistical Supplement to the Social Security Bulletin, program data by race are not available for OASDI after 2009 or for SSI after 2002 because the quality of those data had become inconsistent.6 Although the ACS can help bridge this data gap, some limitations, summarized below, affect our analysis.

This study uses the PUMS of the 2009 ACS. All tables show descriptive statistics of the population, or subsets therein. OASDI benefit and SSI payment receipt are indicated in the ACS if the respondent reports any positive income from either or both of those programs in the past 12 months; however, PUMS data do not distinguish between OASDI retirement, survivor, and disability benefits. All benefit statistics are self-reported. All earnings, OASDI income, and SSI payments are in 2009 dollars.

There are three main limitations to the data. First, the accuracy of self-reported OASDI and SSI receipt has not been assessed.7 Some recipients may confuse SSI payments with OASDI benefits or fail to report payments, and in any event, PUMS data do not distinguish among OASDI benefit types. Second, our analysis of program participation among children is restricted to those aged 15–17 because the ACS, like the Current Population Survey, does not collect data on OASDI and SSI beneficiaries younger than age 15. Finally, because of limitations in survey questions used to determine disability, Burkhauser and others (2012) suggest that the 2009 ACS data fail to identify about one-third of the population aged 25–61 receiving either Social Security Disability Insurance (DI) benefits or SSI disability payments. As we discuss later, the “missing” respondents include those who report a work-activity limitation-based disability only.

In our analysis, “African American” refers to respondents who indicated they were not Hispanic on the ethnicity question and who selected “Black, African American, or Negro” on the racial question. Unless otherwise indicated, all comparisons between African Americans and the overall population, including OASDI beneficiaries and SSI recipients, are statistically significant at the 99 percent confidence level or better.

Demographic and Economic Characteristics in the ACS

Table 1 presents descriptive statistics for the African American and total US populations. It includes breakdowns by sex, age, marital status, educational attainment, disability status, income relative to poverty level, and annual earnings. We highlight some comparisons between African Americans and the total population, but do not discuss each of the breakdowns individually. Because the statistics for the total population include those for African Americans, the differences between the African American population and the remainder of the US population would exceed those reflected in Table 1.

Table 1. Characteristics of African American and total US populations, 2009
Characteristic African Americans (non-Hispanic) Total US population (including African Americans) African Americans as a percentage of total population
Number Percent Number Percent
Total 37,178,119 100.0 307,006,556 100.0 12.1
Male 17,709,631 47.6 151,373,350 49.3 11.7
Female 19,468,488 52.4 155,633,206 50.7 12.5
14 or younger 8,337,765 22.4 61,856,045 20.1 13.5
15–17 1,881,283 5.1 12,616,206 4.1 14.9
18–24 4,422,035 11.9 30,535,943 9.9 14.5
25–61 18,452,296 49.6 153,548,701 50.0 12.0
62–74 2,694,462 7.2 29,778,517 9.7 9.0
75–84 1,006,955 2.7 13,166,384 4.3 7.6
85 or older 383,323 1.0 5,504,760 1.8 7.0
Marital status a  
Married 8,501,133 29.5 120,802,465 49.3 7.0
Widowed 1,727,101 6.0 14,968,097 6.1 11.5
Divorced 3,374,788 11.7 26,023,175 10.6 13.0
Separated 1,293,557 4.5 5,399,414 2.2 24.0
Never married 13,943,775 48.3 77,957,360 31.8 13.4
Educational attainment b  
Less than high school diploma 4,155,344 18.4 29,789,530 14.7 13.9
High school graduate 7,119,700 31.6 57,515,438 28.5 12.4
Some college but no degree 5,626,667 25.0 43,149,458 21.4 13.0
Associate's degree 1,642,387 7.3 15,169,433 7.5 10.8
Bachelor's degree or higher 3,992,938 17.7 56,374,503 27.9 7.1
Disability status c  
Disabled 5,358,302 14.4 38,644,826 12.6 13.9
Nondisabled 31,819,817 85.6 268,361,730 87.4 11.9
Income relative to poverty level d  
Below 100% 9,128,370 25.7 42,743,551 14.3 21.4
100% to 124% 2,273,138 6.4 13,543,406 4.5 16.8
125% to 149% 2,218,563 6.2 13,996,115 4.7 15.9
150% or above 21,927,499 61.7 228,720,630 76.5 9.6
Annual earnings e  
$1–$18,320 7,006,085 39.6 56,042,665 34.6 12.5
$18,321–$40,712 6,320,458 35.7 51,435,419 31.7 12.3
$40,713–$65,139 2,814,554 15.9 29,151,412 18.0 9.7
$65,140–$106,799 1,224,188 6.9 16,859,732 10.4 7.3
$106,800 or more 331,261 1.9 8,692,313 5.4 3.8
NOTE: Rounded components of percentage distributions do not necessarily sum to 100.0.
a. Individuals aged 15 or older. Omits individuals with marital status not specified.
b. Individuals aged 25 or older.
c. ACS defines disability more broadly than SSA because Social Security pays benefits only for total disability. See note 12 for details.
d. Individuals aged 15 or older. Excludes respondents with missing values.
e. Individuals aged 16 or older with positive earnings. Excludes respondents with missing values.


More than 22 percent of African Americans were younger than age 15 in 2009, compared with about 20 percent of the total population, and about 40 percent of African Americans were younger than age 25, compared with about 34 percent of the total population. In 2009, the median age for African Americans was 31.9 years, compared with 36.8 years for the total population, a 5-year gap (not shown). Significant differences also appear in older ages: 10.9 percent of African Americans were aged 62 (the youngest age at which an individual is eligible for Social Security retirement benefits) or older, compared with 15.8 percent of the total population.

Marital Status

African Americans tend to be younger, and they are less likely to be married—only 29.5 percent of African Americans were married, compared with 49.3 percent of the total population.8 The gap in marital rates suggests that African Americans are less likely to qualify for the Social Security spouse and survivor benefits that could otherwise enhance their economic well-being in old age (Meyer, Wolf, and Himes 2004; Tamborini, Iams, and Whitman 2009; Butrica and Smith 2012). Typically, an aged person is eligible for spousal or widow benefits if he or she is currently married, widowed, or divorced after a marriage that lasted 10 years or more.9 Spousal or widow benefits may support an individual whose earnings history is insufficient to qualify for his or her own benefits, or may supplement benefits based on one's own earnings record. Research has also documented the importance of marital status to adult well-being beyond its effect on Social Security (Waite and Gallagher 2000; Lillard and Panis 1996; Wilmoth and Koso 2002).10


The educational attainment patterns of African Americans and the general population also differ: 18.4 percent of African Americans aged 25 or older did not have a high school diploma in 2009, compared with 14.7 percent of the total population. In addition, only 17.7 percent of African Americans had a bachelor's degree or higher, compared with 27.9 percent of the overall population. Educational attainment has important implications for lifetime earnings and labor market patterns. Higher education is related, for example, to greater lifetime earnings, lower unemployment rates, and higher income taxes (Bureau of Labor Statistics 2010; Levin and others 2006). Using Social Security administrative data linked with National Health Interview Survey data for disabled adults aged 25–62, Kirk (2012) finds that high-school dropouts comprise over one-quarter of DI beneficiaries and over one-half of SSI recipients.11

Disability, Health, and Mortality

Disability has an obvious impact on participation in Social Security. Using the ACS definition of disability, African Americans were more likely to be disabled than were individuals in the total population, 14.4 percent versus 12.6 percent.12 These findings are consistent with other research showing that African Americans experience a higher incidence of chronic diseases that can lead to disability. For example, although the leading causes of death for the US population—namely, heart disease (25 percent) and cancer (23 percent)—are similar regardless of race and ethnicity, African Americans have higher rates of diabetes, hypertension, and obesity.13

Differences in mortality reflect some of these health disparities. Although the gap in life expectancy between African Americans and whites has narrowed since 1990, it still persists. As of 2007, that gap was 6 years for men and 4 years for women (National Center for Health Statistics 2011).

It is worth emphasizing that the current ACS undercounts DI and SSI disability program participation. Burkhauser and others (2012) suggest that the 2009 ACS data miss about one-third of the population aged 25–61 receiving DI benefits or SSI disability payments. Since the Census Bureau introduced a new six-question sequence on disability for the 2009 ACS, the data have understated the size of the population with disabilities because none of those questions ask about work-activity limitations. The ACS also underrepresents DI/SSI beneficiaries in certain diagnostic groups such as intellectual disability.


African Americans had lower earnings in 2009 than the overall population. Approximately 40 percent of African American workers aged 16 or older were low earners—earning $18,320 or less annually—compared with about 35 percent of the total population of workers.14 Table 2 shows that average or mean annual earnings were also lower for African Americans ($30,764) than for the overall population ($40,360). Annual mean earnings were consistently lower for African Americans than for the total population, and a higher percentage of African Americans was below the poverty level, regardless of age (Charts 1 and 2). As Social Security benefits are based on lifetime earnings in covered employment, lower earnings can lead to lower benefits. Note, however, that Social Security's benefit formula provides higher replacement rates (benefits as a percentage of preretirement earnings) for people with low lifetime earnings. It also helps people with partial careers—at least 10 years of work history or 40 quarters of coverage accrued—to be eligible for retirement benefits. Thus, although benefits received by African Americans tend to be lower than those of the overall population, they replace a higher percentage of their preretirement earnings.

Table 2. Selected economic characteristics of African American and total US populations, 2009
Characteristic African Americans (non-Hispanic) Total US population (including African Americans)
Annual earnings ($)
Mean 30,764 40,360
25th percentile 10,800 12,000
Median 24,000 28,900
75th percentile 40,000 50,000
Annual mean earnings, by age ($)
16–24 10,893 11,872
25–34 27,676 34,820
35–44 37,608 48,551
45–54 38,967 52,353
55–64 37,901 50,438
65 or older 25,839 34,804
Poverty rate, by age (%)
15 or younger 36.4 20.0
16–24 27.5 20.2
25–34 22.4 14.2
35–44 18.1 11.3
45–54 18.2 9.7
55–64 17.8 9.1
65 or older 17.9 9.1
NOTE: Earnings data apply only to individuals aged 16 or older. Earnings percentile calculations consider only individuals with positive earnings.
Chart 1.
Annual mean earnings for African American and total US populations, by age, 2009 (in dollars)
Bar chart linked to data in table format, which is provided as part of Table 2.
NOTE: Total US population includes African Americans.
Chart 2.
Poverty rates for African American and total US populations, by age, 2009 (in percent)
Bar chart linked to data in table format, which is provided as part of Table 2.
NOTE: Total US population includes African Americans.


Unemployment is closely tied to educational attainment. Among adults in the labor force, a higher percentage of dropouts are unemployed than are high school completers (Department of Labor 2012). In 2010, African Americans aged 25 or older without a high school diploma had a 22.5 percent unemployment rate, almost three times higher than the 7.9 percent unemployment rate for those with a bachelor's degree or higher (Table 3 and Chart 3). Whites without a high school diploma had a 13.9 percent unemployment rate, compared with 4.3 percent for those with a bachelor's degree or higher.

Table 3. Unemployment rates for African American and white adults aged 25 or older, by educational attainment, 2010 (in percent)
Education African American White
Less than high school diploma 22.5 13.9
High school graduate 15.8 9.5
Some college or associate's degree 12.4 7.6
Bachelor's degree or higher 7.9 4.3
SOURCE: Bureau of Labor Statistics, Current Population Survey.
NOTE: Racial categories comprise respondents who selected only one racial group, and may include individuals of Hispanic origin.
Chart 3.
Unemployment rates for African American and white adults aged 25 or older, by educational attainment, 2010 (in percent)
Bar chart linked to data in table format, which is provided in Table 3.
SOURCE: Bureau of Labor Statistics, Current Population Survey.
NOTE: Racial categories comprise respondents who selected only one racial group, and may include individuals of Hispanic origin.


Incarceration is a factor often overlooked in research on unemployment inequality. Unlike the ACS, many surveys exclude institutionalized people (such as those in correctional facilities). That population is increasingly relevant, as incarceration rates have reached record levels (Thomas and Ryan 2008). Table 4 shows that for the 2005–2009 period, African Americans comprised the largest nonwhite share of persons in adult correctional facilities (41.1 percent), in contrast with their 12.6 percent share of the US population. African American men were incarcerated at a rate more than six times higher than white non-Hispanic men in 2009 (West 2010). Incarceration may offer a partial explanation of the relatively low earnings among some African Americans.15,16 Relative to those who have never been incarcerated, previously incarcerated men have a 30–40 percent reduction in earnings, reduced job tenure, reduced hourly wages, and higher unemployment (Western and Pettit 2010). However, previously incarcerated African Americans have even lower earnings than other previously incarcerated men (Hutcherson 2009; Joseph 2010; Wakefield 2010; Western and Wildeman 2009).

Table 4. Population residing in group quarters, by race: Annual average 2005–2009 (percentage distribution)
Race a Total population Population residing in group quarters Population in adult correctional facilities
White 76.1 69.3 47.7
Black or African American 12.6 22.0 41.1
American Indian and Alaska Native 0.8 1.1 2.0
Asian 4.5 3.2 0.9
Native Hawaiian and other Pacific Islander 0.2 0.2 0.2
Other single race 5.8 4.3 8.2
Population size 294,792,853 8,009,634 2,058,113
SOURCE: Census Bureau (n.d. a).
NOTES: "Group quarters" category excludes barracks and shelters.
Rounded components of percentage distributions do not necessarily sum to 100.0.
a. Omits individuals of two or more races.

African American and All OASDI Beneficiaries and SSI Recipients

This section presents selected characteristics of African American OASDI beneficiaries and SSI recipients compared with those of all OASDI beneficiaries and SSI recipients. The tables, based on the 2009 PUMS, show estimates for the full ACS beneficiary-data age range (15 or older), the working-age group (18–61), and the retirement-age group (62 or older). Data are tabulated by sex, age, marital status, education, disability status, and income relative to poverty level. As before, we focus our discussion on notable differences between the African American and total beneficiary populations.

Social Security

In 2009, African Americans comprised almost one-tenth (9.5 percent) of all OASDI beneficiaries aged 15 or older, over one-sixth (17.9 percent) of working-age OASDI beneficiaries, and almost one-twelfth (8.1 percent) of retirement-age OASDI beneficiaries (Table 5). The mean annual Social Security benefit in 2009 for retirement-age African American beneficiaries was about $10,430, compared with about $11,910 for all retirement-age beneficiaries (Table 6). Benefits for African Americans were also lower than those for the total population at the 25th, 50th (median), and 75th percentiles. Similar benefit differences appeared in the full and working age ranges.

Table 5. Characteristics of African American and all Social Security beneficiaries, by age group, 2009
Characteristic African American beneficiaries (non-Hispanic) All beneficiaries (including African Americans) African Americans as a percentage of all beneficiaries
Number Percent Number Percent
  Ages 15 or older
Total 4,245,247 100.0 44,872,872 100.0 9.5
Male 1,748,749 41.2 19,608,072 43.7 8.9
Female 2,496,498 58.8 25,264,800 56.3 9.9
15–17 70,174 1.7 360,114 0.8 19.5
18–24 76,152 1.8 339,038 0.8 22.5
25–61 973,366 22.9 5,526,891 12.3 17.6
62–74 1,914,188 45.1 21,699,117 48.4 8.8
75–84 878,909 20.7 11,971,488 26.7 7.3
85 or older 332,458 7.8 4,976,224 11.1 6.7
Marital status  
Married 1,409,370 33.2 23,447,631 52.3 6.0
Widowed 1,150,130 27.1 11,338,922 25.3 10.1
Divorced 701,035 16.5 5,460,736 12.2 12.8
Separated 206,806 4.9 690,889 1.5 29.9
Never married 777,906 18.3 3,934,694 8.8 19.8
Educational attainment a  
Less than high school diploma 1,451,799 35.4 9,939,176 22.5 14.6
High school graduate 1,314,215 32.1 15,636,652 35.4 8.4
Some college but no degree 729,447 17.8 8,460,733 19.2 8.6
Associate's degree 174,966 4.3 2,025,873 4.6 8.6
Bachelor's degree or higher 428,494 10.5 8,111,286 18.4 5.3
Disability status b  
Disabled 2,078,134 49.0 18,640,861 41.5 11.1
Nondisabled 2,167,113 51.0 26,232,011 58.5 8.3
Income relative to poverty level c  
Below 100% 911,417 22.3 4,576,275 10.6 19.9
100% to 124% 372,461 9.1 2,543,064 5.7 14.6
125% to 149% 337,439 8.3 2,645,418 6.1 12.8
150% or above 2,460,672 60.3 33,608,374 77.5 7.3
  Working age (18–61)
Total 1,049,518 100.0 5,865,929 100.0 17.9
Men 475,571 45.3 2,827,293 48.2 16.8
Women 573,947 54.7 3,038,636 51.8 18.9
18–24 76,152 7.3 339,038 5.8 22.5
25–61 973,366 92.7 5,526,891 94.2 17.6
Marital status  
Married 255,973 24.4 2,078,798 35.4 12.3
Widowed 84,663 8.1 473,841 8.1 17.9
Divorced 179,535 17.1 1,164,034 19.8 15.4
Separated 74,801 7.1 250,720 4.3 29.8
Never married 454,546 43.3 1,898,536 32.4 23.9
Educational attainment a  
Less than high school diploma 265,840 27.3 1,335,669 24.2 19.9
High school graduate 349,088 35.9 1,995,948 36.1 17.5
Some college but no degree 239,280 24.6 1,245,051 22.5 19.2
Associate's degree 52,513 5.4 369,137 6.7 14.2
Bachelor's degree or higher 66,645 6.8 581,086 10.5 11.5
Disability status b  
Disabled 644,281 61.4 3,733,155 63.6 17.3
Nondisabled 405,237 38.6 2,132,774 36.4 19.0
Income relative to poverty level c  
Below 100% 358,522 35.2 1,426,353 25.0 25.1
100% to 124% 105,779 10.4 533,693 9.3 19.8
125% to 149% 88,402 8.7 455,128 8.0 19.4
150% or above 466,374 45.8 3,298,978 57.7 14.1
  Retirement age (62 or older)
Total 3,125,555 100.0 38,646,829 100.0 8.1
Men 1,238,337 39.6 16,599,587 43.0 7.5
Women 1,887,218 60.4 22,047,242 57.0 8.6
62–74 1,914,188 61.2 21,699,117 56.1 8.8
75–84 878,909 28.1 11,971,488 31.0 7.3
85 or older 332,458 10.6 4,976,224 12.9 6.7
Marital status  
Married 1,153,010 36.9 21,366,997 55.3 5.4
Widowed 1,065,260 34.1 10,864,716 28.1 9.8
Divorced 521,500 16.7 4,296,415 11.1 12.1
Separated 131,836 4.2 439,885 1.1 30.0
Never married 253,949 8.1 1,678,816 4.3 15.1
Educational attainment  
Less than high school diploma 1,185,959 37.9 8,603,507 22.3 13.8
High school graduate 965,127 30.9 13,640,704 35.3 7.1
Some college but no degree 490,167 15.7 7,215,682 18.7 6.8
Associate's degree 122,453 3.9 1,656,736 4.3 7.4
Bachelor's degree or higher 361,849 11.6 7,530,200 19.5 4.8
Disability status b  
Disabled 1,422,091 45.1 14,858,480 38.4 9.6
Nondisabled 1,703,464 54.5 23,788,349 61.6 7.2
Income relative to poverty level c  
Below 100% 535,011 17.9 3,086,551 8.3 17.3
100% to 124% 259,155 8.7 1,981,165 5.1 13.1
125% to 149% 242,029 8.1 2,162,716 5.8 11.2
150% or above 1,957,561 65.4 30,072,804 80.6 6.5
NOTE: Rounded components of percentage distributions do not necessarily sum to 100.0.
a. Individuals aged 25 or older.
b. ACS defines disability more broadly than SSA because Social Security pays benefits only for total disability. See note 12 for details.
c. Excludes respondents with missing values.
Table 6. Annual Social Security benefit amounts for African American and all beneficiaries, by age group, 2009 (in dollars)
Amount African American beneficiaries (non-Hispanic) All beneficiaries (including African Americans)
  Aged 15 or older
Mean 9,974 11,518
25th percentile 6,500 7,200
Median 9,600 11,100
75th percentile 13,000 15,100
  Working age (18–61)
Mean 8,893 10,299
25th percentile 5,400 6,000
Median 8,300 9,500
75th percentile 11,800 13,300
  Retirement age (62 or older)
Mean 10,430 11,910
25th percentile 7,000 7,600
Median 10,000 11,900
75th percentile 13,400 15,600
NOTE: Earnings data apply only to individuals aged 16 or older. Earnings percentile calculations consider only individuals with positive earnings.

African American beneficiaries tend to be younger than beneficiaries overall. Almost 23 percent of African American beneficiaries were in the 25–61 age range, compared with 12.3 percent of all Social Security beneficiaries (Table 5). The proportion of African American beneficiaries younger than age 62 nearly doubled that of beneficiaries overall. This higher proportion of beneficiaries younger than age 62 may be partly explained by much higher proportions of disabled working-age African Americans and African American children (aged 15–17) receiving Social Security disability and survivor benefits. For example, almost one-fifth (19.5 percent) of minor children aged 15–17 receiving Social Security benefits were African American. In addition, about 17 percent of African American children aged 15–17 reported both having a disability and being a Social Security beneficiary; overall, less than 14 percent of that age group reported both (not shown).17

About one-third (35.4 percent) of African American beneficiaries aged 25 or older never completed high school, compared with less than one-fourth (22.5 percent) of beneficiaries overall. The proportion of African American beneficiaries that completed a bachelor's degree (10.5 percent) was about half that of the overall group (18.4 percent). However, the education gap may be shrinking for younger blacks. For beneficiaries aged 25–61, more than one-fourth (27.3 percent) of African Americans never completed high school, a proportion comparable with the 24.2 percent of beneficiaries overall. By contrast, for those aged 62 or older, more than one-third (37.9 percent) of African American beneficiaries never completed high school, a much wider gap from the 22.3 percent of beneficiaries overall.

At ages 62 or older, African American beneficiaries were more than twice as likely as all beneficiaries to have incomes below the federal poverty level (17.9 percent versus 8.3 percent). They were also almost twice as likely as all beneficiaries to have income below 125 percent of the poverty level and 1.8 times as likely to have income below 150 percent of the poverty level. This pattern was similar for those aged 15 or older. An even greater proportion of working-age African American beneficiaries had incomes below the poverty level (35.2 percent), although it did not double the proportion of the overall beneficiary population (25.0 percent).18


Individuals younger than age 65 are eligible for SSI if they are disabled or blind and meet the income and asset tests; when they reach age 65, they need to meet only the income and asset tests. To be consistent with the age ranges analyzed for OASDI beneficiaries, we use age 62 as the lower bound for our retirement-age group even though it differs from SSI's age-65 threshold.19

African Americans comprised over one-fifth (20.9 percent) of SSI recipients aged 15 or older, almost one-fourth (22.7 percent) of working-age SSI recipients, and roughly one-sixth (16.8 percent) of retirement-age SSI recipients (Table 7). Almost one-third (30.0 percent) of child SSI recipients aged 15–17 in 2009 were African American. In addition, about 94 percent of African American children aged 15–17 who reported being SSI recipients also reported being disabled, which was similar to the percentage for SSI recipients in that age group overall (not shown).20 Average annual payments received by the total SSI population were about 7 percent higher than those received by African Americans for both the total 15-or-older and working-age recipient populations (Table 8). For those aged 62 or older, average annual SSI payments received by all recipients were about 18 percent higher than those received by African Americans.

Table 7. Characteristics of African American and all SSI recipients, by age group, 2009
Characteristic African American recipients (non-Hispanic) All recipients (including African Americans) African Americans as a percentage of all recipients
Number Percent Number Percent
  Ages 15 or older
Total 1,112,190 100.0 5,325,239 100.0 20.9
Male 467,256 42.0 2,270,695 42.6 20.6
Female 644,934 58.0 3,054,544 57.4 21.1
15–17 32,334 2.9 107,921 2.0 30.0
18–24 68,037 6.1 267,467 5.0 25.4
25–61 709,790 63.8 3,153,961 59.2 22.5
62–74 182,946 16.4 973,624 18.3 18.8
75–84 80,199 7.2 540,772 10.2 14.8
85 or older 38,884 3.5 281,494 5.3 13.8
Marital status  
Married 150,130 13.5 1,210,869 22.7 12.4
Widowed 142,024 12.8 758,834 14.2 18.7
Divorced 182,662 16.4 1,061,613 19.9 17.2
Separated 77,099 6.9 278,047 5.2 27.7
Never married 560,275 50.4 2,015,876 37.9 27.8
Educational attainment a  
Less than high school diploma 459,248 45.4 2,075,230 41.9 22.1
High school graduate 336,472 33.3 1,618,883 32.7 20.8
Some college but no degree 151,575 15.0 753,494 15.2 20.1
Associate's degree 30,703 3.0 186,080 3.8 16.5
Bachelor's degree or higher 33,821 3.3 316,164 6.4 10.7
Disability status b  
Disabled 1,050,143 94.4 5,063,477 95.1 20.7
Nondisabled 62,047 5.6 261,762 4.9 23.7
Income relative to poverty level c  
Below 100% 504,118 47.5 2,004,752 39.2 25.1
100% to 124% 110,773 10.4 530,573 10.4 20.9
125% to 149% 90,568 8.5 399,045 7.8 22.7
150% or above 354,751 33.5 2,181,512 42.6 16.3
  Working age (18–61)
Total 777,827 100.0 3,421,428 100.0 22.7
Men 358,227 46.1 1,585,446 46.3 22.6
Women 419,600 53.9 1,835,982 53.7 22.9
18–24 68,037 8.7 267,467 7.8 25.4
25–61 709,790 91.3 3,153,961 92.2 22.5
Marital status  
Married 104,999 13.5 693,308 20.3 15.1
Widowed 31,269 4.0 135,676 4.0 23.0
Divorced 123,644 15.9 715,893 20.9 17.3
Separated 54,831 7.0 200,279 5.9 27.4
Never married 463,084 59.5 1,676,272 49.0 27.6
Educational attainment a  
Less than high school diploma 285,229 40.2 1,166,531 37.0 24.5
High school graduate 256,477 36.1 1,146,970 36.4 22.5
Some college but no degree 122,132 17.2 550,659 17.5 22.2
Associate's degree 23,690 3.3 130,882 4.1 18.1
Bachelor's degree or higher 22,262 3.1 158,919 5.0 14.0
Disability status b  
Disabled 743,349 95.6 3,298,002 96.4 22.5
Nondisabled 34,478 4.4 123,426 3.6 27.9
Income relative to poverty level c  
Below 100% 369,304 49.4 1,423,845 43.0 25.9
100% to 124% 79,201 10.6 345,130 10.4 22.9
125% to 149% 61,771 8.3 242,069 7.3 25.5
150% or above 237,539 31.8 1,303,325 39.3 18.2
  Retirement age (62 or older)
Total 302,029 100.0 1,795,890 100.0 16.8
Men 88,387 29.3 615,876 34.3 14.4
Women 213,642 70.7 1,180,014 65.7 18.1
62–74 182,946 60.6 973,624 54.2 18.8
75–84 80,199 26.6 540,772 30.1 14.8
85 or older 38,884 12.9 281,494 15.7 13.8
Marital status  
Married 45,131 14.9 517,309 28.8 8.7
Widowed 110,755 36.7 623,098 34.7 17.8
Divorced 59,018 19.5 345,454 19.2 17.1
Separated 22,268 7.4 77,650 4.3 28.7
Never married 64,857 21.5 232,379 12.9 27.9
Educational attainment  
Less than high school diploma 174,019 57.6 908,699 50.6 19.2
High school graduate 79,995 26.5 471,913 26.3 17.0
Some college but no degree 29,443 9.7 202,835 11.3 14.5
Associate's degree 7,013 2.3 55,198 3.1 12.7
Bachelor's degree or higher 11,559 3.8 157,245 8.8 7.4
Disability status b  
Disabled 276,375 91.5 1,665,414 92.7 16.6
Nondisabled 25,654 8.5 130,476 7.3 19.7
Income relative to poverty level c  
Below 100% 121,370 43.3 544,720 32.1 22.3
100% to 124% 26,273 9.4 172,312 10.2 15.2
125% to 149% 25,698 9.2 145,025 8.6 17.7
150% or above 107,226 38.2 833,643 49.2 12.9
NOTE: Rounded components of percentage distributions do not necessarily sum to 100.0.
a. Individuals aged 25 or older.
b. ACS defines disability more broadly than SSA because Social Security pays benefits only for total disability. See note 12 for details.
c. Excludes respondents with missing values.
Table 8. Annual SSI payment amounts for African American and all recipients, by age group, 2009 (in dollars)
Amount African American beneficiaries (non-Hispanic) All beneficiaries (including African Americans)
  Aged 15 or older
Mean 6,748 7,222
25th percentile 4,200 4,200
Median 7,200 7,200
75th percentile 8,100 8,700
  Working age (18–61)
Mean 7,038 7,520
25th percentile 4,700 4,700
Median 7,400 7,500
75th percentile 8,100 9,000
  Retirement age (62 or older)
Mean 6,007 7,097
25th percentile 3,200 3,600
Median 5,300 6,300
75th percentile 8,100 9,100
NOTE: Earnings data apply only to individuals aged 16 or older. Earnings percentile calculations consider only individuals with positive earnings.

Almost three-fourths of African American SSI recipients were younger than age 62 in 2009, compared with about two-thirds of all SSI recipients (Table 7). This younger African American SSI population may result from the higher disability rates reported among African Americans in general.

Slightly more than one in eight (13.5 percent) African American SSI recipients aged 15 or older were married, compared with nearly one in four (22.7 percent) among all SSI recipients. Half of African American SSI recipients never married (50.4 percent), compared with more than one-third (37.9 percent) of all SSI recipients. Among working-age SSI recipients, three-fifths (59.5 percent) of African Americans never married, compared with almost half (49.0 percent) of recipients overall. Among those aged 62 or older, about one-fifth (21.5 percent) of African American SSI recipients never married, substantially higher than the rate among SSI recipients overall (12.9 percent).

Among SSI recipients aged 15 or older, 78.7 percent of African Americans had no education beyond a high school diploma, compared with 74.6 percent of the general SSI population, a gap of 4.1 percentage points. Among those aged 18–61, the disparity is only 2.9 percentage points (76.3 percent versus 73.4 percent). Among those aged 62 or older, the education gap widens to 7.2 percentage points, 84.1 percent versus 76.9 percent.

African American SSI recipients had higher poverty rates than the overall SSI population (47.5 percent versus 39.2 percent), a gap of 8.3 percentage points, among those aged 15 or older. The working-age group had the greatest proportion in poverty (49.4 percent), compared with 43.0 percent of the general SSI population, a 6.4 percentage point gap. The gap for those aged 62 or older was the largest at 11.2 percentage points (43.3 percent for African Americans versus 32.1 percent overall). Among African American child SSI recipients aged 15–17, 41.5 percent were below the poverty threshold, versus 33.5 percent of SSI recipients in that age group overall (not shown).


This note uses the PUMS from the 2009 ACS to document the demographic, economic, and disability characteristics of the African American population. African Americans are, on average, younger, less likely to be married, and less educated. They have lower economic well-being, as seen in their lower earnings and in their higher poverty, unemployment, and incarceration rates. They suffer more health problems, as seen in higher disability and illness rates, which in turn lead to higher unemployment and lower earnings.

All of these demographic and socioeconomic characteristics affect the relationship between African Americans and OASDI and SSI. Lower earnings are most relevant to Social Security in that benefits are based on earnings. However, other characteristics can also affect benefit eligibility and income. For example, the higher never-married rate of African Americans reduces the share of their population eligible for the spouse and survivor benefits that could supplement the benefits their own work records provide.

Our results show that African Americans are less likely than the overall population to be OASDI beneficiaries and more likely to be SSI recipients. One in ten OASDI beneficiaries, and one in five SSI recipients, is African American. Higher rates of poverty, disability, and mortality among African Americans mean that those who are eligible are more likely to rely on Social Security survivor and disability benefits than are other Americans.

Future research could focus on the differences by race and ethnicity in the elderly population's reliance on Social Security income, in the use of auxiliary spouse and widow(er) benefits among retirees, and in the timing of claiming Social Security benefits. However, some of this work may not be possible with the ACS, given its limitations. The current analysis suggests that the ACS is a valuable source of information, but researchers need more robust, detailed data to better address OASDI and SSI receipt among traditionally vulnerable groups.


1 The ACS is a nationwide continuous survey. The Census Bureau distributes about 250,000 questionnaires every month. The ACS samples about 3 million addresses each year, resulting in nearly 2 million final interviews. The 1-, 3-, and 5-year survey data products are released to the public on a phased schedule the following year, generally in the order of shortest to longest survey cycle (although in 2009, the 5-year data were released prior to the 3-year data). The ACS is updated every year. By producing 1-, 3-, and 5-year estimates, the ACS reflects data collected over a period of time as opposed to “point-in-time” estimates, such as the decennial census, that approximate the characteristics of an area on a specific date.

2 See for information about the survey, including the questionnaire and data collection procedures.

3 For further details on how the ACS differs from other government surveys, see and

4 Navarro, King, and Starsinic (2011) compare ACS estimates based on mandatory versus voluntary responses.

5 Pettit and Sykes (2012) find that young African American male high-school dropouts are more likely to be incarcerated than they are to be working. The incarceration rate affects earnings levels, which are important in determining Social Security benefit amounts.

6 From 1935 to 1980, the Social Security number application (Form SS-5) allowed an applicant to select from only “white,” “black,” or “other” race categories; those who did not report their race were classified as “unknown.” SS-5 data using those codes were incorporated into SSA's Master Beneficiary Record. Then, from 1980 through August 2009, the SS-5 provided six race/ethnicity categories: white; black (not Hispanic); Hispanic; Asian, Asian-American, or Pacific Islander; American Indian or Alaskan Native; and unknown. In 1987, SSA began to enroll new participants at birth, extracting data from birth certificates rather than requiring applicants to file Form SS-5; however, SSA did not extract the race/ethnicity information because it was considered unnecessary for administering the program. Since 1989, the only persons filing a Form SS-5 have been those requesting a new number or a name change (Scott 1999). The changes to the racial category codes, the sharp decline in the number of people for whom any race/ethnicity data are collected, and other factors have combined to produce inconsistencies in SSA program data by race that render long-term statistical comparisons meaningless (SSA 2011).

7 See Sears and Rupp (2003) for a discussion of self-reporting and other issues related to the accuracy of survey data.

8 Although the Modeling Income in the Near Term model projects a white majority among future never-married retirees, a sizable percentage is expected to be African American. For 2020, the model also projects that the never-married will account for 21 percent of the African American population aged 62 or older, but only 9 percent of the entire population in that age group (Tamborini 2007).

9 A retired worker's spouse is entitled to a Social Security benefit equal to half the worker's primary insurance amount (PIA) while the worker is still alive, and 100 percent of the PIA after the worker's death.

10 Social welfare agencies use data by marital status to identify the number of children needing special services, such as children in single-parent households who fall below the poverty threshold. The Census Bureau typically categorizes families as either married couples, female householder with no husband present, and male householder with no wife present. Notably, in 2009, 44.3 percent of female-headed households with children younger than age 18 fell below the poverty threshold, compared with 11.0 percent of married-couple families (Census Bureau n.d. b).

11 In determining eligibility for DI benefits, SSA considers educational attainment as a factor that may affect a claimant's ability to “do any other type of work” (see step 5 at

12 Starting in 2008, the ACS classified individuals as disabled on the basis of whether they report difficulty with specific functions in one of six categories: hearing, vision, cognition, ambulation, self-care, and independent living. For children younger than age 5, hearing and vision difficulty are used to determine disability status. For children aged 5–14, disability status is determined from hearing, vision, cognitive, ambulatory, and self-care difficulties. People aged 15 or older are considered to have a disability if they have a difficulty with any one of the six functional categories. By contrast, Social Security pays benefits only for total disability, not for partial or short-term disability. If individuals are unable to do the work that they did before, cannot adjust to other work because of their medical condition, and possess an impairment that has lasted or is expected to last for at least 1 year or to result in death, they meet the SSA's stricter definition of disability.

13 Persons with diabetes are more prone to physical disability and deterioration. African Americans with diabetes are generally sicker than their white counterparts and have worse outcomes (Chin, Zhang, and Merrell 1998; Chiu and Wray 2011), which may result in higher disability rates (leading to DI benefits or SSI payments) or mortality (leading to survivor's benefits). African Americans also have higher rates of hypertension than the general population has (Collins and Winkleby 2002; Quereshi and others 2005; Wright and others 2002). Like diabetes, hypertension leads to increased morbidity and mortality rates (Kannel 1996). Additionally, African Americans are adversely affected by high obesity rates (Anderson and Whitaker 2009; Cossrow and Falkner 2004; Whaley, Smith, and Hancock 2011).

14 In 2001, SSA's Office of the Chief Actuary established guidelines for a hypothetical earnings pattern. Starting with a national average wage estimated at $40,711.61 for 2009, persons earning 45 percent of the average wage ($18,320.22) or less are considered low earners, and those earning 160 percent of average wage ($65,138.58) or more are high earners (Nichols, Clingman, and Glanz 2001). We use those figures and the maximum taxable earnings under Social Security ($106,800 in 2009) to define the earnings-level cutoffs.

15 In 2000, roughly 27 percent of African American male high school dropouts aged 18–55 were in prison or in jail—nearly as high a proportion as the 33 percent who were employed (Raphael 2008). As of 1997, over 40 percent of the prison population had not completed high school (Harlow 2003). Given the explosive growth of this population, the exclusion of prisoners from survey data can produce biased research. ACS data enable new research that compares educational attainment among prisoners and the general population across subpopulations, including by age and by race and ethnic group (Ewert and Wildhagen 2011).

16 OASDI and SSI benefits are suspended when individuals are jailed or imprisoned, and they remain suspended until the inmate is released. For SSI recipients, payments are terminated when the individual is incarcerated for a year or more. Using a special prerelease procedure, institutions can enable SSA to begin processing inmates' claims prior to their release, so that benefits can resume or begin as early as possible after release (SSA 2008, 2010).

17 Recall that the ACS definition of disability differs from the stricter SSA definition. See note 12.

18 Among OASDI beneficiaries, 9.5 percent are black according to ACS data, and 10.4 percent are black according to SSA. This small difference could be due to differing ACS and SSA racial definitions (see note 6). Using data from SSA's Annual Statistical Supplement to the Social Security Bulletin, 2010, we calculate that ACS captures almost 90 percent of the total OASDI population and about 81 percent of the black OASDI population aged 15 or older. The ACS undercount of OASDI beneficiaries could be due to various reasons, including inaccurate self-reporting, respondent confusion between OASDI and SSI, sampling error, and different data collection methodologies (see notes 7 and 12).

19 SSI is distinct from OASDI although SSA administers both. SSI provides payments only to persons with very few resources in the form of income and assets. Individuals with resources in excess of $2,000 ($3,000 for a married couple) are generally not SSI-eligible although certain resources—generally a home, an automobile, and household goods and personal effects—are excluded. SSI payments are reduced if an individual or couple is living in another person's household and is receiving both food and shelter from the person in whose household they are living.

20 Children receive OASDI benefits indirectly if one or both of their parents are disabled, retired, or deceased, but they receive SSI payments directly if they are disabled (SSA 2012a, 2012b).


Abbott, Julian. 1977. “Socioeconomic Characteristics of the Elderly: Some Black-White Differences.” Social Security Bulletin 40(7): 16–42.

———. 1980. “Work Experience and Earnings of Middle-Aged Black and White Men, 1965–71.Social Security Bulletin 43(12): 16–34.

Anderson, Sarah E., and Robert C. Whitaker. 2009. “Prevalence of Obesity Among US Preschool Children in Different Racial and Ethnic Groups.” Archives of Pediatrics & Adolescent Medicine 163(4): 344–348.

Bridges, Benjamin, and Sharmila Choudhury. 2007. “Social Security as a Retirement Resource for Near-Retirees, by Race and Ethnicity, Nativity, Benefit Type, and Disability Status.” ORES Working Paper No. 109.

———. 2009. “Examining Social Security Benefits as a Retirement Resource for Near-Retirees, by Race and Ethnicity, Nativity, and Disability Status.” Social Security Bulletin 69(1): 19–44.

Bureau of Labor Statistics. 2010. Labor Force Characteristics by Race and Ethnicity, 2009. Report 1026. Washington, DC: BLS.

Burkhauser, Richard V., T. Lynn Fisher, Andrew J. Houtenville, and Jennifer R. Tennant. 2012. “Using the 2009 CPS-ASEC-SSA Matched Dataset to Show Who Is and Is Not Captured in the Official Six-Question Sequence on Disability.” Paper presented at the 14th Annual Joint Conference of the Retirement Research Consortium, Washington DC, August 2–3.

Butrica, Barbara A., and Karen E. Smith. 2012. “Racial and Ethnic Differences in the Retirement Prospects of Divorced Women in the Baby Boom and Generation X Cohorts.” Social Security Bulletin 72(1): 23–36.

Census Bureau. 2013. “American Community Survey Sample Size—Data.”

———. No date a. “American FactFinder, Table S2610B: Characteristics of the Group Quarters Population by Group Quarters Type, 2005–2009 American Community Survey 5-Year Estimates.”

———. No date b. “Current Population Survey, Annual Social and Economic (ASEC) Supplement. Table POV03: People in Families with Related Children Under 18 by Family Structure, Age, and Sex, Iterated by Income-to-Poverty Ratio and Race: 2009.”

Chin, Marshall H., James X. Zhang, and Katie Merrell. 1998. “Diabetes in the African-American Medicare Population: Morbidity, Quality of Care, and Resource Utilization.” Diabetes Care 21(7): 1090–1095.

Chiu, Ching-Ju, and Linda A. Wray. 2011. “Physical Disability Trajectories in Older Americans With and Without Diabetes: The Role of Age, Gender, Race or Ethnicity, and Education.” The Gerontologist 51(1): 51–63.

Choi, Namkee G. 1997. “Racial Differences in Retirement Income: The Roles of Public and Private Income Sources.” Journal of Aging & Social Policy 9(3): 21–41.

Collins, Rakale, and Marilyn A. Winkleby. 2002. “African American Women and Men at High and Low Risk for Hypertension: A Signal Detection Analysis of NHANES III, 1988–1994.Preventive Medicine 35(4): 303–312.

Cossrow, Nicole, and Bonita Falkner. 2004. “Race/Ethnic Issues in Obesity and Obesity-Related Comorbidities.” Journal of Clinical Endocrinology & Metabolism 89(6): 2590–2594.

Department of Labor. 2012. The African American Labor Force in the Recovery.

Ewert, Stephanie, and Tara Wildhagen. 2011. “Educational Characteristics of Prisoners: Data from the ACS.” Paper presented at the Annual Meeting of the Population Association of America, Washington, DC, March 31–April 2. SEHSD Working Paper No. 2011-8.

Farley, Reynolds. 1988. “After the Starting Line: Blacks and Women in an Uphill Race. Demography 25(4): 477–495.

Favreault, Melissa M. 2010. “Why Do Some Workers Have Low Social Security Benefits?” The Retirement Policy Program Discussion Paper No. 10-03. Washington, DC: Urban Institute Press.

Gendell, Murray, and Jacob S. Siegel. 1996. “Trends in Retirement Age in the United States, 1955–1993, by Sex and Race.” Journal of Gerontology: Social Sciences 51B(3): S132–S139.

Gibson, Rose C. 1987. “Reconceptualizing Retirement for Black Americans.” The Gerontologist 27(6): 691–698.

———. 1991. “The Subjective Retirement of Black Americans.” Journal of Gerontology: Social Sciences 46(4): S204–S209.

———. 1994. “The Age-By-Race Gap in Health and Mortality in the Older Population: A Social Science Research Agenda.” The Gerontologist 34(4): 454–462.

Gustman, Alan L., and Thomas L. Steinmeier. 2004. “Understanding Patterns of Social Security Benefit Receipt, Pensions Incomes, Retirement and Saving by Race, Ethnicity, Gender and Marital Status: A Structural Approach.” MRRC Working Paper No. WP 2004-082. Ann Arbor, MI: University of Michigan Retirement Research Center.

Harlow, Caroline Wolf. 2003. “Education and Correctional Populations.” Bureau of Justice Statistics Special Report NCJ 195670. .

Hayward, Mark D., Samantha Friedman, and Hsinmu Chen. 1996. “Race Inequalities in Men's Retirement.” Journal of Gerontology: Social Sciences 51B(1): S1–S10.

Hendley, Alexa A., and Natasha F. Bilimoria. 1999. “Minorities and Social Security: An Analysis of Racial and Ethnic Differences in the Current Program.” Social Security Bulletin 62(2): 59–64.

Huntley, Diane. 1979. “Educational and Economic Characteristics of Student Beneficiaries: African American-White Differences.” Social Security Bulletin 42(9): 3–10.

Hutcherson, Donald Tyrone, II. 2009. Street Dreams: The Effect of Incarceration on Illegal Earnings. Doctoral dissertation. Columbus, OH: The Ohio State University.

Joseph, Mark L. 2010. “Understanding the Economic Costs of Incarceration for African American Males.” In Social Work with African American Males: Health, Mental Health, and Social Policy, edited by Waldo E. Johnson, 311–324. New York, NY: Oxford University Press.

Kannel, W. B. 1996. “Blood Pressure as a Cardiovascular Risk Factor: Prevention and Treatment.” Journal of the American Medical Association 275(20): 1571–1576.

Kirk, Adele. 2012. “Understanding the Growth in Federal Disability Programs: Who Are the Marginal Beneficiaries, and How Much Do They Cost?” CRR Working Paper No. 2012-1. Chestnut Hill, MA: Center for Retirement Research at Boston College.

Levin, Henry M., Clive Belfield, Peter Muennig, and Cecilia Rouse. 2006. “The Public Returns to Public Educational Investments in African American Males.” Paper presented at the Economics of Education: Major Contributions and Future Directions Conference, Dijon, France (June).

Lillard, Lee A., and Constantijn W. A. Panis. 1996. “Marital Status and Mortality: The Role of Health.” Demography 33(3): 313–327.

Meyer, Madonna Harrington, Douglas A. Wolf, and Christine L. Himes. 2004. “Linking Benefits to Marital Status: Race and Diminishing Access to Social Security Spouse and Widow Benefits in the U.S.CRR Working Paper No. 2004-05 Chestnut Hill, MA: Center for Retirement Research at Boston College.

National Center for Health Statistics. 2011. Health, United States, 2010—With Special Feature on Death and Dying. DHHS Pub. No. 2011-1232. Hyattsville, MD: NCHS.

Navarro, Alfredo, Karen E. King, and Michael Starsinic. 2011. Comparison of the American Community Survey Voluntary Versus Mandatory Estimates. Washington, DC: Census Bureau.

Newcomb, Chad. 2003/2004. “Demographic and Economic Characteristics of Children in Families Receiving Social Security.” Social Security Bulletin 65(2): 28–48.

Nichols, Orlo R., Michael D. Clingman, and Milton P. Glanz. 2001. “Internal Rates of Return Under the OASDI Program for Hypothetical Workers.” Actuarial Note No. 144. Baltimore, MD: SSA, Office of the Chief Actuary.

O'Rand, Angela M. 1996. “The Precious and the Precocious: Understanding Cumulative Disadvantage and Cumulative Advantage Over the Life Course.” The Gerontologist 36(2): 230–238.

Parsons, Donald O. 1980. “Racial Trends in Male Labor Force Participation.” American Economic Review 70(5): 911–920.

Pettit, Becky, and Bryan Sykes. 2012. “Measuring Racial Inequality in the ACS.” In The Benefits (and Burdens) of the American Community Survey (ACS), edited by The Committee on National Statistics, 76–79. Washington, DC: The National Academies Press.

Quereshi, Adnan I., Fareed K. Suri, Jawad F. Kirmaji, and Afshin A. Divani. 2005. “Prevalence and Trends of Prehypertension and Hypertension in United States: National Health and Nutrition Examination Surveys 1976 to 2000.” Medical Science Monitor 11(9): 403–409.

Raphael, Steven. 2008. “Boosting the Earnings and Employment of Low-Skilled Workers in the United States: Making Work Pay and Removing Barriers to Employment and Social Mobility.” In A Future of Good Jobs? America's Challenge in the Global Economy, edited by Timothy J. Bartik and Susan N. Houseman, 245–304. Kalamazoo, MI: W. E. Upjohn Institute for Employment Research.

Scott, Charles G. 1999. “Identifying the Race or Ethnicity of SSI Recipients.” Social Security Bulletin 62(4): 9–20.

Sears, James, and Kalman Rupp. 2003. “Exploring Social Security Payment History Matched with the Survey of Income and Program Participation.” Paper presented at the Federal Committee on Statistical Methodology Research Conference, Washington, DC (November 18). .

[SSA] Social Security Administration. 2008. “Social Security: Entering the Community After Incarceration—How We Can Help.” SSA Publication No. 05-10504.

———. 2010. “Social Security: What Prisoners Need to Know.” Publication No. 05-10133.

———. 2011. “Removal of Race from Tables in the Annual Statistical Supplement to the Social Security Bulletin and the Earnings and the Employment Data for Workers Covered Under Social Security and Medicare, by State and County—Decision.” Internal memorandum (February 17).

———. 2012a. “Benefits for Children.” Publication No. 05-10085.

———. 2012b. “Benefits for Children with Disabilities.” Publication No. 05-10026.

Stapleton, David, Todd Honeycutt, and Bruce Schechter. 2012. “Out of Sight, Out of Mind: Including Group Quarters Residents With Household Residents Can Change What We Know About Working-Age People with Disabilities.” Demography 49(1): 267–289.

Tamborini, Christopher R. 2007. “The Never-Married in Old Age: Projections and Concerns for the Near Future.” Social Security Bulletin 67(2): 25–40.

Tamborini, Christopher R., Emily Cupito, and Dave Shoffner. 2011. “A Profile of Social Security Child Beneficiaries and Their Families: Sociodemographic and Economic Characteristics.” Social Security Bulletin 71(1): 1–15.

Tamborini, Christopher R., Howard M. Iams, and Kevin Whitman. 2009. “Marital History, Race, and Social Security Spouse and Widow Benefit Eligibility in the United States.” Research on Aging 31(5): 577–605.

Thomas, Pierre, and Jason Ryan. 2008. “U.S. Prison Population Hits All-Time High: 2.3 Million Incarcerated.” ABC News, June 6.

Thompson, Gayle B. 1975. “Blacks and Social Security Benefits: Trends, 1960–73.Social Security Bulletin 38(4): 30–40.

Waite, Linda J., and Maggie Gallagher. 2000. The Case for Marriage. New York, NY: Broadway Books.

Wakefield, Sara. 2010. “Invisible Inequality, Million Dollar Blocks, and Extra-Legal Punishment: A Review of Recent Contributions to Mass Incarceration Scholarship.” Punishment & Society 12(2): 209–215.

West, Heather C. 2010. Prison Inmates at Midyear 2009—Statistical Tables. Report No. NCJ 230113. Washington, DC: Bureau of Justice Statistics.

Western, Bruce, and Christopher Wildeman. 2009. “The Black Family and Mass Incarceration.” The Annals of the American Academy of Political and Social Science 621(1): 221–242.

Western, Bruce, and Becky Pettit. 2010. “Incarceration and Social Inequality.” Daedalus: Journal of the American Academy of Arts and Sciences Summer 2010: 8–19.

Whaley, Arthur L., Michelle Smith, and Anyaliese Hancock. 2011. “Ethnic/Racial Differences in the Self-Reported Physical and Mental Health Correlates of Adolescent Obesity.” Journal of Health Psychology 16(7): 1048–1057.

Wilmoth, Janet, and Gregor Koso. 2002. “Does Marital History Matter? Marital Status and Wealth Outcomes Among Preretirement Adults.” Journal of Marriage and Family 64(1): 254–268.

Wright, Jackson T., Jr., Lawrence Agodoa, Gabriel Contreras, Tom Greene, Janice G. Douglas, James Lash, Otelio Randall, Nancy Rogers, Michael C. Smith, and Shaul Massry. 2002. “Successful Blood Pressure Control in the African American Study of Kidney Disease and Hypertension.” Archives of Internal Medicine 162(14): 1636–1643.