Report | Retirement

Two Steps Back: African Americans and Latinos will lose ground under Social Security ‘reform’

Issue Brief #212

 Download PDF

Share this page:


Two Steps Back

African Americans and Latinos will lose ground under Social Security “reform”

by Ross Eisenbrey and William Spriggs

The United Airlines pension debacle, the president’s Social Security proposal, and the coming debate about tax incentives for saving are all prompting a re-examination of U.S. retirement policies and programs. At such a time, it is essential to understand the significant racial differences in the effects of these policies. The policy mix over the past two decades, which has favored savings in individual tax-deferred accounts at the expense of defined-benefit pension plans, has contributed to retirement income inequality and left millions of minority households without sufficient resources to retire in comfort or dignity. Two of every five minority households nearing retirement can expect retirement income that is less than double the poverty level.

The one bright spot in this picture is Social Security, which is pivotal to minorities’ retirement security, and which grew faster for African American and Latino households than for whites from 1989 to 2001. Social Security’s progressive benefit structure and virtually universal coverage helped ameliorate the lack of gains in other retirement wealth for these minorities. But Social Security “reform” that makes Social Security more like a private defined-contribution account and less like a defined-benefit plan is likely to hurt minorities and exacerbate their vulnerability to employment cycles.

The retirement gap

Minority households nearing retirement can generally expect a much lower standard of living than non-Latino whites. In one sense, this is not surprising, since median African American and Latino households earn substantially less than white ones. African Americans, for example, make up 11.5% of workers contributing to Social Security, but receive only 8.9% of total taxable earnings. The mean income of minority households ages 47-55 was just 44% of similar white households. For minorities ages 55-64, mean income was only 40% of whites’. But racial differences in wealth holdings point to a future in which minority retirees will fare even worse in retirement than they do during their working lives.

As shown in Figure 1, the non-Social Security assets gap between white households and African American or Latino ones, and the effect it will have on their retirement income, are substantial. The expected mean retirement income from wealth, expected pension benefits, and Social Security benefits for non-Latino white households nearing retirement (ages 47-64) was $80,800 in 2001. African Americans and Latinos could expect, on average, only a little more than one-third as much, $28,100 (Weller and Wolff 2005).

(figure 1)

Although the average expected income for African Americans and Latinos is low, most minority households preparing for retirement are doing even worse: in 2001, retirement income for 57% of African American and Latino households nearing retirement added up to less than twice the poverty line. White households were far less likely to fall below this threshold, though their numbers are somewhat surprising: 23% (almost one in four) had expected retirement income less than twice the poverty line.

Troubling as these numbers are, they are an improvement over 1989, when 68% of African American and Latino households approached retirement with the expectation that their income would be less than twice the poverty line. Nevertheless, despite this improvement, the projected retirement income gap grew larger between whites and minorities from 1989 to 2001.

Mean expected income rose from $22,200 to $28,100 for African Americans and Latinos, but even faster—from $59,100 to $80,800—for white households near retirement age. Thus, from 1989 to 2001, the gap between the mean projected retirement income of white and minority households widened from $36,900 to $58,600. In coming years, therefore, if U.S. retirement policy is not changed, the income of minority retirees will continue fall even further behind that of white retirees.

The three kinds of retirement wealth

These estimates of projected income in retirement depend on the projected wealth that households have accumulated. The major components of that wealth are: (1) Social Security wealth, which provides a steady cash income stream to the household in retirement; (2) defined-contribution (DC) and defined-benefit (DB) plan “pension” wealth; and (3) other wealth, including housing wealth and non-pension financial wealth. The composition of household wealth differs significantly by race.

The experience from 1989 to 2001 indicates that policies that preserve and strengthen Social Security benefits, rather than increasing reliance on DC plans or private retirement accounts, are the best hope for promoting adequate retirement income for minority households.

Social Security wealth   
The most important source of retirement income for the typical household is Social Security, and it is especially important for minority households. The share of income from Social Security for median elderly couples and individuals with Social Security income, from 2001 to 2003, was 66% for whites, 82% for African Americans, and 84% for Latinos. The experience of both white and minority households with respect to Social Security wealth was extremely positive for each age group over the period from 1989 to 2001 (Figure 2). African Americans and Latinos saw their mean Social Security wealth grow 49% for households ages 47-55 and 99% for ages 56-64. For whites, the gains were greater in total dollars but less in percentage terms: 54% for households ages 47-55 and 55% for those age 56-64.

Figure 2

Social Security wealth depends on the wages earned over a lifetime—generally, the more one earns, the greater the income in retirement. Rising wages and the improved employment picture from 1989 to 2001 thus led to widespread and substantial increases in Social Security wealth. For minorities especially, the strong labor market of the 1990s brought more employment at higher wages.

For minority households, Social Security was even more important. The increase in mean Social Security wealth between 1989 and 2001 accounted for more than 75% of the overall increase in retirement wealth for each of the three age groups of African American and Latino households. And for large numbers of minorities, Social Security is the only source of cash income in retirement: 38% of Latinos and 37% of African Americans have no other source of retirement income, compared to 18% of whites (Sun and Ghilarducci 2004).

Social Security is the most effective means of assuring at least a modest retirement income for the vast majority of American households, regardless of race, for three main reasons. First, unlike pensions, Social Security is virtually universal. Second, it compels saving, which is otherwise very hard for low-income households, among which minorities are over
represented. And, third, its benefit structure is progressive, replacing a higher percentage of the pre-retirement wages of low earners than of higher earners. Unlike the current regressive system of tax deferrals for private savings through IRAs, 401(k)s, and other private accounts, Social Security acts as an equalizer for the regressive nature of private savings.

The ameliorative effect of Social Security.   Because of the progressive structure of Social Security benefits, the racial gap in retirement income for households 65 and older in 2001 was smaller than the earnings gap during their working lives. The average Social Security retirement benefit for African American retirees is 85% of the benefit for white retirees.

African Americans earn less than whites while they are working, and African American retirees have lower non-Social Security income than whites. But because Social Security replaces more income for low-earners than for those better off, the income of African Americans falls less after 65 than for whites. The ratio of household income for those 65 and older to those 55 to 64 is 56% for African Americans and 53% for whites. In 2004, according to the Census Bureau, the income gap (from all income sources) between African American and white households headed by seniors over 65 was 74 cents on the dollar, the smallest gap for any age group of African American households.

Pension wealth   
Despite the record-breaking performance of the stock market in the 1990s, DB and DC pension wealth grew much less evenly than Social Security wealth among different racial and age groups. Pension wealth actually declined for many African American and Latino households over this period. In fact, on average, African American and Latino households age 65 and over lost pension wealth, as did those ages 47-55. A major factor in this loss is the decline in defined-benefit pension plan coverage over that period, which disproportionately affected unionized workers who, in turn, are disproportionately minorities. Because of their lower wages, minorities—especially Latinos—are generally less able and less likely than whites to save for their retirement through defined-contribution plans.

Defined-benefit coverage fell dramatically from 1989 to 2001, as employers substituted DC plans or simply dropped their DB plans without replacing them at all. The decline in DB coverage was especially steep for people ages 47 to 55. Those with any DB plan dropped from 68% to 40%, and those with only DB coverage went from 55% to 10%.

For minorities, the problem was not just loss of DB coverage, but lack of pension coverage of any kind. In 2001, 63% of Latinos and 44% of African Americans worked for employers who didn’t offer pension plans, compared to 41% of whites. By 2000, according to the Department of Labor, only 44% of African American male workers and 28% of Latino male workers had any pension coverage at all, compared with 58% of white men. Only 43% of African American women and 30% of Latino women had pension coverage, compared with 52% of white women.

One group of minority households—those between the ages of 56 and 64—nevertheless managed a healthy 36% gain in mean pension wealth between 1989 and 2001, in line with their white counterparts. Unfortunately, even after that healthy increase, these households averaged only $68,000 in pension wealth, a very modest source of retirement income. The group that gained the most in pension wealth over this period was white households ages 47-55, whose pension wealth increased 103%, from $77,000 to $156,000. At least in part, their success probably reflects the fact that such households started out owning relatively little DB pension wealth in 1989. While the other cohorts each lost over a third of their DB pension wealth, the 47-55 year-olds lost less because they had less to lose (Weller and Wolff 2005, table 4). Every age group saw large percentage gains in their mean DC plan wealth over this period, both because coverage in such plans grew strongly and because the stock market tripled in value over this period.

Because DB pension coverage is still declining and minorities are far less likely than whites to have pension coverage, Social Security is more reliable than employer-provided pensions as a means to ensure that minority households have adequate income in retirement. Even when DC plans are available, low-wage minority workers are less able to participate and make contributions because housing costs and health care consume so much of their income. Without refundable tax credits or a substantial employer match, many minority households will never be able to add much DC wealth to their retirement savings.

The structure of IRA and 401(k) plans exacerbates racial differences in the labor market in two other ways. First, because of labor market discrimination, young African Americans are much less likely to have a job than whites. Among 20- to 29-year-olds in January 2005, 75% of whites had a job versus 60% of African Americans.

So a greater share of whites can contribute to IRAs and 401(k)s in their early years when savings count much more than later in life because the interest earned on those accounts has more years to compound. Additionally, beyond labor market problems, African Americans need more education to earn the same income as less-educated whites. In 2003, the median white high school graduate earned $26,518, but African Americans needed high school plus some college coursework to earn the same income. Therefore, to earn the same income later in life, African Americans must spend some of their early years getting additional education while whites are already earning and making retirement contributions to an IRA or 401(k).

The same problem does not occur with Social Security. Under the current formula, Social Security uses a worker’s 35 highest earning years to calculate their benefit. Thus, the difficulties African Americans have in their early years play a lesser role in determining Social Security benefits. In addition, the Social Security benefit formula gives lower earners a higher share of their pre-retirement earnings in benefits than it gives higher income workers.

Second, stocks return their highest yield if they are bought when prices are low and sold when they are high. But African Americans’ job opportunities are more sensitive than other groups’ to fluctuations in the economy, as William Rodgers at Rutgers and others have documented. When the economy is growing and the stock market does well, African Americans are able to find jobs and build up IRA and 401(k) portfolios, but they do so when the market is high. When the economy slows and the market is low, African American unemployment jumps high and they are less able to put money into these accounts. Weller and Wolff (2004) shows this phenomenon would substantially reduce the returns on African Americans’ accounts compared to those of whites.

Defined-benefit plans, whose structure more resembles Social Security than defined-contribution plans, are better suited to addressing the problems faced by African American and Latino workers than are simple saving plans. Tax incentives to encourage private saving that are not progressive in nature and do not compensate for business cycle or labor market effects will exacerbate rather than close the retirement savings gap faced by African American and Latino seniors.

Other wealth   
From 1989 to 2001, other wealth—from h
ousing/real estate, savings, cash, money market accounts, bonds, the cash surrender value of life insurance plans, and stocks—increased, on average, $28,500 for Americans nearing retirement, a modest 14% overall. As a percent of total retirement wealth, however, middle-income Americans’ other wealth decreased from 46% in 1989 to 40% in 2001. Rises in the stock and housing markets had little effect on typical middle-income Americans because they had little invested in the stock market (outside of a 401(k) or similar plan) and their increased home equity was offset by higher home borrowing.

For minorities the role of other wealth in preparing households for retirement is further reduced. The wealth gap between African Americans and non-African Americans has long been substantially greater than the income gap. According to Havens and Schervish (2004), by 2001, the average African American household had wealth that was 19% of the national average but income that was 55% of the national average. In 2001, only 333,000 out of 13.2 million African American households owned as much as $500,000 in wealth. The picture was no brighter for Latinos.

Unlike white households, for minorities the great majority of “other wealth” is concentrated in home equity. The Pew Hispanic Center (2004) reports that, in 2002, the mean net worth of Latino households was approximately $65,371, with 61% derived from home equity. The overwhelming importance of homeownership is also evident for African American households for whom home equity comprised 63% of mean total net worth of $45,711 in 2002. By contrast, the Pew Hispanic Center found that home equity was only 38.5% of the mean net worth of $221,871 of white households. Stocks and mutual funds accounted for 21.9% of the mean wealth of white households in 2002, but less than 12% of the mean wealth of minority households.

Until the wages and salaries of America’s minorities improve enough to allow them to make substantial investments in financial assets, other wealth is unlikely to be a source of important retirement income for African American and Latino households.

Conclusion

An equitable national retirement system that aims to provide minority households with sufficient income to retire in comfort and dignity must either change or, at the least, take into account the facts that Latinos and African Americans on average earn substantially lower wages and are far less likely to be covered by employer-provided pensions than whites. As the experience from 1989 to 2001 has made clear, even a historically strong stock market and increased coverage of minorities by DC pension plans contribute less to wealth at retirement and projected retirement income of minorities than Social Security. Strategies like private accounts, that fail to compensate for the lower incomes and less certain employment of minorities, have been and will continue to be less able to guarantee adequate retirement income to minority households. Strengthening Social Security’s financing and improving its benefits appear to be the most effective way to secure a decent retirement for the greatest number of minority households.

Bibliography

Havens, John J. and Paul Schervish. 2004. Wealth Transfer Estimates for African American Households. Boston, Mass.: Center on Wealth and Philanthropy, Boston College. October.

Pew Hispanic Center. 2004. The Wealth of Hispanic Households, 1996-2002. Washington, D.C.: PHC, October, p.17.

Sun, Wei and Teresa Ghilarducci. 2004. Hispanics’ Low Pension Coverage and Disenfranchisement from the U.S. Financial System. Notre Dame, Ind.: University of Notre Dame.

Weller, Christian and Edward Wolff. 2005. Retirement Income: The Crucial Role of Social Security. Washington, D.C.: EPI.



See related work on Retirement

See more work by Ross Eisenbrey and William E. Spriggs