Increasing Life Expectancy Could Undercut Social Security Viability, Scholar Finds
Samir Soneji, PhD, is an alumnus of the Robert Wood Johnson Foundation (RWJF) Health & Society Scholars program, and an assistant professor at the Dartmouth College Institute for Health Policy and Clinical Practice and the Norris Cotton Cancer Center. His study on the statistical security for Social Security was published in the August 2012 issue of Demography. Read the study.
Human Capital Blog: This study is a follow-up to your previous research. Can you briefly describe what you’ve studied up to this point?
Samir Soneji: Previously we studied the impact of historical smoking and obesity patterns on future mortality and life expectancy trends. For men there’s been a steady decline in cigarette smoking, and so also a gain in life expectancy. Women have also experienced a decline in cigarette smoking, but not as quickly. The rise in obesity has been much more recent than the historic decline in smoking, and we don’t know yet the impact of that rise. There’s a lag—the effect of today’s obesity may affect the population in 15-20 years, or later. One possibility may be that the rise in obesity may partially offset what’s been achieved by the historic reductions in smoking. Taking these factors into account, we found that both men and women will have an increase in life expectancy in the next 25 to 30 years.
HCB: Your new study looks at the solvency of Social Security. Tell us more about what you were analyzing.
Soneji: The Social Security Administration and Medicare use the same mortality and demographic forecasts to determine the number of beneficiaries, and the number of working age adults who are contributing payroll taxes to support those retirees.
Our previous research has shown that the government’s mortality forecasts could be off by several years. If we’re living longer than official government projections, that’s a good thing—it’s a good sign for population health—but there are some consequences of that. There will be more retirees supported by a smaller working age population. We wanted to examine the impact of these new mortality forecasts on Social Security solvency and financing.
HCB: How did you make your projections?
Soneji: We collaborated with a modeler to make a simulation model that’s very similar to the official Social Security model. We left everything the same as in the official models except the mortality forecast—so fertility, disability, payroll taxes, age at retirement, maximum taxable earnings, and other demographic and economic inputs, and policy levers all stayed constant. The only difference between our model and the Social Security solvency projections were the mortality inputs.
One of the challenges of the Social Security solvency projection is its transparency. It makes several assumptions about how diseases are related and their decline over time, and that goes against the advice of a lot of demographers. We created a much more transparent mortality forecast.
HCB: And what did you find?
Soneji: Because life expectancy is improving faster than Social Security projects—mostly because of the decline in smoking—there’s a greater strain on the system. From a solvency perspective, that means the larger of the two trust funds, the Old Age & Survivors Insurance program, will start to run at a loss several years earlier than official projected. Since the early ’80s, the Social Security administration has brought in more revenue than it needs to pay out in benefits, in anticipation of aging Baby Boomers, so they’ve amassed surplus revenue in the form of a trust fund, which are really special-issue Treasury bonds. That trust fund will be depleted faster and sooner than officially projected. We predict it will start to go into debt (be drawn down but not depleted) about four to five years earlier than Social Security projects.
HCB: What do you hope people take away from your study? Is there a message for policy-makers?
Soneji: First, we hope that we communicate the very likely reality that the population will age even faster than officially projected. That’s good news for public health, medical treatment, prevention and screening, but there’s a cost as well to large programs like Social Security and Medicare. Demographic shifts in the 20th century are at the heart of current concerns over Social Security solvency in the 21st century.
In terms of policy, we hope policy-makers see the changing composition of our population, and know that they can’t rely on demographic factors alone to extend the solvency of Social Security. We have to finally address the policy levers that we can change, like age at retirement, payroll taxes and the maximum taxable earnings. There are dozens of other countries with similar pay-as-you-go pension programs that face similar solvency concerns from aging populations. In some countries, benefit levels are indexed to life expectancy.
Social Security is the single largest federal program in the U.S. and it touches the lives of workers contributing payroll taxes and disabled workers, retired workers, and their dependents receiving benefits. We know it’s a challenge to change such a massive program, but given its profound benefits, it needs to be continued and remain solvent.