Social Security’s Impact | Sylvain Catherine

How Does Social Security Affect Wealth Inequality?

Dan Loney: When you think about retirement, there’s no doubt that part of the discussion ends up being around Social Security and what it means for Americans. But when you factor in wealth inequality, is there a difference in what it is now compared to a few decades ago? That’s part of the research done by our guest here today, Sylvain Catherine, who is an assistant professor of finance here at the Wharton School.

Sylvain, let’s start with the backstory on wealth inequality tied to Social Security.

Sylvain Catherine: As you mentioned, when people think of retirement, a big part of it is Social Security. For most Americans, most of their income during the retirement period does not come from the stock of wealth that they have at the beginning, but comes from the Social Security benefits that they receive.

Now, all those promises that the government makes, they have a value. You could think of it like, if you were to go on the private market, you could buy an annuity. And that annuity would basically offer exactly the same type of terms as Social Security. It would provide a monthly payment until the end of your life. So, there is a market value for what the government provides.

Once you try to value those benefits, the ones that you have already accrued because you have contributed into the system, what’s the value of this? How does it change the level of inequalities that we see today? Does it change also the trends in wealth inequality? Because when we look at wealth excluding Social Security, we see a steady increase in wealth inequality since more or less the mid-1980s. But what we find in our paper is that once you factor in Social Security, this positive trend in wealth inequality basically disappears.

Loney: When you think about value for Social Security, how has that changed over the last several decades?

Catherine: It has changed enormously, and this has implication both for households but also for the government. Because, of course, what we consider as an asset for households is going to be a liability for the government. But we are talking about something right now that is close to $50 trillion, where the total stock of wealth excluding Social Security in the U.S. would be slightly more than $100 trillion. So, you have one-third of the total that is Social Security, which was not considered in inequality statistics before.

Loney: When you talk about the different income brackets, there’s probably much more of a reliance on Social Security as a component of support in your retirement years for lower-income families than for higher-income families.

Catherine: Exactly. In general, as you move up in the income distribution, people receive higher benefits. But that slope, that relationship, is much less pronounced than if you look at wealth in general. Because there is much less inequality in Social Security benefits, adding it to the bucket of the things that you consider as wealth totally changes the picture that you have when you trace the level of inequality over time.

Loney: With everything we’ve seen, especially in recent years, how do we value Social Security benefits right now? Maybe even more importantly, how might we value them in the future? Because there are so many questions about what Social Security might be in the next decade or so.

Catherine: That’s an important consideration. We have known for decades that at some point around the year 2033 or 2035 — this data has moved a little bit — the Social Security Trust Fund, which has the reserves of the Social Security system, is going to run out of money. When this happens, Social Security will not be able to pay the benefit in full. Based on the current estimate, it will be able to pay 75% of the benefits. At this point, there will need to be a reform. Either you will have to raise the level of payroll taxes that form Social Security, or you will have to lower the benefits or move the retirement age.

Now, in our paper, we take this into account. We have some tests, like what if we pay the benefit in full, what if we cut them by 20%? And we look at how it changes the trends in wealth inequality. But you have to remember that this is not news. If you look at the Social Security reports that the Social Security Administration publishes every year about the health of the system, we have known that for decades. That was, in a sense, already known, 40 years ago, so it doesn’t change the trends that much.

Loney: But the concept of inequality has probably shifted over the last few decades as well.

Catherine: What do you mean?

Loney: When you think about the impact of the inequality because of how Social Security has changed, we’re seeing probably a much greater impact than we have in the past.

Catherine: There are several reasons for that. One is that you have many more workers that are covered by Social Security, like young retirees, than 40 years ago. Second, life expectancy has increased. And the third thing, which is, I think, the more interesting point, is that the level of interest rate has declined enormously. When you try to buy an annuity from an insurance, for example, the level of the interest rate is going to affect the value of this annuity. The lower the interest rates, the higher the value of the assets, such as future Social Security benefits.

Loney: We went for such a long period of time with interest rates basically at zero. Then we saw that rise around the pandemic, and we still are seeing a higher rate of inflation than we saw, say, 20 years ago.

Catherine: What really matters here is a level of real interest rates. That is the difference between the level of nominal interest rates that you would see on a mortgage contract and the level of inflation. Social Security benefits are indexed on inflation. When inflation goes up, it doesn’t really hurt retirees as much as it hurts the rest of the population, because they are protected against it. Now, the spread between the nominal interest rate and inflation is still very low relative to what it was in the ‘80s.

The Future of Social Security and Retirement Policies

Loney: We’ve obviously had a lot of policy discussion around Social Security, especially lately. One of the things that the current administration brought forward was not taxing Social Security benefits. All of these ideas filter into the mix as to that component of inequality and Social Security benefits.

Catherine: Yeah. This kind of policy has big redistribution effects — not so much between rich and poor people, but more so between different generations. For example, cutting what they call the Social Security tax is effectively increasing the Social Security benefits. Now, there is no free lunch. Social Security benefits in a pay-as-you-go retirement system are paid by workers, so if you decide that the current cohort of retirees will receive more benefits, that means that the deal is mechanically getting worse for younger generations. That’s the dimension on which the redistribution along this kind of policy operates. It’s between generations.

Loney: It’s not a component of wealth that’s transferable to that next generation as well.

Catherine: That’s a choice for your parents or your grandparents. If they decide that these additional benefits, like not having to pay the tax on Social Security taxes — because just to be clear, the Social Security tax is not going to the federal government. It feeds back into the Social Security Trust Fund, so I think the right way to think about it is an increase in benefits.

If your parents who are currently retired decide to save that money instead of spending it, then you would get it back for inheritance. Now, not all Americans are equally situated on that dimension. Because, first of all, that Social Security tax is only going to go to towards high-income Social Security recipients. Effectively, what would happen is that you would have current workers, whose parents have low Social Security benefits — for them, it means that they will have to fund that increase in generosity for current retirees. But their parents will not be able to save that money to return it to them in the form of inheritance. So yes, it’s a transfer from young generations to the older generation. And potentially, that transfer is kind of offset by the inheritance mechanism, but not equally for all young Americans.

Loney: What I found interesting in reading through the paper is the impact of Social Security benefits to inequality on the top 10%. Take us through why that’s so important.

Catherine: Because when you look at the bottom 90% — the rest of the population — for them, Social Security is more or less half their wealth. Whereas for the top 10%, it’s much smaller. And for the top 1%, it’s like, totally insignificant. Therefore, when you ignore Social Security, you kind of totally overlook the biggest chunk of wealth that the bottom 90% has. And you fully count the wealth of the top 10%. Which is why including it really changes the picture.

Loney: You also mentioned at the end of the paper that we’re still at a point where there’s a lot more research to do in terms of looking into this and these impacts as we move forward, correct?

Catherine: Right. Because our paper is mostly an accounting paper. We are just trying to measure how much wealth people have. It’s not a measure of the causal effect of Social Security on the distribution of wealth. Because if Social Security did not exist, the behavior of people would totally change. Presumably, net wages would be higher because you wouldn’t have to pay the Social Security tax. And there is a question of what people would do with the money. Would they have saved more privately, or would they have chosen to consume it?

Depending on the assumptions that you are making here, the causal effect of Social Security on wealth inequality could be very different. Our paper is just measuring things. It’s just trying to get the accounting right, but it doesn’t really tell you what Social Security is actually doing in terms of affecting inequality.

Loney: Is there a natural area that you would like to build off of this research to look at next?

Catherine: As you mentioned, one big source of concern for the next 10 years is this funding gap. And I alluded to, there are several ways to solve that gap. Now, whether you solve it by reducing the benefit, changing the retirement age, or by increasing taxes, that does not affect all Americans in the same way. Since it’s a big fraction of their wealth, the actual policy decision will have a big effect on the way the cost, the burden, is distributed among Americans. That’s something that we are trying to do right now.

Loney: That puts more emphasis on the policy side over the next decade to truly understand the path that we’re going to see.

Catherine: Yes. Right. Once you get the accounting right, you can think of, “OK, if I change the rule, who gets affected? Who benefits?” In net, it’s going to be a burden, because there is this money that is just missing. But depending on the policy choice that you make to address this problem, who is going to be affected in terms of which cohort — and we mean whether it’s high-income people or low-income people — that really depends on the way we decide to solve that problem.

Loney: Is there enough of an understanding by policymakers about the accounting side, so they can make the proper decisions on the policy side?

Catherine: In a sense, that’s what we are trying to provide in our future research. I think a lot of policymakers know about this. Now, I think the problem is that Social Security has been described as the third rail of American politics. As long as the trust fund still has money, I don’t think any of them will say anything about this. Because it’s just like, there is only political cost. But we know that fact. We are in 2025. In eight years, based on the latest estimate, there won’t be any money left. At that point, policymakers will have to say, “We prefer to raise the retirement age, or we prefer to cut the benefit, or we prefer to raise the taxes.” Those are going to be our choices.