May 12, 2003

GAO REPORT SHOWS SOCIAL SECURITY IS FAVORABLE TO PEOPLE OF COLOR BUT SOME CHANGES IN IT COULD HARM MINORITY COMMUNITIES
by Kilolo Kijakazi

Summary

PDF of this report View Related Analyses

More Topics... Back to Home Page Search the Site  - Publications by Subject  - Upcoming Releases  - Unemployment Insurance About the Center  - Board of Directors  - Internship Programs  - Job Opportunities  - Staff Bios Special Projects  - DC Fiscal Policy Institute  - International Budget Project  - State Fiscal Analysis Initiative  - State Policy Documentation Project  - Start Healthy/Stay Healthy Join E-Mail List Donate to the Center

If you cannot access the files through the links, right-click on the underlined text, click "Save Link As," download to your directory, and open the document in Adobe Acrobat Reader.

The General Accounting Office (GAO) has just released a report analyzing the experience of African Americans and Latinos under Social Security and whether they benefit more or less than whites under the system.[1]  The study found that “In the aggregate, blacks and Hispanics have higher disability rates and lower lifetime earnings, and thus receive greater benefits relative to taxes [paid] than whites.”  The report also warns that Social Security reform proposals “tying benefits more directly to contributions [a characteristic of many proposals to replace Social Security partially with individual accounts] may affect the progressivity of the system and are likely to disproportionately affect equity for minorities …”

An Associated Press account on April 24 summarized the study’s findings as follows:  “Minorities are greatly helped by disability and survivor benefits in the Social Security system and would be hurt by an overhaul that instead linked benefits mostly to contributions, congressional auditors said Wednesday.  The General Accounting Office report issued Wednesday appeared to contradict contentions by supporters of Social Security-linked personal investment accounts that minorities fare worse in the current system.”

 

Brief Description of the Study

One question about Social Security, both historically and currently, is whether the program is less beneficial to people of color, especially African Americans, than to whites.  The basis for the question is largely the fact that African Americans have a higher mortality rate than the general population and consequently may not receive Social Security benefits for as many years, on average, as other groups (and may not receive benefits at all if they die before reaching retirement age).

The GAO examined this question primarily by comparing the payroll tax contributions made by workers to the Social Security benefits that the workers and their families subsequently receive.  The GAO made use of a Social Security Administration (SSA) microsimulation model that combines longitudinal survey data on workers with data on the actual payroll tax contributions these workers make and the benefits they receive.[2]

 

Findings

While the Social Security program is designed to be neutral with respect to race, ethnicity, and gender, differences occur in the relationship between taxes paid and benefits received, as a result of differences in lifetime earnings, rates of disability, and mortality rates among groups.

Lifetime Earnings

African American and Latino workers, on average, have lower levels of lifetime earnings than white workers.  For example, the GAO report indicates that 38 percent of Latino workers and 35 percent of African American workers born between 1931 and 1940 had lifetime earnings that fell into the bottom fifth of earnings received by workers born in these years.  Conversely, only nine percent of Latino workers and 11 percent of African American workers had earnings that placed them in the highest fifth of workers.  By contrast, 17 percent of white workers had earnings that placed them in the lowest fifth, and 22 percent had earnings in the highest fifth.

As a result of their over-representation among workers receiving the lowest earnings and their under-representation among workers receiving the highest level of earnings, African Americans and Latinos gain from Social Security’s progressive benefit formula.  The formula provides low-wage earners with benefit levels that replace a higher percentage of their pre-retirement wages than they replace for high-wage earners.  For lower-wage earners, the progressive benefit formula thus increases the level of benefits received relative to payroll taxes paid.  The progressive benefit formula increases the ratio of benefits received-to-taxes paid for African Americans and Latinos in the aggregate, because they are disproportionately represented among low-wage earners.

Incidence of Disability

African Americans and Latinos have higher rates of disability and consequently are more likely to receive benefits from the Social Security Disability Insurance (DI) program.  The GAO reports that African Americans made up 10.3 percent of its sample but 16.9 percent of the DI beneficiaries.[3]  Similarly, Latinos made up 8.4 percent of the sample but constituted 10.1 percent of the DI beneficiaries.

The GAO report explains that, in general, workers with a higher probability of becoming disabled have a higher benefits received-to-taxes paid ratio than those who are not disabled.  This is because Social Security Disability Insurance beneficiaries begin to receive benefits, and to cease contributing payroll taxes, at an earlier age than workers who are not disabled.  The higher incidence of disability among African Americans and Latinos increases their benefit-to-tax ratio.

Since low-wage DI beneficiaries also gain from the program’s progressive benefit formula, low-wage, disabled workers benefit from two aspects of the program.  African Americans and Latinos are disproportionately represented among both disabled workers and low-wage earners.  Social Security’s progressive benefit structure and its Disability Insurance component are the critical reasons that people of color, as a group, receive greater benefits relative to taxes paid than whites do.

It is important to note that DI benefits are paid not only to disabled workers but to the families of such workers as well.  Social Security Administration (SSA) data indicate that African-American children make up 20 percent of the children receiving benefits because a parent is disabled. [4]  African American children make up about 15 percent of all children in the country, so they benefit disproportionately from Social Security’s DI component. [5]  (Comparable SSA data are not available for Latino children.)

Incidence of Mortality

While a greater prevalence of low-wage earnings and a higher incidence of disability serve to raise benefits relative to tax contributions for both African Americans and Latinos, the impact of mortality rates differs between these communities.  African Americans have a higher mortality rate than the general population.  While African Americans represent 10.3 percent of GAO’s sample, they made up about 16 percent of the workers who died before age 62.[6]  By contrast, Latinos not only live longer than African Americans, on average, but also live longer than white people.

Workers with higher mortality rates do not contribute payroll taxes for as many years as workers with lower incidences of mortality, but they receive Social Security benefits for a shorter period of time, if they receive these benefits at all.  Not surprisingly, the GAO found that the benefit-to-tax ratio is expected to be lower for workers who die before they retire than for workers who live to receive retirement benefits.

Among workers who live until retirement age, African Americans are expected to receive benefits for fewer years than white or Latino retirees, because they have shorter life expectancies.  An African American man who was 20 years old in 2001 is expected ultimately to have 12 years of retirement before dying, assuming that he retires at age 62.  By comparison, a white man who was 20 in 2001 is expected to have 16 years of retirement and a Latino is expected to have 19 years.

There are, however, some partially compensating factors.  While African American workers have a higher probability of dying before retirement and of living fewer years after retirement, their families are more likely, as a consequence, to receive Social Security survivor benefits.  The GAO reports that 23 percent of the children who received Social Security survivor benefits in 2001 were African American.  Since, as noted earlier, only about 15 percent of the children in the country are African American, this means that African American children receive Social Security survivor benefits at a substantially higher rate than white children do.  Thus, while African American workers receive retirement benefits for fewer years, on average, than other groups, this is partially offset by the increased receipt of Social Security survivor benefits by members of their families.[7]

Combined Effects of the Three Key Factors

The combined effect of these three socioeconomic factors (lifetime earnings, disability rates, and mortality rates) is that African American and Latino workers enjoy a modestly higher benefits received-to-taxes paid ratio than white workers.  (All three groups receive more in benefits, on average, than they contribute in payroll taxes.)  Both African American and Latino workers are helped by the progressive benefit formula and by Social Security’s disability component.  African Americans also gain significantly from the survivor benefit, while Latinos are helped by having more years of retirement benefit receipt (due to their greater longevity).

 

Warning to Reformers

Given these findings, the GAO report also contains a cautionary note about the potential detrimental effect of some types of Social Security reform.  The study states:

                        Reforms that … [tie] benefits more directly to contributions may affect the progressivity of the system and are likely to disproportionately affect equity for minorities as measured in this report.  Further, disability, survivors’, and dependents’ benefits also could be affected by any changes to the structure of the program.[8]

 

Conclusion

The study makes clear the fallacy of claims that Social Security is a bad deal for people of color.  By analyzing payroll tax and benefit histories of actual workers, the GAO has confirmed that people of color receive more in benefits than they contribute in payroll taxes and receive greater benefits relative to taxes paid, in the aggregate, than do white workers.  Reforms that alter the benefit structure of Social Security by linking benefits more closely to payroll contributions and weakening the progressive nature of the Social Security benefit structure, or by reducing disability and survivor benefits significantly as part of a redesign of the system, could have disproportionate adverse effects on people of color.

---

End Notes:

[1] General Accounting Office. Social Security and Minorities: Earnings, Disability Incidence, and Mortality Are Key Factors that Influence Taxes Paid and Benefits Received, April 2003 (GAO-03-387).

[2] The microsumulation model used in the study is SSA’s Modeling Income in the Near Term or MINT3 model.  The payroll tax and benefit histories contained in the model are confidential; the model consequently is not available to outside users.  To conduct this study, GAO specified the analyses to be done on the model, and SSA’s Office of Policy carried out the analysis.  A limitation of the model is that it does not include beneficiaries who are children, surviving spouses with young children, or aged parents, although it attempts to control for the exclusion of these beneficiaries. 

[3] General Accounting Office, op. cit., p. 12.

[4] Social Security Administration, Annual Statistical Supplement, 2002.  Washington, DC: U.S. Government Printing Office, 2003, Table 5.A1.4, p. 181.

[5] U.S. Bureau of the Census.  Statistical Abstract of the United States:  2000.  www.census.gov/prod/2001pubs/statab/sec01.pdf, Table 19.

[6] General Accounting Office, op. cit., p. 14.

[7] General Accounting Office, op. cit. p.15.

[8] General Accounting Office, op cit., p. 16