Social Security has a reputation around Washington as the third rail of American politics ("touch it and you're dead"). Last year, three senators boldly stepped onto the middle of the tracks.
Among the changes they are urging their colleagues to adopt are
The first change was proposed by Senator Daniel Patrick Moynihan, Democrat of New York, and all three were introduced as part of a sweeping Social Security reform package by Senators Bob Kerrey, Democrat of Nebraska, and Alan K. Simpson, Republican of Wyoming.[1]
Galvanizing these recent legislative efforts was the report in late 1994 of the Bipartisan Commission on Entitlement and Tax Reform, which warned of an explosion in spending on Social Security and Medicare when the baby-boom generation retires.
According to the commission, unless reforms are enacted, spending on entitlements and interest on the national debt will consume every dollar the federal government collects in taxes by 2012. The Social Security trust fund, which today runs a surplus, will begin running a deficit by 2013 and will be completely insolvent by 2029, a calamity that would force either severe benefit cuts for retirees or a massive payroll tax increase for workers.
Clearly reforms must come, but altering Social Security's complex structure of taxes and benefits is a monumental undertaking that will touch every American, living and yet to be born. Before beginning the task, policymakers need to know how the current system affects various groups--who gains and who loses--and what the likely consequences are of changes aimed at preventing the depletion of the fund.
The study focuses on the Old Age and Survivors Insurance (OASI) portion of Social Security, which provides benefits to retired workers, their spouses and their widows. OASI makes up about 88 percent of Social Security payments.
Most retired Americans believe they have earned their Social Security benefits and fiercely oppose any talk of changing the system. After all, they have contributed payroll taxes into the Social Security system most of their lives.
But it is a profound misunderstanding to think that the retirement benefits an individual receives are funded by payroll contributions that person made during his or her working career. Social Security--established in the 1930s to combat poverty among the elderly--is neither a pension plan nor, despite its name, a genuine trust fund. Rather, it is a redistribution system: Each generation of workers pays taxes to support the previous generation in return for a promise that the next generation will support it.
The system works provided that each generation has enough children to underwrite benefits. The baby-boom generation is producing proportionately fewer children. Today, there are 3.3 workers paying taxes to support each retiree; when the baby boomers retire, there will be 1.8.
In 1994, 139 million working Americans and their employers pumped $345 billion into the Social Security system. Under present trends, many of the 139 million have no reasonable expectation of receiving anything like current levels of benefits from the fund when they retire. Social Security has hundreds of billions of dollars in unfunded promises it will be unable to pay if the system goes bust.
As Panis and Lillard point out, it is well established that Social Security is not an actuarially fair program for all individual participants. Those who retired many years ago earn a better return on their contributions than the recently retired, thanks to low payroll tax rates during the program's startup phase and amendments in the 1970s that increased benefits. There are also large differences in rates of return within birth cohorts (groups born in particular years) purposely built into the system to ensure disproportionate benefits to elderly individuals who, because of low incomes or above-average life spans, are more needy than others.
But the goal of equity--a fair rate of return for all participants--has never been abandoned. In fact, the researchers observe, the system has always striven for a balance between the safety net and fairness pillars of the program. In this, the RAND analysis shows, Social Security has been more successful along both dimensions than previously thought. It is still progressive--wealth flows from the prosperous to the poor--but far less so than other studies have indicated.
Previous analyses of how different socioeconomic groups fare have shown that over lifetimes, large amounts of money are transferred from high- to low-income groups, from single individuals to married couples, and from men to women. Most of these studies use standard survival tables that project mortality rates on the basis of age, sex and race.
Using a richer set of socioeconomic characteristics that include marital status and income, Panis and Lillard find that the dollar transfers from high-income to low-income individuals are a great deal smaller than studies based on standard survival tables suggest. The reason: High-end earners live longer and thus collect Social Security benefits over a longer period.
This longevity of the well-to-do partially offsets the progressive nature of the benefit formula that both takes more in contributions from high-income individuals during their working years and taxes their benefits in retirement.
Racial differences are also much smaller than expected. It is true that blacks receive lower returns from Social Security contributions than whites because they die at a younger age. But there is little difference in the mortality rates of blacks and whites with the same income; thus Social Security transfers from blacks to whites with the same earnings are also small. In fact, as a group, blacks may earn a better return on their contributions than whites because of their lower average incomes.
Differences in lifetime rates of return (here translated into 1995 dollars) among various social and economic groups are still substantial. For the cohort born in 1930 and retiring in 1995, the study came to the following conclusions:
In the second part of the study, the researchers assess the effects of specific reforms on people in various income categories, born in the years 1930, 1950 and 1970. The 1930 cohort just retired last year and would be spared the effects of some reforms; the 1970 cohort only recently entered the workforce and would feel the full impact of any amendment. Differences by sex, race and marital status are not considered because none of the reforms would have a disproportionate effect on any of these groups.
Raising the retirement age. Currently, 65 is the age at which workers may retire with full benefits, but in 1983 Congress voted to gradually increase the retirement age to 67 by 2022. New proposals have suggested raising it to 70 to reflect the growth in life expectancy.
People retiring today obviously would not be affected by changes in the retirement age. Rather, younger people would bear the brunt of this amendment. For example, an increase in the retirement age to 70 would cost a typical young couple, born in 1970, about $70,000--less for low-wage earners, more for high-wage earners.
Increasing contributions. Increasing the contribution rate has no effect on current retirees. But a couple who recently entered the labor force could expect to pay roughly $120,000 more in payroll taxes if the total (employer plus employee) contribution rate were increased by 5 percentage points.
Reducing benefits. Reducing benefits affects the 1930 through 1970 birth cohorts almost equally; couples may expect roughly $80,000 less in benefits as a result of a 30 percent across-the-board cut in benefits.
Panis and Lillard note that all of the reforms currently on the table would make the benefit formula more progressive--that is, they would increase the amount of money going from the wealthy to the poor. Although that achieves the program's original aim of protecting the elderly poor, it has important implications for Congress in its deliberations on reforming Social Security.
"Some individuals stand to lose more than others from the reforms," says Panis, "and while it is tempting to support only those measures that protect one's own interests, a well-balanced package spreads the pain more evenly and would be more likely to find broad acceptance."
In choosing the elements of such a package, he concludes, "Congress must also keep in mind the trade-offs between protecting the elderly poor and providing a fair rate of return to all workers."
Socioeconomic Differentials in the Returns to Social Security, Constantijn W. A. Panis, Lee A. Lillard, RAND/DRU-1327-NIA, 1996, 25 pp., no charge.