Revised, October 8, 1998
African Americans, Hispanic Americans, and Social Security:
the Shortcomings of the Heritage Foundation Reports
by Kilolo Kijakazi
Table of Contents
Click here for PDF version of this report
I. Executive Summary
The Heritage Foundation has released two reports on Social Security's rate of return for minorities. One report addresses African Americans; the second speaks to Hispanic Americans. Heritage argues that the African-American community will secure poor and even potentially negative rates of return from Social Security because African Americans have lower life expectancies. Heritage asserts that Social Security also is a bad deal for Hispanic Americans because they are a disproportionally young population and will bear a significant share of the cost of financing Social Security in the future. According to these reports, these communities would fare better under a retirement program of individual accounts invested in the private market.
The Heritage reports contain critical flaws, however, in their assumptions about both Social Security and individual accounts. The erroneous assumptions result in large errors in Heritage's analysis of rates of return for the population in general and for minorities in particular. As this paper explains, Heritage's rate-of-return calculations substantially understate the rate of return from Social Security and heavily overstate the rate of return from the individual accounts that Heritage favors. As a result, the Heritage calculations are inaccurate.
A major part of this paper is drawn from an assessment of the Heritage paper on African Americans that the Social Security Administration's highly respected Office of the Chief Actuary conducted. The Office of the Chief Actuary identified major errors in Heritage's work, including the fact that Heritage exaggerated the taxes that African Americans pay into Social Security while substantially understating the Social Security benefits that African Americans receive. The actuaries concluded that Heritage "grossly underestimates the expected rates of return from Social Security benefits" for the overall population and especially for African Americans. The actuaries also examined Heritage's subsequent work on Hispanic Americans and found it seriously flawed as well. The analyses of Heritage's work by the Office of the Chief Actuary are attached as appendices.
Assumptions that Understate Social Security's Rate of Return
- The Social Security Administration's Office of the Chief Actuary found that Heritage both overstated the payroll taxes that workers pay for Social Security and understated the Social Security retirement benefits that workers receive, with these errors being particularly large for African Americans. The Office of the Chief Actuary found that in estimating the number of years an individual would work and make payroll tax contributions, Heritage failed to take into account the fact that some workers die before retirement and consequently contribute payroll tax for a smaller number of years than those who work into their 60s. Since a larger portion of African-American workers than of white workers die before 65, failing to take this factor into account led Heritage to make particularly large errors in its rate-of-return calculations for African Americans. For example, the actuaries found that the number of years a 20-year-old African-American man is expected to make payroll tax payments into the Social Security system is six years fewer than the Heritage methodology would predict. This flaw in its methodology led Heritage to overstate significantly the Social Security taxes that African Americans pay.
The actuaries found that Heritage made still larger errors in computing the number of years for which African-American workers receive Social Security retirement benefits. The actuaries found that a 20-year old African-American man would be expected to receive such benefits for nearly four times as many years as is assumed under the Heritage methodology. The actuaries' analysis shows that for 20-year-old African-American men, the number of years of drawing Social Security retirement benefits will equal 20.7 percent of their work years. The Heritage methodology, however, erroneously assumes the number of benefit years would equal just 4.9 percent of these individuals' work years.
The actuaries also found the Heritage methodology resulted in a large understatement of Social Security's rate of return "for younger workers who are far from retirement age in 1997," a finding of particular significance for Hispanic Americans, who are a younger-than-average population.
Excerpts from the Analysis of Heritage's Work On African Americans
by the Social Security Administration's Office of the Chief Actuary
"... the [Heritage] conclusions are highly misleading due to two major errors in methodology, plus a number of incorrect or inappropriate assumptions. Due to these errors, rates of return for the general population in this study are at substantial variance with those produced for the 1994-6 Advisory Council on Social Security.
"More importantly, the methods utilized by the authors bias expected rates of return downward to a far greater degree for African Americans than for the general population. This study thus erroneously indicates differences in rates of return by race that are greatly exaggerated. In fact, results from more careful research reflecting actual work histories for workers by race indicate that the non-white population actually enjoys the same or better expected rates of return from Social Security than for the white population."*
* Steve Goss, Deputy Chief Actuary, Social Security Administration, "Problems with 'Social Security's Rate of Return,' A Report of the Heritage Center for Data Analysis," January 27, 1998, p. 1.
- Heritage inappropriately assumed that if Social Security is not privatized, it will be restored to long-term actuarial balance entirely by raising payroll taxes and that this tax increase will begin in 2015, a decade earlier than the Social Security actuaries project would be necessary. No proponent of restoring long-term balance to Social Security without privatization has recommended correcting the imbalance purely through payroll tax increases. Virtually all reform plans that do not include privatization components would increase Social Security revenue through such measures as expanding Social Security coverage to the 30 percent of state and local government employees not currently covered and/or investing a portion of the Social Security reserves in the equities market. In calculating Social Security's rate of return, however, Heritage assumed that the only method used to close the imbalance would be payroll tax increases. Heritage selected what is probably the single approach to closing the Social Security imbalance that would affect rates of return for minorities (and especially Hispanic Americans) most adversely.
Moreover, even if calculations of rates-of-return for Social Security are based on the implausible assumption that the imbalance will be closed entirely through payroll tax increases, Heritage's approach is deficient; Heritage assumed that payroll taxes would be raised years before that would be necessary. The Social Security actuaries project it would not be necessary to raise taxes until 2025 to maintain a pay-as-you-go system without any changes in eligibility or benefits, and even then not by as much as Heritage assumed. In its assessment of this aspect of the Heritage paper, the Office of the Chief Actuary wrote: "In order to provide a pay-as-you-go financed Social Security program, tax rates would not need to be raised until around 2025, and by less than the [Heritage] authors assumed. The early, too-large increase in tax rates results in rates of return that are estimated to be too low for Social Security."
- Heritage failed to calculate the rate of return for the full Social Security system. It ignored Social Security disability and survivors benefits in making its calculations. African Americans benefit disproportionately from the Social Security disability insurance and survivors' insurance programs. A considerably larger proportion of African Americans than of whites receive these benefits. While African Americans make up 11 percent of the civilian labor force, they account for 18 percent of the workers receiving Social Security disability benefits. Social Security Administration analyses have found that African Americans receive substantially more benefits from Social Security Disability Insurance in relation to the taxes they have paid than whites do. Moreover, nearly half of all African Americans who receive Social Security benefits of any kind 47 percent receive either disability or survivors benefits. By contrast, 28 percent of white Social Security beneficiaries receive disability or survivors benefits.
In calculating Social Security's rate of return, Heritage eliminated these two components of the Social Security system, the components most favorable to the African-American community. That makes Social Security appear less valuable to African Americans than it is.
Social Security and Hispanic Americans
Social Security provides a particularly large portion of the income of older Hispanic Americans. While nine percent of all couples age 65 and older rely on Social Security for all of their income, 23 percent of Hispanic couples do. Similarly, while 24 percent all of aged single individuals receive all of their income from Social Security, 40 percent of aged Hispanic single individuals do.
Social Security also is of particular value to Hispanic Americans for another reason. Hispanic retirees live longer, on average, than other Americans the average American who reaches 65 (including both men and women) will live an additional 17½ years, while the average Hispanic reaching 65 lives an additional 20.5 years. Those with a longer life span receive more monthly benefit checks from Social Security. Furthermore, Social Securitys annual cost-of-living adjustment a feature most private annuities do not have is of greater value for those who live longer.
Hispanics thus benefit doubly from Social Security; they benefit both from Social Securitys provision of benefits that keep pace with inflation and cannot run out no matter how long one lives, and also from Social Securitys progressive benefit formula, which ensures that individuals who earned lower wages and/or had fewer years in the workforce receive larger monthly benefit amounts, in proportion to the wages they earned and the taxes they paid, than other workers do. Since Hispanic retirees on average have lower wages and fewer covered years of employment and also live longer than other workers, they receive benefit levels that return the taxes they paid in fewer years than average retirees do, while also receiving benefits for more years than the average retiree.
Hispanics consequently are one of the groups for which Social Security is most beneficial. A recent Social Security Administration fact sheet notes that Hispanic American beneficiaries "on average receive a higher rate of return on taxes paid."* A recent analysis by the Deputy Chief Actuary of the Social Security Administration explains that "a somewhat higher rate of return for Hispanic Americans is to be expected, based on the higher life expectancy for Hispanic Americans, and the fact that Hispanic Americans have lower than average earnings."** Indeed, tables in Heritages own report show Social Security provides higher rates of return to Hispanics than to the general population, although Heritage fails to acknowledge this in the text of the report.***
In addition, a conclusion drawn by the General Accounting Office that "because Social Security provides a lifetime annuity that is indexed for inflation, it becomes an increasingly important source as retirees grow older and exhaust other income sources" is particularly true for Hispanics.**** By contrast, under individual accounts, Hispanics could, because of their greater life expectancy, face a greater-than-average risk both of having their accounts run out before they die and of losing a substantial amount of the purchasing power of the funds in their accounts to inflation as the years pass.____________________
* Social Security Administration, "Social Security and Hispanic-Americans: Some Basic Facts, July 1998."
** "Comments on Heritage Rates of Return for Hispanic Americans," Memorandum from Stephen C. Goss, Deputy Chief Actuary, to Harry C. Ballantyne, Chief Actuary, Social Security Administration, April 2, 1998.
*** William W. Beach and Gareth G. Davis, "Social Securitys Rate of Return," Heritage Foundation, January 1998, Table 1, p. 7.
****General Accounting Office, Social Security: Different Approaches for Addressing Program Solvency, July 1998, p. 78.
Assumptions that Overstate Returns from Individual Accounts
While understating rates of return for Social Security, Heritage overstates rates of return for individual accounts.
- Heritage failed to factor in transition costs the costs that workers would have to pay under a privatized system to continue financing the Social Security benefits of current retirees and near-retirees. If retirement benefits are privatized, the payroll taxes that are currently used to finance Social Security retirement benefits will instead be deposited in individual accounts. That will create a financing gap funds will be needed to fulfill the government's obligation to pay Social Security benefits to current retirees and those nearing retirement. As a result, workers would have to pay a new tax to continue financing the Social Security benefits of current and soon-to-be retirees.
As senior researcher Paul Yakoboski of the Employee Benefit Research Institute recently testified, "Because the current Social Security system is largely pay-as-you-go, most of what workers pay into the system funds today's benefits....[O]n top of paying current benefits, workers moving to a privatized system would have to pay "twice" for the benefits going to today's beneficiaries and again to their own [personal] accounts."(1) For this reason, the General Accounting Office has noted that if Social Security retirement benefits were privatized, "the [payroll] contributions needed to fund both current and future retirement liabilities would clearly be higher than those currently collected."(2) Heritage fails to factor these added tax payments into its calculations. That leads it to overstate the rate of return that individual accounts would provide.
This error is of particular importance in analyzing rates of return for Hispanics. A study conducted by the Employee Benefit Research Institute, a highly regarded research institution, incorporated transition costs into its calculations. It found that for workers who are 21 today and receive low wages, the rate of return would be lower under the individual accounts options it examined than under all options it examined to restore long-term balance to Social Security without individual accounts. The EBRI data are a further indication that Heritage's claim that Hispanics would fare better under an individual accounts system because they are disproportionately young is inaccurate.
Excerpts from the Social Security Actuaries' Analysis of
Heritage's Work on Hispanic Americans
"...In a subsequent statement by Heritage...it was suggested that Hispanic Americans could expect far lower rates of return from Social Security than would be possible from individual accounts. The statement compared the (erroneously) computed expected rates of return from Social Security to the expected return from an individual account invested either all in bonds or partly in stocks. But the comparison ignores both the cost of transition to an individual-account program, and the administrative expenses associated with individual accounts. Transition costs associated with the continuation of benefit payments for current OASDI beneficiaries and today's older workers would be substantial. Adding in these costs, which are excluded from the Heritage analysis, would greatly reduce the overall rate of return...under the Heritage proposal.
"Finally, the authors fail to note...that, by their own calculations, see page 7 of January 15 report, Hispanic Americans would be expected to receive a substantially higher rate of return from Social Security than would the general population, on average."*
* "Comments on Heritage Rates of Return for Hispanic Americans," Memorandum from Stephen C. Goss, Deputy Chief actuary, to Harry C. Ballantyne, Chief Actuary, April 2, 1998.
- The Heritage estimates fail to take into account administrative costs, the fees that investment companies charge on individual accounts, and the costs of converting accounts to annuities, an omission that results in a substantial overstatement of the rates of return for individual accounts. Heritage compared the rate of return for Social Security to the rate of return for private accounts like IRAs. Use of IRAs results in significant administrative costs and management fees, which are paid out of the proceeds in the IRA accounts and consequently reduce the amounts available in those accounts to pay retirement benefits. Moreover, additional costs are incurred when the funds in these accounts are converted to lifetime annuities upon retirement.
Heritage ignores these costs. Its estimates of the rate of return that individual accounts would generate rest on the invalid assumption that no such costs would be incurred.
Based on data on IRA and 401(k) accounts, two eminent Social Security experts Henry Aaron of the Brookings Institution and Peter Diamond of M.I.T. have estimated that the administrative costs for retirement accounts like IRAs or 401(k)s would consume 20 percent of the amounts that otherwise would be available in these accounts to pay retirement benefits.(3) They note that a one percent annual charge on funds in such accounts eats up, over a 40-year work career, 20 percent of the funds in the accounts.(4) The 1994-1996 Advisory Council on Social Security estimated an annual charge of one percent on the assets in privately managed individual accounts.
Furthermore, recent financial data indicate that a one percent annual charge is a conservative estimate. In 1997, the average annual charge on stock mutual funds was 1.2 percent of the amounts invested in those funds. In addition, Diamond has noted that administrative and management costs consume approximately 20 percent of the amounts in individual accounts in Chile's privatized retirement system and more than 20 percent of the funds in privatized retirement accounts in Great Britain and Argentina.
Some of these costs are fixed-dollar expenses that do not vary with the size of an account. As a result, such costs would generally consume a larger percentage of the amounts in smaller-than-average accounts (and a smaller percentage of the amounts in large accounts). This suggests these costs would, on average, consume more than 20 percent of the funds in the accounts of lower-wage workers. That is of particular significance to minorities since, as a group, they receive lower-than-average wages and would consequently have smaller-than-average accounts.
To these costs must be added the costs of converting an individual account to an annuity upon retirement. The leading research on this matter indicates that an additional 15 percent to 20 percent of the value of an individual account is consumed by the costs that private firms charge for converting accounts to annuities.(5) The General Accounting Office recently noted that "While individual annuities are available, they can be costly especially relative to annuities provided through Social Security."
(Heritage may envision that account-holders would not have to convert the funds in their accounts to annuities when they retire. Lack of annuitization, however, would create large risks that substantial numbers of retirees would eventually exhaust their accounts and be thrust into poverty in very old age. Lack of a lifetime annuity would be a particularly severe problem for elderly widows and for Hispanic Americans, since they have longer-than-average life expectancies.)
Taking all of these costs into account both administrative and management fees and the costs of converting accounts to annuities Aaron estimates that at least 30 percent and as much as 50 percent of the amounts amassed in individual accounts similar to IRAs or 401(k)s would be consumed by these costs rather than being available to provide retirement income. The Heritage reports essentially assume that none of the amounts in these accounts would go for such costs.
- Heritage ignores the variation in rates of return that would occur with individual accounts and the likelihood that low-income workers and minorities would receive below-average rates of return. Retirees who are particularly lucky or wise in their investments could receive retirement income from individual accounts that more than offsets their loss of Social Security benefits. But retirees who are less lucky or wise, including those who retire and convert their account to a lifetime annuity in a year the stock market is down, would likely face large reductions in the income they have to live on in their declining years.
A recent GAO report takes note of these issues. "There is a much greater risk for significant deterioration of an individual's 'nest egg' under a system of individual accounts," the GAO wrote. "Not only would individuals bear the risk that market returns would fall overall but also that their own investments would perform poorly even if the market, as a whole, did well.(6)
This is a concern for workers in general surveys have found Americans are not very knowledgeable about financial markets and a particular concern for lower-wage workers, who generally would not be able to afford as good investment advice as individuals at higher income levels. Moreover, lower-income groups have less investment experience and would be more likely to invest in an overly conservative manner because they could not afford to expose the funds in their accounts to much risk. African Americans and Hispanic Americans make up disproportionate shares of the low-income population. As a result, they would be likely to receive a somewhat lower-than-average return on amounts invested even while, as explained above, they would likely pay an above-average percentage of their holdings in fees. Heritage overlooks these factors.
To persuade the public that Social Security should be replaced with individual accounts, Heritage has targeted the African-American and Hispanic-American communities and portrayed them as winners under such a conversion. Yet Heritage's rate-of-return calculations do not withstand scrutiny. These calculations significantly understate the rate of return to Social Security and significantly overstate the return to an individual accounts program.
When these problems are corrected, African-American and Hispanic-American communities no longer are the big winners Heritage portrays them as being under privatization of Social Security retirement benefits. They would be more likely to come out behind than ahead. The largest winners under Heritage's proposal for IRA-type individual accounts would more likely be financial investment companies, which could reap administrative and management fees running into the tens of billions of dollars on the accounts of nearly 150 million workers.
II. Assessment of the Heritage Reports
The authors of the Heritage reports assert that the retirement component of the Old Age Survivors and Disability Insurance (OASDI) program, or Social Security, should be replaced with private individual accounts that are invested in stocks and bonds. The basis for their assertion is that rates of return from the private market would be greater than those for Social Security. The rate of return reflects the difference between the value of money paid into a retirement or insurance program over time and the value of benefits received from the program over time. The higher the rate of return, the greater the amount of money an individual will receive compared to what the individual paid in.
Heritage compares the payroll taxes that individuals contribute to Social Security during their work lives to the benefits they receive after retirement. The Heritage authors also look at the funds individuals would get back if their payroll taxes were paid into individual accounts modeled on IRAs, instead of Social Security, and invested in the private market.
The authors argue that rates of return are lower under Social Security than they would be under individual accounts for the population in general and are particularly low for African-American workers and their families. According to Heritage, the higher mortality rate for African Americans compared to whites, especially for African-American men, will result in fewer years of benefits after retirement and consequently lower rates of return. In addition, the Heritage report on Hispanic Americans states that because Hispanics constitute a relatively young population, they will bear a disproportionately larger share of the cost of funding Social Security in the future.
To determine rates of return, the Heritage authors used a model that calculates the benefits of a hypothetical individual or family. The accuracy of any such model depends on the validity of the assumptions built into it. The Heritage model suffers from deeply flawed assumptions. The faulty assumptions are listed below and then examined one by one in the remainder of this analysis.
Assumption 1: All workers will live to be 65, and the age at which they are expected to die does not increase as they grow older. Assumption 2: All workers will remain in the labor force until the age at which they can draw full Social Security benefits; none will retire earlier and draw actuarially reduced Social Security benefits. Assumption 3: Converting to a private individual accounts system will incur no transition costs. Assumption 4: Administrative costs, fees charged by investment companies, and the cost of converting accounts to annuities can be omitted from projections on rates of return. Assumption 5: Social Security can be assessed based solely on its retirement provisions. Assumption 6: It is not necessary to adjust life expectancy by income level when examining differences in longevity by race. Assumption 7: The long-term actuarial balance for Social Security will be restored solely by raising payroll taxes and this hike will begin in 2015, when the trust fund has a several trillion dollar surplus.
Assumption 1: All workers will live to be 65 and the age until which an individual is expected to live does not increase as the individual grows older.
The Social Security Administration's Office of the Chief Actuary has analyzed the approach used by Heritage to estimate rates of return for Social Security and found major errors that led the Heritage authors to underestimate significantly the rates of return for workers in general and for African Americans in particular.(7) The Heritage method inaccurately estimates both the number of years an individual is expected to work and the number of years the individual is expected to collect Social Security retirement benefits.
Heritage used in its calculations the Life Tables developed by the Department of Health and Human Services. These tables show life expectancy to be 54.3 more years for a 20-year-old white man and 47.2 more years for a 20-year-old African-American man.(8) Under the Heritage methodology, both such men are assumed to work for 45 more years and to retire at the age at which Social Security pays full benefits, currently 65.
Such an assumption, however, fails to take into account the fact that some of these men will die before age 65; Heritage essentially assumed these men have a 100 percent chance of reaching 65. The Social Security Administration actuaries note that the probability of the men living this long is 78.1 percent for a 20-year-old white man and 60.3 percent for an African-American man.
The SSA actuaries determined that the true expected number of work years averages 42.2 years for the 20-year-old white man and 39.1 years for the African-American man, rather than the 45 years Heritage assumed. By overestimating the likelihood of living until 65, the Heritage authors overestimated the average number of years these individuals will make payroll tax contributions. Heritage thus overstated the Social Security payroll taxes these individuals will pay, with the overstatement being greatest for African Americans.
Expected Work and Retirement Years for 20-Year-Old Males in 1997
Years Working and Paying Taxes
African- American Men
African- American Men
Source: Steve Goss, "Problems with 'Social Security's Rate of Return,' A Report of the Heritage Center for Data Analysis," January 27, 1998, p. 2.
At the same time, Heritage underestimated the average numbers of years these workers will draw Social Security retirement benefits because Heritage failed to adjust the expected age of death for individuals who do survive to 65.(9) Continuing with the previous example, a man who is 20 years of age is expected to live to 74.3 if he is white and to 67.2 if he is African American, given their life expectancies at age 20 of 54.3 additional years and 47.2 additional years, respectively. The Heritage methodology assumes that white men who work until 65 will receive 9.3 years of retirement benefits (from age 65 to 74.3), while African-American men who work until 65 will receive just 2.2 years of retirement benefits (from 65 to 67.2 years of age). This is flatly wrong. The age that an individual is expected to attain changes over his or her lifetime. If a man reaches age 65, he is expected to live 15.5 additional years if he is white and 13.5 more years if he is African-American. Heritage thus heavily underestimated the number of years that these men, and especially African-American men, will draw Social Security retirement benefits.
Because of these and other flaws, the actuaries found that the Heritage methodology "... grossly underestimates the expected rates of return from Social Security retirement benefits. Moreover, the extent of this bias is greater for workers with lower life expectancy, in particular African Americans. The error is also greater for younger workers who are far from retirement age in 1997."(10)
An appropriate method for estimating the expected number of years of retirement for a 20-year-old man must incorporate both the probability of surviving to 65 and the life expectancy for those who reach 65. Since a 20-year-old white man has a 78.1 percent chance of reaching 65, his expected number of retirement years is 12.1 (that is, 0.781 times the 15.5 additional years such a man is expected to live if he reaches 65). The 20-year-old African-American man who has a 60.3 percent probability of surviving until 65 is expected to have 8.1 retirement years (0.603 times 13.5).
The table above compares work years and retirement years under the Heritage methodology with the corrected values determined by the SSA actuaries. The actuaries' findings result in nearly four times as many estimated years of retirement benefits for young African-American men as the Heritage methodology assumes. Had Heritage used the appropriate methodology for computing work years and retirement years, the rates of return that Social Security provides for men, especially African-American men, would be shown to be considerably higher than Heritage claims.
Another way of examining the data is to compare the number of work years to the number of retirement years for both African-American men and white men. The actuaries' calculations show that the number of years of retirement for 20-year-old African-American men equals 20.7 percent of their working years. For 20-year-old white men, the retirement period equals 28.7 percent of their work years. The Heritage methodology yields distorted results, with the assumed retirement period for 20-year-old African-American men equaling only 4.9 percent of their work years and the assumed retirement period for 20-year-old white men equaling 20.6 percent of their work years.
Because Heritage overestimates the number of years for which men, especially African-American men, contribute payroll taxes while substantially underestimating the number of years for which these men and again, especially African-American men receive Social Security benefits, it produces large underestimates of the rates of return that Social Security provides. (Note: This problem also results in an underestimate by Heritage in the rates of return that Social Security provides to Hispanic-Americans, although the underestimate is greatest for African Americans. The Social Security actuaries did not compute the specific degree of error contained in the Heritage estimates for Hispanics.)(11)
Assumption 2: All workers will remain in the labor force until the age at which they can draw full, rather than reduced, Social Security benefits.
The rates of return that Heritage has estimated for Social Security reflect an assumption that each worker will remain in the labor force and contribute payroll taxes until the age at which workers may retire and draw full Social Security benefits. (This age, sometimes mislabeled the "normal retirement age," is currently 65.(10)) This Heritage assumption fails to take into account that about two-thirds of all workers, including minority workers, take early retirement. Having the option to receive Social Security benefits before 65 is particularly important to African Americans given their shorter average life span.
The Social Security system allows workers either to retire with full benefits at a given age, currently 65, or to retire early with reduced benefits. A worker can take early retirement at age 62. Workers who retire at 62 pay payroll taxes for three fewer years. They also begin receiving benefits three years earlier, with their monthly benefit level permanently reduced to compensate for the increased number of years during which they will receive benefits.
The reduction in the monthly benefit amount for those who retire early is based on actuarial tables and is intended to make the amount of benefits received from age 62 to the point of death equivalent, on average, to the amount of benefits retirees would receive if they waited until the "normal retirement age" to retire. Over the population as a whole, the Social Security early retirement option is close to a wash the lower monthly benefits paid are designed to offset the increased number of years for which benefits will be received. (On average, individuals do slightly less well by retiring early.)
The story is different, however, for African Americans. Given the shorter average life span of African Americans, the benefits that African-American early retirees receive from age 62 to the end of their lives exceed the benefits they would receive, as a group, if they waited until 65 to retire. Starting to receive benefits several years earlier increases the total benefits they receive and raises their average rate of return.
The Heritage calculations overlook this. These calculations understate the Social Security retirement benefits that African-American workers are expected to receive because Heritage assumes African-American workers do not start drawing benefits until age 65.
Elements of the Social Security Benefit Structure Favorable to
Low-income Workers are of Importance to Minorities
Low-income people tend to have a higher mortality rate than high-income people. This factor would result in lower rates of return from Social Security retirement benefits for low-income workers if no other factors came into play. But other factors do come into play. Social Security has a highly progressive benefit formula. Workers with relatively low earnings receive a much higher proportion of their wages as a retirement benefit than high-wage earners do. As a result, low-wage workers get back their payroll tax contributions in substantially fewer years than high-wage earners do.
This is of particular importance to minorities. Since African Americans and Hispanic Americans make up a disproportionately large share of low-wage workers (and a disproportionately small share of highly paid workers), the Social Security benefits they receive tend to return their payroll tax contributions in fewer years than is true, on average, for non-minorities.
A study by James Duggan, Robert Gillingham, and John Greenlees of the Treasury Department used data from 1951 to 1988 on actual workers and retirees (rather than the hypothetical cases Heritage constructed) to estimate rates of return for Social Security retirement and survivors benefits by race. The study reflected both the shorter life expectancies of African Americans and the progressivity of the Social Security benefit structure. Contrary to Heritages claim that Social Security provides lower rates of return to African Americans than to whites, the researchers found African-American workers received slightly higher real rates of return than white workers.* The study by Duggan and his colleagues is generally regarded as the most significant study on this issue because it is the only study of rates of return by race that is based on data on actual Social Security earnings and benefit histories. (No other such studies are based on these data because researchers outside the Social Security Administration and the Treasury are not allowed access to these data for confidentiality reasons.)It also may be noted that the Earned Income Tax Credit, a refundable credit for working families and single individuals with low or moderate wages, offsets some or all of the payroll taxes that many low-income workers pay. This credit was established in 1975 primarily to offset the payroll tax burdens of low-income workers. In 1998, low-income workers with one child can receive an Earned Income Tax Credit equal to 34 percent of earnings on up to $6,680 of their wages, for a maximum credit of $2,271, while workers with more than one child may claim a credit equaling 40 percent of earnings on up to $9,390 of their wages, for a maximum credit of $3,756. Poor workers without children qualify for a credit equal to 7.65 percent of earnings on up to $4,460 of their wages. If the portion of the EITC that offsets Social Security payroll taxes is taken into account in calculating rates of return for low-wage workers, these rates rise significantly. The Heritage reports do not take the EITC into account.
* James E. Duggan, Robert Gillingham, and John S. Greenlees, "Returns Paid to Early Social Security Cohorts," Contemporary Policy Issues, October 1993.
Assumption 3: Converting to a private individual accounts system will incur no transition costs.
Social Security is largely a "pay-as-you-go" system; the payroll taxes of workers currently in the labor force pay the benefits of current retirees. Under the privatized system that Heritage proposes as an alternative, payroll taxes from workers would flow into their own individual accounts and would not be available to pay retirement benefits for current retirees. Yet funds would have to be made available to honor the promise of benefits to current retirees and workers near retirement who have paid into the system throughout their working lives.
These funds would presumably be generated by raising payroll taxes. The 1994-1996 Social Security Advisory Council split into three factions when making recommendations for restoring long-term balance to Social Security. The faction most in favor of privatization proposed diverting five percentage points of the payroll tax from the Social Security trust funds to individual accounts. The members of this faction of the Advisory Council identified the transition costs that would result and proposed that to cover these costs, the payroll tax be increased by 1.52 percentage points for 72 years. (Even with this increase in the payroll tax, the federal government would need to borrow an additional $1.9 trillion over the next 40 years to help cover the transition costs under this proposal. These borrowed amounts would be repaid between 2035 and 2070.)
A study by the respected Employee Benefit Research Institute (EBRI) also estimated the transition cost of redirecting five percent of the payroll tax to individual accounts.(13) (The option EBRI examined differs in some respects from the privatization option these members of the Advisory Council proposed.) EBRI estimated that the transition costs would require an increase in contributions equal to five percent of payroll over a 40-year period or a 3.3 percent payroll tax increase over a 70-year period.
EBRI appropriately took these transition taxes into account when computing rates of return for privatization proposals, something Heritage failed to do. The Heritage calculations essentially assume that the retirement component of Social Security can be fully privatized without incurring transition costs, a virtual impossibility.(14)
EBRI found that when transition costs are included, rates of return are lower for various groups including low-income workers now 22 years of age under individual accounts than under any of the options it examined for maintaining Social Security. The options it modeled for maintaining Social Security included restoring actuarial balance through benefit reductions only, payroll tax increases only, and a combination of the two.
This finding is of particular significance for Hispanic Americans. Heritage's report on Hispanic Americans contends that because this community is disproportionately young, it would be unduly burdened by an increase in the payroll tax to sustain Social Security and would fare better under individual accounts. Part of Heritage's argument rests, however, on Heritage's failure to take into account the transition costs to a privatized system, which would be borne disproportionately by younger populations. The EBRI findings suggest that the rate of return to young, low-wage Hispanics under an individual accounts system would likely be lower than the rate of return from maintaining Social Security either by raising Social Security payroll taxes and/or by cutting Social Security benefits.
Furthermore, if part of the long-term imbalance in Social Security finances is closed by investing a portion of Social Security reserves in equities rather than by raising taxes or cutting benefits, the rate of return that the Heritage individual accounts alternative would generate for young populations would compare even more poorly to Social Security. Investing a portion of Social Security reserves in equities is a course that former Congressional Budget Office director Robert Reischauer, Brookings senior fellow Henry Aaron, and former Social Security commissioner Robert Ball, among others, have recommended.
Assumption 4: Administrative costs, fees charged by investment companies, and the cost of converting accounts to annuities can be omitted from calculations of rates for return for individual accounts.
When computing rates of return for individual accounts, the authors of the Heritage reports fail to take into account the fees that investment companies charge to cover their administrative costs, the companies' management fees, and the fees charged for converting individual accounts into lifetime annuities when a worker retires. These costs reflect the expenses that investment companies incur in advertising and undertaking other promotions to attract investors, in communicating with investors, and in record-keeping, as well as the profits the companies make.
Social Security pays no commissions, and it charges no fees for converting funds to annuities. Its administrative costs are extremely low, in large part because of the economies-of-scale achieved by having a single agency manage accounts for 147 million workers and 44 million beneficiaries.(15) In addition, Social Security does not incur costs in advertising to attract customers and does not take a portion of the funds it handles for profits for private shareholders.
This much-greater administrative costs incurred under a privatized system involving numerous financial firms must be paid from funds in the individual accounts. As noted earlier, Peter Diamond and Henry Aaron have pointed out that a one percent annual charge on funds in an individual account similar to an IRA or a 401(k) account the type of individual account Heritage favors would, over a 40-year work career, consume approximately 20 percent of the funds in the retirement account. Based on the most recent financial data on mutual funds, a one percent annual charge is a conservative estimate. Lipper Analytical Services reports that in 1997, the average charge on no-load stock funds equaled 1.21 percent of the amounts invested in these funds, while the average charge on front-load stock funds equaled 1.25 percent of the amounts invested, in addition to the load charge.(16)
Individual Accounts May Yield Large Profits for Financial Securities Industry
In testimony June 18 before the Subcommittee on Social Security of the House Ways and Means Committee, Ric Edelman, head of Edelman Financial Services, commented on the large fees, costs, and profits that would be associated with individual accounts. Edelman stated:
"If these massive investments are handled on an individual level, five major industry sectors will enjoy tremendous profits...
a) First, the money management industry will earn enormous fees by investing and managing these assets. I am referring to mutual funds, annuities, institutional money managers, clearing firms, banks, insurance companies and brokerage firms....
b) Financial advisors would become the nations hottest new profession (if it isnt already). There will be incredible fee and commission income from advising consumers on how to invest their Social Security assets...If you enact this proposal, let me be the first to thank you in advance for helping me and all financial advisors to become rich, or rather, even richer.
c) The burgeoning financial education field, of which I am a part, and which currently is a small cottage industry, would become a major metropolis. Tens of millions of workers would be new targets for educational products and services.
d) The financial media, which is very well represented today by Money Magazine, would find a treasure chest of new information to convey to its readers. Theyll have plenty to write about for a long time.
e) The advertising industry would receive its biggest bonus since prescription drug advertising hit television. Whereas currently Wall Street attempts to reach only the affluent, who have assets to invest, this proposal would place investable assets into the hands of virtually every working American. Wall Street will want to reach those consumers, and this means unmatched spending on advertising, marketing, promotional and public relations campaigns...."
The amounts consumed would generally be greater than 20 percent of the amounts in smaller-than-average accounts, the accounts that workers with below-average wages would have. That would include the majority of African-American and Hispanic-American workers.
A sizeable additional cost is incurred when a worker converts the funds in his or her account to a lifetime annuity at the time the worker retires. (An annuity is a payment generated for the remainder of the beneficiary's life, regardless of how long the individual lives.) An insurance company selling an annuity must cover its overhead, advertising and promotional costs, commissions to agents, and profit for shareholders.
It also must cover itself against the risk of "adverse selection" the risk that the lifetime benefits the company pays to people who purchase annuities will exceed the amounts the company charged for the annuities if many of the beneficiaries live to very old ages and receive annuity benefits from the company for many years. This is a particular risk for firms selling annuities, because the people most likely to purchase annuities are people who are more likely than the average retiree to live to a quite old age. Firms cover themselves against losses of this nature by pushing up the price of the annuities they sell.
The leading research in the field indicates that the various costs and charges involved in converting private accounts to annuities eat up 15 percent to 20 percent of the value of the accounts.(17)
The costs charged for annuities can vary considerably across insurance firms. A retiree would need to be fairly astute in selecting an insurance company from which to purchase an annuity to obtain the best value for his or her investment. Relatively few families, however, have experience with buying annuities. The 1995 Survey of
Consumer Finances by the Federal Reserve Board shows that only 3.8 percent of all families have assets such as personal annuities and trusts, and only one percent of nonwhite and Hispanic families have them.(18) African-American and Hispanic-American families, especially those with low incomes, may be among those who, under an individual accounts system, would be most likely to incur costs for annuities that extract a higher-than-average percentage of the funds in their accounts. That would further reduce the rate of return on individual accounts for these communities.(19)
Financial Securities Analyst Explains That Administrative
Costs Can Substantially Reduce Returns From Accounts
In testimony July 24 before the Subcommittee on Finance and Hazardous Materials of the House Commerce Committee, Joel M. Dickson, Senior Investment Analyst with the Vanguard Group, explained the large role that administrative costs can play in lowering rates of return that individual accounts can provide, especially for the smaller-than-average accounts of workers with below-average earnings. Dickson stated:
"Much of the recent debate on individual accounts has centered around how Social Security participants can earn higher rates of return like those experienced in the private financial markets. Unfortunately, there is often a large gap between the assumed market return on an investment and the actual return realized by investors. The reason? Cost... What may seem to be a relatively small cost difference over a short period of time can become an enormous difference over a long investment horizon....
"Consider a $1,000 investment in two equity funds; one earns a market return of 10%; the second earns only 8%, which represents the markets return reduced by an all-too-realistic total cost of 2% (arising from charged expenses and portfolio transaction costs). After the first year, the difference in account values between the two options is only $20. Of course, a participant investing over the course of their working lifetime must consider much longer time horizons. Over 40 years, the first fund soars to $45,000. The second fund, earning 2% less per year, grows to $22,000 less than half what the market earned....
"The key pricing variable for private accounts is the average account size. To the extent that costs of servicing the accounts could not be covered by asset-based management fees, other charges like account maintenance fees are likely. One of the criticisms often made about privately-managed, individual accounts is that lower-income workers may not be as financially savvy, resulting in lower average returns. I think a bigger issue is that lower-income and part-time workers will have small account balances, potentially subjecting them to higher fees because asset-based revenue cannot begin to cover the costs of maintaining these accounts. Thus, they are likely to earn lower returns than higher-income workers that make the same investment choices for reasons unrelated to financial market sophistication" [emphasis added].
Assumption 5: Social Security can be assessed based solely on its retirement provisions.
The Heritage reports do not calculate the rate of return on the full Social Security system. They attempt to isolate the retirement portion of Social Security by estimating both the payroll tax contributions and the Social Security benefits associated only with retirement. They then compare the retirement component of the Social Security system to private individual accounts.
Social Security is not designed, however, to be solely a retirement program. It is a comprehensive insurance system that protects against risks which are unforeseen or for which workers are not sufficiently prepared. Social Security does not simply cover a worker's retirement; it also provides benefits to the worker's spouse and dependents when the worker retires or becomes disabled, as well as survivors benefits if the worker dies. The divorced spouse of a retired or deceased worker also is generally entitled to benefits.
African Americans benefit disproportionately from the disability and survivors components of Social Security. By removing these parts of Social Security from its calculations, Heritage disregards the elements of Social Security most favorable to them.
Furthermore, Heritage removes survivors insurance from its rate-of-return calculations in a questionable manner that makes the rate of return for Social Security retirement benefits appear lower than it is. In computing the rate of return for Social Security retirement benefits, Heritage disregards the portion of the payroll tax that it claims to be the part that supports benefits for survivors of workers who die before retirement. Heritage compares the remaining part of the Old Age and Survivors Insurance payroll tax to the retirement benefits that workers would receive.
In making these calculations, however, Heritage attributes too much of the payroll tax that African Americans and Hispanic Americans pay to retirement benefits and too little to survivors benefits. In addition, while Heritage claims its individual accounts option preserves Social Security disability benefits, it fails to take into account that replacing Social Security retirement benefits with individual accounts would make it more difficult to maintain Social Security disability benefits without raising payroll taxes or cutting disability benefits sharply. It also fails to take into account the potential under its plan for many spouses of retired workers and many divorced women to be faced with sharp reductions in retirement income.
One of every seven Social Security beneficiaries is a beneficiary of Social Security disability benefits. African Americans make up a disproportionate number of these participants.(20) While African Americans account for 11 percent of the civilian labor force,(21) they comprised 18 percent of the workers receiving Social Security disability benefits in 1996.
When a worker becomes disabled, the workers' dependents also become eligible for Social Security benefits. African Americans made up 23 percent of the children and 15 percent of the spouses who received Social Security benefits in 1996 because workers in their families were disabled. Studies by the Social Security Administration show that African American receive substantially more benefits from Social Security Disability Insurance in relation to the taxes they have paid than whites do.(22) Leaving disability benefits out of Social Security rate-of-return calculations, as Heritage does, understates the value of Social Security to the African-American community.
Furthermore, Heritage never explains how, if Social Security retirement benefits are privatized, Social Security disability insurance can remain unchanged unless payroll taxes are raised or disability benefits cut. The Heritage reports gloss over the fact that Social Security disability insurance faces more immediate funding problems than the retirement component of Social Security. Looked at in combination, the Social Security trust funds will remain solvent until 2032, but the Disability Insurance trust fund on its own will be exhausted in 2019. Addressing the financing problems in Social Security disability insurance are easier if the full Social Security system is maintained, because funds can be reallocated from the Social Security Old Age and Survivors Insurance trust fund to the disability insurance trust fund if necessary. If the revenue for the retirement part of Social Security has been diverted to individual accounts, however, it will not be available to help sustain the disability insurance program.
Aggravating this problem, if only the disability and survivors parts of Social Security remain, Social Security administrative costs will rise as a percentage of benefits paid. This is true both because economies-of-scale will be reduced and because the costs of administering disability insurance are three times as high, measured as a percentage of benefits paid, as the administrative costs of Social Security as a whole.(23)
For these reasons, the benefits provided under Social Security disability insurance would likely need to be cut back more sharply under the Heritage approach than under a reform plan that maintains the payroll tax levels now devoted to Social Security retirement benefits. To maintain adequate coverage in the face of such reductions, workers might need to supplement their scaled-back Social Security disability insurance by purchasing private disability insurance if they could afford it.
If workers needed to purchase private disability insurance, those with the lowest incomes would, on average, be likely to face the highest premiums if they qualified for such insurance at all. Disability insurance premiums are based in part on the occupation and health of the applicant.(24) The greater the physical risk of the occupation and the less healthy the worker, the greater the probability of not qualifying for coverage and the higher the premium for those who do qualify. Workers with lower wages are more likely to be employed in jobs that require physical exertion.(25) In addition, low-income earners have a greater propensity for poor health since they generally have fewer resources with which to obtain health care. As a result, some low-income workers could find themselves unable to secure coverage or unable to obtain coverage they could afford.
The likely reductions in Social Security disability insurance under the Heritage approach would be particularly detrimental to African Americans. This would be true both because of their greater use of Social Security disability benefits and because, as a population with a disproportionate share of low-wage workers, African Americans could face higher-than-average costs for purchasing private disability insurance to supplement their reduced Social Security disability benefits. Hispanic Americans could face higher costs for private disability insurance for the same reason.
Social Security survivors benefits go to the spouse, the children, and sometimes the dependent grandchildren and parents of a deceased worker. An elderly widow, widower, or divorced spouse of a deceased worker can, for the rest of his or her life, receive a survivors benefit equal to 72 percent to 100 percent of what the deceased worker's monthly benefit would have been.(26) A non-elderly widow, widower, or divorced spouse receives 75 percent of the deceased worker's benefit if he or she cares for the deceased worker's child. Eligible children and grandchildren also receive monthly checks equal to 75 percent of the deceased worker's benefit. So do the parents of the worker if they are caring for the deceased worker's children.(27)
In computing Social Security rates of return, Heritage attempted to isolate the retirement portion of Social Security from the survivors portion. As with the removal of disability insurance from the computations, this lowers rates of return for African Americans. As a result of above-average mortality rates among African Americans, the African-American community benefits disproportionately from the feature of Social Security that provides benefit payments to non-elderly survivors.
For example, although African-American children comprise about 16 percent of all children in the United States,(28) they made up 24 percent of the children receiving survivors benefits in 1996.(29) African-Americans also accounted for 21 percent of the spouses with children who received survivors benefits. Benefits for non-aged survivors are one of the aspects of Social Security most favorable to African-Americans and low-income workers.
Furthermore, Heritage used a highly questionable methodology in removing the survivors portion of Social Security from its calculations. Removing survivors insurance from calculations of Social Security's rate of return necessitates determining what portion of the combined Old Age and Survivors Insurance (OASI) payroll tax goes for survivors insurance so that this part of the payroll tax may be removed from the computations.(30) There is no ready way, however, to make such a determination; there are no separate payroll tax components for retirement and survivors benefits.
To compute rates of return for Social Security retirement benefits, Heritage subtracted the portion of the OASI tax it claims is used for benefit payments to the survivors of workers who die before retirement. But the methodology Heritage used in allocating the payroll tax into survivors and retirement components is flawed; it allocates to survivors insurance too little of the payroll taxes that African Americans and Hispanic Americans pay, while allocating an overly large share of these payroll taxes to Social Security retirement benefits. Overstating the amount of Social Security taxes said to go for retirement benefits makes the rate of return on these benefits look lower than it actually is.
To determine what share of the payroll tax goes for survivors benefits, Heritage identified a low-cost term life insurance premium offered by one life insurance company. Heritage simply assumed that the cost of this premium represents the portion of the payroll tax that goes for survivors benefits.
The cost of a life insurance policy varies greatly, however, depending on an applicant's health. A term life insurance policy worth $200,000 would cost a 38-year-old man in good health approximately $225 to $350 per year,(31) while the same individual in poor health could pay from $900 to several thousand dollars per year or be uninsurable. Heritage based its estimate of the portion of the payroll tax that goes for survivors benefits on the cost of life insurance premiums for healthy workers. As a result, the estimated cost of a life insurance policy that Heritage used to determine the portion of the Old Age and Survivors Insurance payroll tax that goes for survivors benefits is too low for the population as a whole.
The problem is most serious in computing rates of return for low-income workers, since they are more likely than those at higher income levels to be in poor health and thus to incur larger premium costs when they seek to purchase life insurance (if they qualify for insurance at all some in poor health may be unable to secure a policy). Because minorities are disproportionately represented among the lower-income population, as a group they are likely to incur above-average costs for life insurance.
The Heritage assumption that the portion of the payroll tax going for survivors benefits equals the cost of a low-cost life insurance premium for healthy workers consequently is particularly problematic for minorities. It results in an understatement of the portion of payroll taxes paid by minority workers that go for survivors benefits and a corresponding overstatement of the payroll taxes these workers pay for retirement benefits. This is another factor that leads Heritage to underestimate the rate of return that African American and Hispanic Americans receive on Social Security retirement benefits.(32)
The Heritage Analysis and Widows of Retired Workers
Under the Social Security system, survivors benefits also go to survivors of retired workers who die; such benefits go primarily to elderly widows and some members of their families. The Heritage reports are vague about how such widows would be taken care of under the individual accounts that, under the Heritage approach, would replace not only Social Security retirement benefits but also the benefits of survivors of workers who die after retiring.
The Heritage reports suggest a retired worker could purchase an annuity with the accumulated savings in his or her individual account. Although it is possible to purchase a joint annuity for two people that would be paid until both have died, Heritage does not state whether this would be mandatory under its proposal. If no restrictions applied and a married retiree could purchase a one-person annuity, then a spouse who outlived the retired worker to whom she was married would face the risk of receiving no income from the annuity after her husband's death.
Moreover, if conversion of individual accounts to annuities at retirement is not mandatory and nothing in the Heritage reports suggests it would be many retirees could largely exhaust their accounts by the time they die, leaving little for their surviving widows. The risk of increased impoverishment of elderly widows is one of the principal concerns posed by plans to replace Social Security retirement benefits in full or in part with individual accounts.
Another issue the Heritage reports do not address is that the privatization alternative Heritage advances could lead to significant losses in income for many spouses and divorced women. Social Security benefits increase if a beneficiary has a spouse or dependents. The spouse of a retired worker can receive a Social Security benefit equal to 50 percent of the worker's benefit,(33) with the result that the couple can receive a total benefit equal to 150 percent of the worker's benefit. If the worker has dependent children, the children also are eligible for a benefit equal to 50 percent of the retired worker's benefit, up to a maximum amount for the family.
These benefit features would be hard to replicate under a plan where retirement benefits are limited to what can be paid from an individual's account. As the General Accounting Office has noted: " With respect to dependents' benefits, individual account proposals imply reduced spousal benefits."(34)
There also is the question of joint annuities. The Heritage reports indicate that under the Heritage individual accounts plan, a worker could, upon retirement, purchase an annuity with the savings in his or her individual account. But in contrast to Social Security, the amount of the monthly benefit payment for a joint annuity would necessarily be lower than the amount of the monthly benefit payment for one person, rather than higher.(35) A joint annuity must be paid until both people die; therefore, the payout period is likely to be longer than for one person. To compensate for the longer period of coverage, monthly benefit payments are necessarily smaller for a joint annuity than for a one-person annuity.(36)
Heritage also does not explain what would happen to divorced spouses of retired workers under its individual accounts system. Under Social Security, a divorced spouse who was married to a worker for at least 10 years, is 62 or over, and has not remarried is entitled to the same Social Security benefits that a current spouse would receive. Furthermore, benefits to a divorced spouse are outside the family maximum for the worker; should the worker remarry, the benefits paid to the divorced spouse do not affect the benefits paid to the worker's or retiree's current spouse and family. Heritage glides over the question of whether a divorced spouse would receive any benefits from the worker under a system of individual accounts; if so, current spouses and dependents would have to receive lower benefits as a result (since, once again, the amount of the payments would be limited to what is in the accounts). Individual accounts may pose a risk of eliminating divorced spouses as a group of beneficiaries, which could result in a substantial increase in poverty for a substantial number of older women. None of these factors are considered in Heritage's rate-of-return estimates.
Assumption 6: It is not necessary to adjust life expectancy by income level when examining differences in longevity by race.
The Heritage report on African Americans contains examples of hypothetical workers to demonstrate differences between African-Americans and whites with respect to Social Security rates of return. In discussing the African-American community, the Heritage authors use a single, low-income man and a single low-income, 21-year-old mother. When discussing the white community, the authors select an upper-middle-income married couple with two children. Aside from the rather stereotypical nature of the characteristics selected for these hypothetical workers, the authors fail to separate out the impact of income on longevity (and hence on Social Security benefits) from the impact of race on longevity.
Although the Heritage authors acknowledge that life expectancy varies by income, they make no adjustments for these differences when estimating rates of return. In their critique of the Heritage methodology, the Social Security actuaries state: "Without appropriate adjustment of mortality by income, race-differentials in rates of return are highly misleading."(37)
Stated another way, failing to control for income that is, failing to compare mortality rates for whites and minorities with the same income levels exaggerates the impact of race on rates of return. When determining differences in longevity, Heritage failed to compare African-American upper-income, married couples with children to white upper-income, married couples with children. Similarly, it failed to compare single, low-income African-American men and women to single, low-income white men and women.
A study by Duggan, Gillingham, and Greenlees did control for income, among other factors, when calculating rates of return for Social Security retirement and survivors benefits for African-American and white workers.(38) The study used data from 1951 to 1988 on actual workers and retirees, rather than the hypothetical cases Heritage constructed. Contrary to Heritage's claims, the researchers found that African-American workers received slightly higher real rates of return than white workers.
In calculating rates of return for Social Security, assume that long-term actuarial balance will be restored solely by raising payroll taxes and that the tax hike will take effect in 2015.
The Social Security trustees project that over the next 75 years, total Social Security benefits paid will exceed total payroll taxes received by an amount equal to 2.19 percent of taxable payroll for this period.(39) The 1994-1996 Advisory Council issued three competing plans for eliminating the long-term imbalance. Recommendations contained in one or more of these plans included expanding coverage of Social Security to the 30 percent of state and local government workers not currently covered, investing a portion of Social Security assets in the private market, reducing benefits in various ways, and raising the payroll tax in future decades.(40) The Heritage studies are based on the assumption that if Social Security is not privatized, the only method used to correct the imbalance will be raising the payroll tax. Consequently, Heritage assumes too large an increase in payroll taxes in its calculations. The Heritage authors also assume the payroll tax hike will begin in 2015. This is a decade earlier than would be necessary.
Assuming Social Security's long-term deficit will be met solely through increases in payroll tax rates unrealistically drives up the amount of payroll contributions that workers would make and drives down their rates of return. No proposal by any Advisory Council members or other prominent proponents of Social Security reform has recommended correcting the imbalance purely through payroll tax increases.
Moreover, some of the other steps that have been recommended to close the imbalance, such as investing a portion of Social Security assets in the equities market and extending Social Security coverage to more state and local employees, would reduce the imbalance without lowering the typical retiree's rate of return. For example, the Social Security actuaries estimate that investing half of the trust fund in equities would cut the long-term deficit about in half. This would substantially reduce the need for tax increases. Some other recommended measures would trim benefits primarily from beneficiaries who receive large Social Security benefits; such measures would disproportionately affect beneficiaries with high incomes and hence would not lower rates of return appreciably for the typical minority beneficiary. Heritage stacked the deck in computing Social Security rates of return by assuming the single approach to closing the Social Security imbalance that would most adversely affect rates of return for minorities.
In addition, the Heritage authors assumed that payroll tax rates would be raised in 2015, earlier than the actuaries project would be necessary even under a "payroll-tax-increase-only" approach. Social Security is currently running a substantial surplus, with payroll tax revenue exceeding benefit payments. The actuaries estimate that in 2015, the trust fund balances will equal 300 percent of annual benefits, or $1.9 trillion.(41) No tax increase would be needed at that time. In its assessment of the Heritage report, the Office of the Chief Actuary noted: "In order to produce a pay-as-you-go financed Social Security program, tax rates would not need to be raised until around 2025, and by less than the [Heritage] authors assumed. [Heritage's assumption of an] early, too-large increase in tax rates results in rates of return that are estimated to be too low for Social Security." (42)
The Heritage analysis suffers from large flaws. It understates the rate of return for Social Security and overstates the returns under individual accounts. When the deficiencies in the analysis are corrected, the African-American and Hispanic-American communities no longer are the winners Heritage portrays them as being under these accounts. They are more likely to be injured than aided by the elimination of Social Security retirement benefits and their replacement with individual accounts.
February 4, 1998
From: Steve Goss, Deputy Chief Actuary
Social Security Administration
Subject: Problems with "Social Security's Rate of Return" A Report of the Heritage Center for Data Analysis
The analysis by William M. Beach and Gareth G. Davis of the Heritage Foundation attempts to provide an answer to the question "What can Americans expect in future Social Security retirement benefits?" The paper suggests that expected rates of return will decrease substantially in the future and that African Americans do now and will in the future have substantially lower expected rates of return that for the general population. However, the conclusions are highly misleading due to two major errors in the methodology, plus a number of incorrect or inappropriate assumptions.
Due to these errors, rates of return for the general population in this study are at a substantial variance with those produced for the 1994-6 Advisory Council on Social Security (See Appendix 2 of Volume 1 of the report, pages 219-22). For example, for single men (all races combined) born in 1970 with average earnings, real rates of return are estimated at only 0.47 percent by Heritage, compared with 1.28 percent in the Advisory Council Report.
More importantly, the methods utilized by the authors bias expected rates of return downward to a far greater degree for African Americans than for the general population. This study thus erroneously indicates differences in rates of return by race that are greatly exaggerated. In fact, results from more careful research reflecting actual work histories for workers by race indicate that the non-white population actually enjoys the same or better expected rates of return from Social Security that for the white population. (See Duggan, et. al., "The Returns Paid to Early Social Security Cohorts," Contemporary Policy Issues, (October), pp. 1-13.)
Error in Method of Computing Expectation
Expected values indicate the average of all possible outcomes for individuals at a point in time with an uncertain future. Instead, the Heritage study erroneously analyzes a single outcome where an individual is assumed to know how long he or she will live. For persons at each age 22 through 66 in 1997, the authors estimated the remaining life expectancy, i.e., the average number of years of life remaining after 1997. Individuals at each age were then assumed to live for precisely the number of years indicated by this average life expectancy. This approach consistently overestimates the expected number of years of work and consistently underestimates the expected number of years after reaching retirement age. As a result, it grossly underestimates the expected rates of return from Social Security retirement benefits. Moreover, the extent of this bias is greater for workers with lower life expectancy, in particular African Americans. The error is also greater for younger workers who are far from retirement age in 1997.
To illustrate this error, consider the expected years of work and retirement under the Heritage method in comparison with the true expected years for white and black men based on the 1992 United States Life Tables. Life expectancy at age 20 was 54.3 years for white men and 47.2 years for black men. Assuming both planned to work until retirement age at 65, the Heritage approach would assume 45 work years for each, and 9.3 retirement years for white men but only 2.2 years of retirement for black men.
However, the true expected number of work years for both black and white men must be less that 45, because some will die before reaching age 65. In fact, the true expected number of work years (reflecting deaths before 65) is 42.2 years for white men and 39.1 years for black men. Computing, for 20-year-old men the expected number of retirement years after 65 requires both the probability of surviving to 65 and the life expectancy at 65 for those who do survive. For white men age 20, 78.1 percent would expect to survive to 65 at which point they would average an additional 15.5 years of life after 65. Thus, the expected years of life for a 20-year-old white male after reaching 65 would be 12.1 years (= 15.5 x .781). For black men age 20, 60.3 percent would average an additional 13.5 years of life. Their expected retirement years would thus be 8.1 (= 13.5 x .603).
Expected Work and Retirement Years for 20-Year-Old Males in 1997 Work Years Retirement Years White Men Black Men White Men Black Men Heritage Erroneous
45 45 9.3 2.2 True Expectation 42.2 39.1 12.1 8.1
Clearly, computed rates of return for retirement benefits using true expectation will be much higher for all men, and, moreover, the difference between rates of return for black and white men will be dramatically smaller, than if the erroneous Heritage method is used. While this is the largest and most fundamental error in the Heritage analysis, an number of additional errors further bias the results.
Error in Not Reflecting Relationship Between Mortality and Income
The authors acknowledge the fact that life expectancy is highly correlated with income, and that their estimates do not reflect this fact. Estimates made by the Office of the Chief Actuary, Social Security Administration indicate that for the general population, about 20 percent of the difference in the rates of return between low- and high-income workers is removed by properly reflecting mortality differences by income. However, this correlation would be far more important in comparing rates of return between African Americans and the general population.
The authors note that rates of return would be lower for low-income and higher for high-income workers if adjusted appropriately for mortality at the indicated level of income. Average-income workers would presumably have mortality at about the level of the average population, and thus be unaffected by the adjustment. While this is all true for the general population, the African American rates of return would be increased much more for high-income workers, would also increase for average-income workers, and would decline much less for low-income workers. This is true because average income for African Americans is substantially lower than for the general population at this time.
For example, African Americans with earnings equal to the general population average have much higher-than-average earnings within the African American population. As a result, they must be assumed to have much better mortality than the average for all African Americans. Thus, with proper adjustment with mortality by income level, race differentials in life expectancy would diminish greatly, for individuals at the same income levels.
Without appropriate adjustment of mortality by income, race-differentials in rates of return are highly misleading. This is the reason that the Office of the Chief Actuary, Social Security Administration has not as yet developed race-specific money's worth ratios or rates of return.
The authors of the Heritage study exclude disability benefits and taxes from the analysis, even though the Disability Insurance program represents nearly 15 percent of the Social Security. While including disability would not substantially change the expected rate of return for the general population, it would considerably narrow the rate of return for African Americans versus the general population. This follows from the fact that disability benefits are significantly more likely for African Americans than for the population as a whole.
Because Social Security provides survivor benefits when a worker dies before retirement age, and Heritage omits these benefits, the authors also deducted from taxes a premium intended to cover these benefits. The premium appears to exclude those survivor benefits for a spouse that are payable after the youngest child reaches age 16, i.e., widow(er)s benefits that are payable from age 60 to the end of life. Premiums are also likely to be too low if Heritage used premiums for commercially-available life insurance. Life insurance is sold largely to healthy, higher-income persons who can afford the premium and can qualify as insurable. Including lower-income and uninsurable individuals would boost the premium substantially. Making these changes would increase rates of return from Social Security.
The authors raised tax rates for Social Security starting 2015 by enough to cover future benefit costs, ignoring the fact that the Trust Fund balance projected for that time, 300 percent of annual benefits, would continue to grow as a percent of annual benefits under these conditions. In order to produce a pay-as-you-go financed Social Security program, tax rates would not need to be raised until around 2025, and by less than the authors assumed. The early, too-large increase in tax rates results in rates of return that are estimated to be too low for Social Security.
The authors included a "high-income" worker with earnings at 300 percent of the average wage. In fact, the maximum level of earnings that is taxable and creditable for Social Security benefits is now and will be in the future 240 percent of the average wage (less than 240 percent in years before 1978). It is not clear whether the authors included taxes above this maximum benefit level. If they did, then rates of return for Social Security are estimated to be too low for high-income workers.
Date: April 2, 1998
To: Harry C. Ballantyne
From: Stephen Goss
Deputy Chief Actuary
Subject: Comments on Heritage Rates of Return for Hispanic Americans
In an analysis from the Heritage Foundation dated January 15, 1998, the authors Bill Beach and Gareth Davis, suggested that African Americans can expect to get a lower rate of return, and Hispanic Americans a higher rate of return from Social Security, than the general population. My memorandum on February 4, 1998 addressed a number of technical shortcomings in the methods used in this analysis. These shortcomings resulted in an understatement of the expected rates of return for all groups. The understatement was largest for African Americans, and smallest for Hispanic Americans. There correct expected rates of return would be higher, and closer together than presented by Heritage.
In a subsequent statement by Heritage in early March, referring to the results of he January 15 analysis, it was suggested that Hispanic Americans could expect far lower rates of return from Social Security than would be possible from individual accounts. The statement compared the (erroneously) computed expected rates of return from Social Security to the expected rate of return for an individual account invested either all in bonds or partly in stocks. But the comparison ignores both the cost of transition to an individual-account program, and the administrative expenses associated with individual accounts. Transition costs associated with the continuation of benefit payments for current OASDI beneficiaries and today's older workers would be substantial. Adding in these costs, which are excluded from the Heritage analysis, would greatly reduce the overall rate of return for the combination of individual accounts and the OASDI program under the Heritage proposal.
Finally, the authors fail to note in the March statement that, by their own calculations, see page 7 of January 15 report, Hispanic Americans would be expected to receive a substantially higher rate of return for Social Security that would the general population, on average. A somewhat higher rate of return for Hispanic Americans is to be expected, based on the higher life expectancy for Hispanic Americans, and the fact that Hispanic Americans have lower than average earnings. However, the extent to which Heritage shows higher returns for Hispanic Americans is overstated, just as the extent to which African Americans are show to have lower rates of return (Heritage January 15) is overstated.
Stephen C. Goss
1. Statement of Paul Yakoboski before the Senate Finance Committee, July 22, 1998.
2. General Accounting Office, "Social Security: Restoring Long-Term Solvency Will Require Difficult Choices," Testimony of Jane L. Ross before the Senate Special Committee on Aging, February 10, 1998, p. 11.
3. Statement by Peter Diamond, AARP-Concord Coalition Debate on Social Security, Albuquerque, July 27, 1998; Peter Diamond, "The Future of Social Security for this Generation and the Next: Proposals Regarding Personal Accounts," Testimony before the Subcommittee on Social Security of the House Committee on Ways and Means, June 18, 1998; Testimony of Henry J. Aaron before the Senate Committee on the Budget, July 23, 1998.
4. As Diamond has explained, with a one-percent annual charge on holdings in accounts, a dollar deposited in an individual account in the first year of a 40-year work career will be subject to the one percent fee 40 times, while a dollar deposited in the final year before retirement will be subject to the fee once. On average, dollars in the account will be subject to the one percent annual charge 20 times, so approximately 20 percent of the amounts deposited or earned by the account will be consumed by these charges.
5. Olivia S. Mitchell, James M. Porterba, and Mark J. Warshawsky, New Evidence on the Money's Worth of Individual Annuities, Cambridge, MA: National Bureau of Economic Research, April 1997.
6. General Accounting Office, Social Security: Different approaches for Addressing Program Solutions, July 1998, p. 6.
7. Steve Goss, op. cit.
8. U.S. Department of Health and Human Services, Vital Statistics of the United States, 1996: Life Tables, Washington, DC: U. S. Government Printing Office, 1996, Table 6-1, p. 8 and 10.
9. Heritage took into account that during the 45-year period that would pass between the time an individual was 20 and the time he or she was 65, life expectancies in the United States would be expected to climb modestly. But Heritage failed to take into account the more critical, and larger, factor that the expected age of death for those who reach 65 is substantially higher than the expected age of death at age 20.
10. Steve Goss, op. cit.
11. The figures cited here for numbers of work years and retirement years for 20-year-old men under the Heritage methodology were computed by the Social Security actuaries. In its report, Heritage uses 21-year-olds rather than 20-year-olds as an example.
12. Although the term "normal retirement age" is widely used, it is a misnomer the majority of Social Security beneficiaries retire and begin drawing benefits before the "normal retirement age." What is commonly called the "normal retirement age" is simply the age at which full, rather than reduced, Social Security benefits are paid.
13. Kelly A. Olsen, Jack VanDerhei, Dallas L. Salisbury, and Martin R. Holmer, "How Do Individual Social Security Accounts Stack Up? An Evaluation Using the EBRI_SSASIM2 Policy Simulation Model," Issue Brief, March 1998. The EBRI individual accounts options are partially privatized programs that would provide retirees with reduced OASI benefits in addition to their individual accounts.
14. Heritage claims to have dealt with transition costs in a subsequent document, "A New Framework for Cutting Taxes: Reforming the Tax Code and Improving Social Security," July 1, 1998. In fact, this document calls for diverting $867 billion over the next 10 years from the Social Security trust funds to individual accounts (about $1.9 trillion over 10 years) without replacing in the trust funds any of the foregone revenue. This Heritage document contains no mechanism to cover the transition cost.
15. 1998 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds, April 28, 1998, Tables II.C5, p. 38.
16. See Robert McGough, "Robust Fund Industry Isn't Lowering Fees," Wall Street Journal, May 14, 1998.
17. Olivia S. Mitchell, James M. Porterba, and Mark J. Warshawsky, New Evidence on the Money's Worth of Individual Annuities, Cambridge, MA: National Bureau of Economic Research, April 1997.
18. Board of Governors of the Federal Reserve System. "Family Finances in the U.S.: Recent Evidence from the Survey of Consumer Finances," Federal Reserve Bulletin, January 1997.
19. Heritage has belatedly acknowledged that administrative costs should be taken into account. In a September 1998 paper on Social Security and union households, Heritage assumes an annual charge on accounts of $100 plus one percent of the assets in the accounts, an initial setup cost of $1,000 per account, and a fee of 20 percent of the value of the principal for workers who choose to purchase an annuity. None of these costs are assumed, however, in the Heritage papers on rates of return for African Americans and Hispanic Americans, which were written earlier in 1998.
20. Social Security Administration, Annual Statistical Supplement, 1997: To the Social Security Bulletin, Washington, DC: U. S. Government Printing Office , 1997.
21. U. S. Department of Labor, Bureau of Labor Statistics, Employment and Earnings, Washington, DC:
U. S. Government Printing Office, Vol. 43, No. 1, January 1996, Tables A-3 and A-4 pp. 14-15
22. Dean R. Leimer, "Historical Redistribution Under the Social Security Disability Insurance Program," Social Security Administration, Office of Research, Evaluation, and Statistics, July 1998.
23. 1998 Annual Report of the Board of Trustees of the Federal Old Age and Survivors Insurance and Disability Insurance Trust Funds, April 28, 1998, Table II.C5., pp46-48 and Table II.C3, p44.
24. This information is based on conversations with actuaries at Guardian Life Insurance Company.
25. Lawrence Mishell, Jared Bernstein, John Schmitt, The State of Working America 1996-97, Armonk, NY: M.E. Sharpe, 1997. Table 3.5 p. 142.
26. A widow, widower, or divorced spouse aged 65 or older will receive 100 percent of the worker's benefit unless the worker retired early, in which case the survivor will receive the amount the worker would be receiving if the worker lived, but not less than 82.5 percent of the full benefit. If the surviving spouse elects to receive benefits between the ages of 60 and 65, he or she will receive a permanently reduced benefit determined by the age at which he or she begins receiving benefits. If the surviving spouse is disabled, he or she may begin receiving benefits as early as age 50 without a reduction in benefits.
27. A dependent parent who was receiving 50 percent of his or her income from the worker is eligible at age 62 for 83 percent of the worker's benefit; two parents may receive 75 percent each. Survivors benefits are subject to a family maximum determined using a progressive formula. This cap limits the total amount of benefits the family may receive.
28. Bureau of the Census, Poverty in the United States: 1996, P60-198.
29. Social Security Administration, Annual Statistical Supplement, 1997: To the Social Security Bulletin, Washington, DC: U. S. Government Printing Office , 1997.
30. The combined employer and employee payroll tax for Social Security equals 12.4 percent of earnings up to $68,400. Of this amount, 1.7 percent goes into the Social Security Disability Insurance trust fund, and 10.7 percent goes into the Old Age and Survivors Insurance trust fund.
31. These quotes are based on conversations with insurance agents at All State and State Farm insurance companies.
32. Still another problem with the manner in which Heritage estimated the portion of the payroll tax that goes for survivors' benefits is that Heritage based its estimate on the cost of a life insurance policy only for individuals aged 35 to 55. This is too short to be comparable to Social Security, which covers workers under 35 and over 55.
33. A spouse who has worked and is eligible for his or her own retirement benefits may receive 100 percent of his or her own benefit or 50 percent of the spouse's benefit, whichever is higher.
34. General Accounting Office, Social Security: Different Approaches for Addressing Program Solvency, July 1998, p. 60.
35. Mark J. Warshawsky, "The Market for Individual Annuities and the Reform of Social Security," Benefits Quarterly, 1997.
36. General Accounting Office, Social Security Reforms: Implications for Women's Retirement Income, December 1997.
37. Steve Goss, op.cit. p. 3.
38. James E. Duggan, Robert Gillingham, and John S. Greenlees, "Returns Paid to Early Social Security Cohorts," Contemporary Policy Issues, October 1993.
39. 1998 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds, April 28, 1998.
40. Report of the 1994-96 Advisory Council on Social Security, Volume I: Findings and Recommendations, January 1997.
41. 1998 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds, April 28, 1998, Table III.B2, p. 177.
42. Steve Goss, op. cit., p.4.
About the Author:
Dr. Kilolo Kijakazi is a senior policy analyst in the Income Security Division at the Center on Budget and Policy Priorities, where she specializes in Social Security. Dr. Kijakazi previously worked as a policy analyst at the National Urban League and the Food and Nutrition Service of the U. S. Department of Agriculture. She is the author of African-American Economic Development and Small Business Ownership, a book published by Garland Publishing, Inc. in 1997.
for PDF version
of this report
Download the Adobe Acrobat Reader
if you do not already have one.
Additional related reports: