How Progressive Is Social Security and Why?

Number 37 in Series "Straight Talk on Social Security and Retirement Policy"

The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

Social Security was designed to redistribute income from those with higher lifetime earnings to those with lower lifetime earnings. The reason is obvious: the system was created to ensure an adequate retirement income for the elderly. Less obvious is how Social Security's many provisions interact to achieve redistribution. This Straight Talk summarizes the most comprehensive study of those interactions to date, concluding that less-educated, lower-income, and nonwhite groups benefit little or not at all from redistribution in the old age and survivors insurance (OASI) part of Social Security. However, there is substantial redistribution to women, who historically have had lower lifetime earnings than men. A succeeding Straight Talk will examine how the addition of disability insurance restores some progressivity.¹

The Social Security program redistributes income in five major ways:

1. From richer workers to poorer workers through a progressive benefit formula that provides higher returns to the first dollars of worker earnings and lower returns to the last dollars;²
2. From shorter-lived groups (such as men and the less educated) to longer-lived groups (such as women and the better educated) through annuities whose lifetime value depends upon life expectancy;
3. From singles to married couples (and from higher earners to lower earners within couples) through spousal and survivors' benefits, paid as a pure transfer without any additional contributions required;
4. From the healthy to the disabled through disability benefits; and
5. From later generations to earlier generations, since earlier generations paid in at lower tax rates than later generations, yet receive benefits related to their prior earnings (rather than, as in private insurance, their actual contributions).

To test how different factors affect this redistribution, we ran a simulation using survey data that began by giving all individuals certain common traits: the same age of death (84.05), the same age of retirement (63.15), and the same benefit formula (41.2 percent of career average earnings). Then, one by one, we reintroduced the progressive benefit formula, actual ages of retirement, and ages of death, noting the effect of each. We did this separately for men and women.

Why go through this type of exercise? It is a mistake to consider Social Security a monolithic program that will simply be replaced by another monolithic program. Granting a spousal or survivors' benefit as an add-on requiring no additional contributions, providing higher survivors' and spouses' benefits to those married to higher earners, basing benefits on earnings subject to tax rather than actual taxes paid, using progressive rates in the benefit formula, and forcing annuitization are all policy decisions that deserve separate attention.

There are many ways to measure redistribution. Here we report on differences in real internal rates of return (IRRs)³ and assume that, for couples, benefits and taxes are shared equally during those years when both are married and alive.

• Redistribution by gender. When we control for the redistributive effects of mortality, retirement age, and benefit formula, males and females have similar rates of return. When these factors are reintroduced, mortality is the dominant factor affecting returns by gender, raising all groups of women to a return higher than any group of men.
• Redistribution by education. Education can be considered a proxy for potential earnings. For men, OASI is not necessarily "progressive" by education. High school dropouts actually receive lower rates of return than high school and college graduates. For high school dropouts, the progressive benefit formula and the age of retirement raise the relative rate of return, but both effects are overcome by losses due to higher mortality.
• Redistribution by race/ethnicity. Blacks do about the same as whites both before and after all adjustments are made, although again, women clearly gain relative to men. Hispanics appear to do slightly better than both blacks and whites.
• Redistribution by earnings quintile. Although not shown in the table, redistribution by earnings quintile is progressive for both men and women. So why is redistribution by education not progressive for men while it is for women? Although including mortality fails to affect female high school dropouts, it makes a substantial difference for first and second quintile women because many seemingly low-earning women are moderately to well-educated and are married to higher-earning men, while having minimal labor force attachment. Thus, lower lifetime-earning women include significant numbers of both well-educated women with only modest labor force attachment and less-educated women with significant labor force participation at low wages.

In sum, OASI redistributes from higher earners to lower earners through its progressive benefit formula. It also redistributes from men to women by paying benefits in the form of annuities. Single women (including many divorced women), however, fare much more poorly than married women—a fact hidden in the data on all women. OASI does not increase the rates of return of nonwhite, less educated men enough to offset the penalty they face on average because of higher mortality. For the most part, differences in rates of return in OASI across different educational and racial groups are not large for any generation retiring now or in the near future.

Eugene Steuerle is a senior fellow and Adam Carasso is a research associate at the Urban Institute. Lee Cohen is an economist in the Social Security Administration Office of Policy.

Notes

This brief highlights findings from Lee Cohen, C. Eugene Steuerle, and Adam Carasso, "Social Security Redistribution by Education, Race, and Income: How Much and Why," paper prepared for the Retirement Research Consortium, Washington, D.C., May 17-18, 2001.

1. Both analyses use the Modeling Income in the Near Term (MINT) model v.2, jointly developed by the Urban Institute, the Rand Corporation, and the Social Security Administration.

2. Currently, the formula pays 90 percent of the first $606 of average monthly earnings, 32 percent of the next $3,047, and 15 percent of the remainder (up to a maximum).

3. For our purposes, the IRR is calculated as the interest rate workers would need to earn on their lifetime Social Security taxes to make this sum equal to the lifetime retirement and disability benefits they receive, adjusted for inflation.

About the Series

This series is made possible by an Andrew W. Mellon Foundation grant. For more information, call Public Affairs: 202-261-5709. For additional copies of this publication, call 202-261-5687 or visit the Retirement Project's web site: http://www.urban.org/retirement.

Usage and reprints: Most publications may be downloaded free of charge from the web site and may be used and copies made for research, academic, policy or other non-commercial purposes. Proper attribution is required. Posting UI research papers on other websites is permitted subject to prior approval from the Urban Institute—contact [email protected].

If you are unable to access or print the PDF document please contact us or call the Publications Office at (202) 261-5687.

Disclaimer: The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Copyright of the written materials contained within the Urban Institute website is owned or controlled by the Urban Institute.