(August 2000) On August 14, 2000, Social Security turned 65 years old, but it’s not about to retire. First signed into law during the Great Depression, the system has become a cornerstone of American life, providing more than half of family income to two out of three Americans ages 65 and older.
But when Americans look back on Social Security’s accomplishments this month, they should also take a moment to ponder the system’s next 65 years. With baby boomers retiring, life expectancies increasing, and family sizes changing, large population shifts loom on the horizon, and these demographic developments will have powerful consequences for the viability of Social Security. But since no one can predict just how the U.S. population will change, the retirement system’s future is unavoidably murky.
“The key question in a pay-as-you-go system such as ours is the ratio of workers to beneficiaries,” says Steve Goss of Office of the Actuary at the Social Security Administration (SSA). The SSA’s best guess is that this ratio will drop from 3.4:1 today to 2.1:1 in 2030, leaving a relatively smaller workforce to shoulder the burden of financing the system. If no adjustments are made, the Social Security trust fund will go bankrupt in 2037 under this scenario.
But estimates of future worker-beneficiary ratios are complicated by controversies over birth and death rates. While the SSA forecasts that the average number of children born per woman will fall from 2.06 in 1999 to 1.95 in 2025, the Census Bureau’s middle-series projections have this figure rising to 2.2 over that period. If the Census Bureau’s estimate proves more accurate, the relative number of workers in the future will be higher, and — other things being equal — Social Security’s shortfall will not be as severe.
Future life expectancies are another critical variable, but “the uncertainty on this issue is enormous,” says Goss. With scientists poised to make rapid strides in genetics and biotechnology, mortality rates could fall substantially over the next 30 years. On the other hand, SSA analysts wonder whether new medical advances will be as widely available as they were in the twentieth century and what the side effects of new treatments and technologies will be. As a result of such questions, the SSA predicts a relatively modest improvement in life expectancy, from about 76 years in 1999 to roughly 83 years in 2075.
Many demographers expect more dramatic improvements in longevity. Professor Ronald Lee of the University of California at Berkeley in particular thinks that death rates will continue to fall as rapidly as they did in the past century. If so, Americans born in 2075 will be able to look forward to 86 years of life. While longer life is good news, it also puts more financial pressure on Social Security in the absence of adjustments such as raising the retirement age.
Future immigration rates are yet another question mark. Unlike birth and death rates, the actual level of immigration can be controlled reasonably well by policymakers, but forecasters cannot predict the behavior of future politicians. Current U.S. immigration levels — at roughly 900,000 per year — are not high by historical standards, and could increase significantly if limits were eased.
Yet future immigration may not have too big an effect on Social Security’s viability. “If you look at it on a per immigrant basis, immigration looks like it would have a significant effect (on Social Security),” says Lee, noting that most immigrants tend to be of working age. “But if you look at it in the context of the larger population,” he continues, “the effect becomes quite small.” According to his calculations, the U.S. would need an additional five million immigrants per year — six times the current rate-every year for the next 75 years to make the system solvent over that time period.
As if this thicket of individual questions were not enough, these various uncertainties must be considered together before a picture of the future can be developed. To account for the uncertainty, the SSA brackets its best estimate of the future with scenarios that would impose higher and lower costs on the system (see table). The “high cost” scenario, for example, combines lower birthrates with faster increases in longevity, developments that would combine to put even greater pressure on Social Security’s finances.
|Year||Total Fertility Rate||Life Expectancy at Birth (Men)||Life Expectancy at Birth (Women)|
|2075 (high cost)||1.7||85.5||89.2|
|2075 (low cost)||2.2||77.5||81.7|
Source: 2000 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Trust Funds.
But contention lurks here, too. In contrast to the SSA’s scenario-oriented approach, Professor Lee and Shripad Tuljapurkar of Mountain View Research (MVR) have developed “stochastic” forecasts, models of the future that take into account the range of possibilities that could arise from interactions between multiple variables. The end result is not a prediction, but rather a projection that assigns probabilities to ranges of future values. Using this methodology, Lee and MVR’s Michael Anderson have warned that even if 90 percent of the Social Security reserve fund were invested in the stock market today, there would still be a two-thirds chance that the fund would be exhausted by 2099, assuming a 7 percent rate of return over the long run.
Like it or not, the U.S. will have to find a way to prepare for the future in the face of so much uncertainty. While that might sound impossible, other countries are already doing it. Sweden’s retirement system, for one, indexes benefits to life expectancy at retirement. If life expectancy increases but the retirement age does not change, beneficiaries receive proportionately less each year in benefits. To be sure, this particular type of reform would be politically difficult in the U.S., but there is still a need to think of ways to “build uncertainty into policy,” as Lee says. At least, there is if Americans want Social Security to remain viable for another 65 years.
Bingham Kennedy, Jr. is an associate editor at the Population Reference Bureau.