Insider Interview: The Future of Social Security

What is the current state of Social Security’s financial health? And what might be the best ways to shore up the program for future generations? As the Social Security Administration’s chief actuary, Stephen Goss is responsible for producing the actuarial estimates and analysis that can help answer those questions. In this lightly edited conversation with Senior Editor Eleanor Laise, Goss reflects on his 46 years at Social Security and discusses the solvency of the Old-Age and Survivors Insurance (OASI) Trust Fund, which pays retirement and survivor benefits, and the Disability Insurance (DI) Trust Fund, which pays disability benefits. (These are legally separate entities, but Social Security trustees look at a hypothetical combined trust fund to gauge the program’s overall financial stability.)

According to the most recent trustees report, the Social Security trust funds [the hypothetical combination of the OASI and DI trust funds] will be depleted in 2035, at which point they’d be able to pay only 80% of scheduled benefits. What are the main factors driving that depletion of the trust funds? Largely because of demographic changes, we’re moving toward spending more than we’re taking in. During the baby boom period, the birth rate was 3.3 children per woman over a lifetime. Shortly after 1965, we dropped down to an average two children per woman, and that’s a big change.

A lower birth rate changes the age distribution of the population. Back in the day we had three kids per couple, and now we’ll have two kids contributing for every retired couple. Those two kids working will have to pay more to cover the two parents, or the two parents have to get less benefits. That’s what we’re facing.

Subscribe to Kiplinger’s Personal Finance

Be a smarter, better informed investor.

Save up to 74%

Sign up for Kiplinger’s Free E-Newsletters

Profit and prosper with the best of expert advice on investing, taxes, retirement, personal finance and more - straight to your e-mail.

Profit and prosper with the best of expert advice - straight to your e-mail.

Sign up

In the future, because of the drop in birth rates, we need to either reduce benefit levels on the order of 20% to 25% or increase revenue by a third by 2035. Either that, or have a whole bunch more kids.

Next year, for the first time since 1982, Social Security’s total cost will exceed its total income, and the program must start drawing down its assets to pay benefits. How concerned should current and future beneficiaries be about this? This is not unforeseen. For the last several years, we were expecting this little turning point to occur even sooner. But the economy has been relatively good, and our disability applications have dropped enormously, so we have much less cost on the disability side.

The trust funds are explicitly for providing a cushion. The reserve in the trust fund is intended to cover us for periods when the cost of the program exceeds the income coming in. The program has been around paying monthly benefits since 1940. Social Security has never missed a benefit payment and never reached a point where it depleted reserves and was unable to pay benefits on a timely basis in full. One in six Americans is, in any given month, receiving a benefit from Social Security. If all those were to be reduced by 20%, there would be a lot of unhappy people. Congress understands that.

The last time this happened, in the early 1980s, a bipartisan National Commission on Social Security Reform was formed to address it. Do you think this issue is getting enough attention in Washington today? It’s clearly getting plenty of attention. Before Congress now is a proposal by Rep. Larson [D-CT], the current chairman of the House Ways and Means Social Security Subcommittee. He has a comprehensive bill with many sponsors. [Previously, former Texas Republican Congressman] Sam Johnson had a comprehensive bill before Congress. They took very different approaches. Rep. Larson would largely bring in more revenue. Sam Johnson would reduce benefit levels. History suggests that when we have comprehensive legislation, there will be compromise. It’s just a matter of Congress getting to the point where they decide now is the time. We hope we’re doing a service in the office of the actuary. Anybody in Congress who comes up with a proposal, we work with them on estimates.

You mentioned the substantial decline in Social Security disability applications. Can you talk about the significance of that? It’s quite remarkable. The number of disability applications for Social Security rose to just over 2 million in the year 2010. We understood that number was pushed up by there being a bad recession. Any of us would do the same. If you lose your job and think you might have a medically determinable impairment, you’re going to apply.

In 2018 and 2019, our total disability applications will be below 1.4 million [per year]. The largest driver of that is the changing nature of work in our economy. We’ve moved away from high-risk physical jobs to jobs in the service economy and jobs in production where [workers are] monitoring a machine doing the production. We have people exposed to a lot less risk of physical injury and harm than in the past.

In the latest trustees report, we’re projecting being solvent in the trust fund to pay benefits timely in full for disability insurance up until 2052. DI is looking to be the healthier of the two funds at this moment.

What do you say to people who are approaching retirement and want to claim their Social Security benefits as soon as possible because they’re worried about future benefit cuts? Don’t worry about it, beyond writing a letter to your member of Congress and saying, “let’s get on this.” And remember, an increase in the normal retirement age was enacted in 1983, but [the normal retirement] age didn’t even start to rise until 17 years later, in the year 2000. Anybody 45 years old or older [at the time] wasn’t even affected by this. There are very few cases where people want to make changes in law that would negatively affect benefits for people near retirement.

What are the most effective steps that could realistically be implemented in the near term to shore up Social Security? Changes could be enacted right now that won’t come into implementation until 2035, but what if we enact changes now that will either start to generate more revenue sooner than 2035 or reduce benefit levels from what’s scheduled now? There’s an opportunity if we enact something to raise tax rates somewhat, as Chairman Larson would. You could have that apply to people well before 2035 and have more people pay additional taxes. There’s clearly some advantage in having changes enacted sooner and having them effective sooner, because either would mean more generations participating.

Obviously, there’s a lot riding on the accuracy of your forecasts. A 2015 paper by researchers at Harvard and Dartmouth argued that since 2000, Social Security’s forecasting has been systematically biased toward suggesting the program is in better financial shape than is really the case—in part because retirees’ life expectancy is underestimated. What do you say about the accuracy of your projections? We try to make projections of what’s going to happen in the future—and we never refer to them as forecasts. We lay out a set of assumptions, and under those assumptions, make projections of what’s going to happen. We tend to develop assumptions that are on a good center path of what is most expected. In the real world, things don’t move on a good centered path. We haven’t outlawed the economic cycle yet, and sometimes we move into a recession, as we did in late 2007, and sometimes we have a recovery that’s stronger than expected, which we also had. So we do have volatility and swings that occur, and we don’t project those swings in our intermediate assumptions.

Our projections since about 2011-2012 have been shown to be consistently on the pessimistic side, because the economy and the experience of the program have been better than we were anticipating. What we hope is these projections will be evaluated not just on the basis of what happens for the first two or three years of the projection but over a longer period.

You’ve worked at the Social Security Administration for 46 years now. Are you still learning new things about Social Security? One saying in my office is, “any day that goes by when you don’t learn one or two new things is a terrible day.” And I’ve not had many terrible days in these 46 years. After our last trustees report, we were talking to regions and posing the question we pose to everybody: Do you have insight on why disability applications are dropping? What we heard was not only the changing nature of jobs but a sense from regional commissioners that there has been increasing compliance with [the Americans with Disabilities Act]. More and more employers are providing accommodations on the job. [Given] that level of impairment that would have precluded [workers] from finding jobs in the past, there are more jobs now that allow them to stay in employment. It’s a wonderful story all the way around.