Furman on Social Security: Attention must be paid

Consumer confidence in the U.S. economy recently hit an all-time low. New data on Social Security, inflation, and the national debt is unlikely to lift anyone’s spirits.

Earlier this month, trustees of the Social Security Administration said that there will not be enough money to pay recipients their full benefits by 2032, earlier than expected, without more funding and/or cost cuts. The Consumer Price Index showed inflation hit a three-year high in May at 4.2 percent. And the U.S. now has a record-high $31 trillion in publicly held debt, equal to the country’s gross domestic product.

In this edited conversation, Jason Furman, Aetna Professor of the Practice of Economic Policy jointly at Harvard Kennedy School and in the Department of Economics, discusses Social Security’s impending cash crunch, consumer pessimism, and why new data about the national debt is “definitely a problem.” Furman, who was an outspoken critic of “Bidenomics,” served in the Clinton administrations and was President Barack Obama’s chief economist.

---

You recently wrote in The New York Times that Social Security’s solvency crisis is closer than anyone ever imagined. Why is this happening sooner than previously forecast?

Social Security was last significantly retuned in 1983 and the goal was to make it last at least another 75 years. Within about a decade, it became clear that that expectation was too optimistic. Fertility rates, especially, were falling faster than expected, longevity rising a little bit more than expected, and other economic numbers, like interest rates, were lower than expected. Fertility is the biggest one — it fell further than what the actuaries were counting on. Since the 1990s, we have expected that the day of reckoning for Social Security would be coming probably in the 2030s. With the latest trustees’ report, they pulled it forward a little bit.

Some of that was because of choices Congress made: A law that was passed in 2024 expanded benefits for some state employees, and a law that was passed in 2025, the One, Big, Beautiful Bill, effectively expanded benefits de facto for high-income households. So, some of it was the law, and some of it was just the annual technical revision process that goes into these estimates.

How much is needed and what are some fixes that could shore up this gap before 2032?

We need several trillion dollars. If you raised everyone’s payroll taxes by 2 percent, that would be enough. That’s a lot, but U.S. payroll taxes are much lower than most other rich countries. The 12.2 percent we pay is a lot less than most everyone else pays.

I don’t mind raising the cap on taxable earnings, but the thing I worry about is with Social Security, it’s primarily been the people who benefit are the ones who pay. There’s a limit to how many different types of increases we can have on high-income people. So, it can be the answer to some of our fiscal questions; it can’t be the answer to every single fiscal question.

“We need several trillion dollars. If you raised everyone’s payroll taxes by 2 percent, that would be enough.”

Is this looming shortfall primarily a math problem, a political problem, or both?

It’s an elementary math problem and a Ph.D.-level political problem. You could assign students the problem of how to solve this, and if it was in an economics class, it would be extremely easy to figure out what combination of benefit cuts and tax increases adds up to the magic number. But if it’s in a political class, I’m not even sure there’s an answer to the question.

Given how strongly voters feel about the program, why has neither party done much to head off the funding challenges since this problem has been known about for decades?

What’s interesting to me is that the interest in the problem has diminished over time, not grown. President Clinton really did put time, effort, and political capital into it, and some people think but for the Monica Lewinsky scandal this would have been addressed. George W. Bush put some effort into it. I didn’t like his plan, so I’m glad it didn’t happen, but I think it was actually a good-faith effort. Obama put a little bit of effort into it in the beginning of his term in 2010 and 2011 but probably lost interest in the issue after that, didn’t really see any pathway to dealing with it. I think in some ways, as the problem gets closer, the solutions get less attractive, and as a result, the bigger the problem, the less we talk about it.

Last month, the Consumer Price Index showed inflation was up 4.2 percent over May 2025, but 2.9 percent with food and energy stripped away. Is this a good sign? What do those numbers tell you?

It’s important to distinguish the price level from the inflation rate. The price level went up a lot in March. It went up a bunch again in April. It didn’t go up as much in May, but it went up. So, from the perspective of consumers, it was quite a bad report.

If you’re the Federal Reserve trying to figure out whether there’s some new ongoing inflation where each and every year prices are going to keep rising by 3, 4 percent, I thought the report was somewhat reassuring on that score. The amount of inflation within May came down a lot. We’ve already seen gasoline prices starting to fall again. They’re higher than they were, but the direction is now down, not up. With the latest Iran deal, assuming it sticks, oil prices have fallen quite a lot, and that will work its way into gasoline prices over the next month or two.

Is it bad news for consumers? The answer is yes. Is it a new era of ongoing sustained inflation that the Fed needs to raise interest rates to fight? Cautiously, I lean toward no — while being very nervous because it’s been many years of elevated inflation, so you don’t want to be too sure about anything.

How is low confidence affecting consumer spending?

So far, we’re seeing nothing in people’s actual behavior. This is the way they answer questions, but not the way they spend their money.

You can do a statistical model based on all the different economic variables, how much would you predict people spend, and then add a variable in for how confident consumers are. Normally, that variable, how confident consumers are, is a small positive. Everything else being equal, if you’re more optimistic, you spend a bit more, and if you’re more pessimistic, you spend a bit less.

If you do that same exact statistical analysis, but use data for the last five years, you get the wrong sign on confidence. The more negative people are, the more they spend, and the more positive they are, the less they spend. Now, I don’t believe the negative sign is true, but it suggests that in the data in the last five years that the positive sign is definitely not true. It has, to date, been detached from economic activity.

But you raise a question we don’t know the answer to, which is, might it become self-fulfilling? The one place where there’s some evidence it could become self-fulfilling is not on the amount of money consumers spend, but on people increasing their expectations for inflation on a partisan basis and then that becoming self-fulfilling. If you’re a business, you raise prices more because you think there’s going to be more inflation; you’re a worker, you demand a bigger wage increase. There’s been some research which suggests that partisan irrationality on inflation — this was Republican irrationality under Biden — actually had a self-fulfilling increase in inflation. It looks like it’s still happening, although now it is partisan, irrational Democratic beliefs about inflation. There’s not enough data, it’s a year and a half of data, but they might be raising expectations of inflation and becoming self-fulfilling.

Investors do not appear to reflect a similar lack of confidence in where the economy is headed. Why have the markets remained so buoyant?

The market is extremely skewed right now, and its gains are very, very dependent on a small number of AI-related tech companies. They are betting that these companies are going to make huge profits in the future; most of them are losing money now and hemorrhaging money. So, the markets are not in any tension with all of these things, they’re just recording a totally different piece of the economy. Right now, both financially and also in terms of GDP growth, our economy is increasingly reliant on just one subpart of the overall economy. That’s a little bit of a cause for concern in terms of just how resilient things will be.

In its annual report to Congress, the Government Accountability Office pegged the nation’s debt at $31.3 trillion, equal to the size of our economy, and predicted that will grow more than twice as fast as the U.S. economy over the next 10 years. How big of a problem is this?

It is definitely a problem. The deficit, which is the amount you add to the debt in any given year, is larger relative to our economy than any other rich country, and it’s larger relative to our economy than any time in our history with the exceptions of World War II, the 2008-09 financial crisis, and COVID. So, we’re in an unprecedented place for a non-emergency situation, both for our own history and relative to the rest of the world.

You can debate how bad and how risky it is, but you’d have a hard time arguing that this is the way we should try to do things. Part of why it’s up for such debate is that there is almost no historical experience to call on. Right now, there’s not really anything to confidently extrapolate or infer from, and to me, that says that we’d rather not find out the answer to this. And so, we should deal with it.