Doug Elmendorf and Karen Dynan: How Much Can the Federal Budget and the Deficit Continue to Grow?

What should be done now about the federal budget and the deficit, with Doug Elmendorf, dean of the Harvard Kennedy School, and Karen Dynan, professor of the practice of economics.

Doug Elmendorf and Karen Dynan

Photographs by Martha Stewart



Even before the coronavirus shifted the U.S. economy into low gear, demanding a massive stimulus in response, federal debt as a percentage of GDP was as high as it had been since the years following World War II. Simultaneously, given the nation’s aging population, spending on benefits for older Americans was and is expected to skyrocket. Should voters be worried? Why would curbing federal deficits now be a mistake? In this episode, recorded before the coronavirus arrived in the United States but perhaps even more pertinent now, Price professor of public policy Douglas Elmendorf, who is dean of the Harvard Kennedy School, and professor of the practice of economics Karen Dynan argue that this is the moment for the U.S. government to borrow. See the May-June 2020 issue of Harvard Magazine for an article by Elmendorf and Dynan on these topics.


Transcript from the interview (the following was prepared by a machine algorithm, and may not perfectly reflect the audio file of the interview):


A note to our listeners: this episode was recorded on February 12, 2020. 

Jonathan Shaw: Politicians on the campaign trail are famous for making promises—to reform healthcare, to not raise taxes; but ultimately such pledges live and die with the constraints of the federal budget. For most voters, that budget is a black box, just a very big number. Joining us today are two guests who bring brisk analytical skills to their scrutiny of the federal budget and who reach surprising conclusions about what is fiscally possible, and even what is economically desirable to be doing now with respect to the national public coffers. Douglas Elmendorf, who earned his doctorate in economics from Harvard in 1989, is the Don K. Price professor of public policy, and the Dean of Harvard Kennedy School. From 2009 to 2015, prior to his recent return to Cambridge, he was director of the nonpartisan Congressional Budget Office, which is charged with projecting the cost of legislative proposals. He has worked as an economist for the Federal Reserve Board and for the White House Council of Economic Advisers, and as Deputy Assistant Secretary for Economic Policy at the U.S. Treasury.

Karen Dynan, who earned her doctorate in economics from Harvard in 1992, is professor of the practice of economics in the Faculty of Arts and Sciences. She has served as a senior economist to the White House Council of Economic Advisors, and as Assistant Secretary of the Treasury for Economic Policy and chief economist of the United States Department of the Treasury. She is an expert on consumer spending and saving behavior, household finances, and macroeconomic policy. Professors Elmendorf and Dynan have both worked at the Brookings Institution, a Washington D.C. economic think tank, at various times during their careers in a variety of roles. They are also husband and wife. And with that, before we get into the serious questions, I have to ask: at which of the many institutions where you have both studied and worked—Harvard, The White House, the Treasury, the Federal Reserve, the Brookings Institution—did you actually meet? Was it here in Cambridge?

Karen Dynan: Such fond memories. Yeah, we met here in Cambridge. We were both graduate students. I still remember, I was entering data. I was sitting at a carrel at the National Bureau of Economic Research and I was entering data, and I remember when Doug came around the corner to see what was going on. As you can tell, since we worked at so many of the same places and we write together, we have a lot of similar interests, but that’s where it started.

Douglas Elmendorf: I’ll just add that our daughters, I will tell you that they learned too much economics sitting around the kitchen table growing up. But we think it was just a great addition to all their other education.

Jonathan Shaw: Right. In any event, welcome to you both.

Douglas Elmendorf: Thank you.

Jonathan Shaw: First, what is the state of the federal budget and the cumulative deficit now? When economists talk about a nation’s economic health, one of the measures referred to is the ratio of federal debt held by the public to GDP, the national output known as gross domestic product. Where does that number stand now, and how does that compare to past measures of debt to GDP?

Douglas Elmendorf: Federal debt held by the public is now about 80% as large as our annual output, our GDP, and that’s the highest percentage that we’ve seen in our history except for a few years at the end, and just after, the second world war. And that number has attracted a lot of attention and worry and concern.

Jonathan Shaw: Another important number is the ratio of federal revenue to GDP—what the government collects, mainly in the form of taxes. Where does that stand today relative to prior measures?

Douglas Elmendorf: Federal revenue is now just over 16% of GDP. That’s almost the lowest percentage of GDP that we’ve seen in the past several decades.

Jonathan Shaw: So putting those numbers together would suggest that there is some cause for concern, except that there’s a third number that economists focus on when assessing the economic risks of large debts and low revenues, and that is the cost of borrowing. Where does that stand?

Douglas Elmendorf: Yeah, so it turns out that despite a very high level of debt, the interest rate that the treasury needs to pay on all the money it’s borrowed is exceptionally low today. One of the lowest figures we’ve seen, again, in a number of decades.

Jonathan Shaw: So even with low unemployment, which we’re also enjoying right now, what will happen to annual deficits and the cumulative debt under current policies?

Karen Dynan: Under current policies, we’re going to see rising deficits and rising debt. And it’s quite understandable why that’s happening. It’s because our population is aging, and the Baby Boom is going to be retiring, which means they’ll be paying less taxes, but also collecting benefits through the Social Security system. They’ll also need healthcare, and there’ll be government programs providing that healthcare. So that’s a primary kind of driver of higher deficits and debt over the next several decades. So in fact, we’re going to see the ratio of debt to GDP grow to something probably close to 150%, according to projections done by the Congressional Budget Office.

Jonathan Shaw: Wow. Is there a specific debt level that would cause you to worry about U.S. economic stability?

Douglas Elmendorf: There’s no particular debt level that economists can identify as a sort of tipping point. There was a view a decade ago that maybe there was such a tipping point, but the evidence doesn’t really show that. Different countries have been able to sustain different levels of debt at different times, depending mostly on their underlying economic strength and also to some extent on the outlook for debt. So it’s not just so much the level of debt you have at a point in time, it’s where that debt is expected to go and whether people who are buying the debt have confidence that the country and the government of the country will be able to manage its finances in the years ahead.

Jonathan Shaw: As a hypothetical, what would it take to stabilize the ratio of debt to GDP at current levels?

Douglas Elmendorf: To stabilize the debt at the 80% of GDP it is now over the next few decades, you have to make cuts in spending or increases in revenue equal to hundreds of billions of dollars each year. So say the number was $400 billion as a ballpark. If you cut spending or raised taxes—either by 10% or each of them by five percent roughly, that would be close to the right amount to stabilize the debt. It doesn't seem very hard when you put it that way, but then you have to think about the pieces. There’s a lot of statements about federal spending or revenues or debt in the aggregate that sound fine that get much, much harder to work through when you get down to the specifics.

So if you think of federal spending, the biggest part of spending and the part that’s growing most rapidly is, as Karen described, spending on benefits for older Americans. So Social Security benefits and Medicare benefits are very highly valued by the recipients. They would not view a five or ten percent cut as a small matter, and they think with some justification, that they've been paying into these programs during their working lives and are those owed those benefits. If you think about Medicaid benefits—and Medicaid is the healthcare program that provides support for Americans with less income—Medicaid pays a large share of all of the nursing home bills in this country, so if you think of making big changes, big cutbacks in Medicaid spending, you have to worry a lot about what’s going to keep our nursing homes afloat. So you think about the specific elements of spending that are actually quite large. There’s a lot of political support for them, and so cutbacks that may seem like not a very large percentage end up being much harder to carry through when you get to the specifics.

Jonathan Shaw: And are those also the government programs that have grown most relative to the size of the economy?

Douglas Elmendorf: Yes, absolutely. We have an aging population already. The baby boomers have started to retire, and healthcare costs have risen faster than our economic growth for many years. And that’s not just a matter of public programs, it’s also true in private healthcare as well. The government programs’ healthcare spending per person has grown actually a little less rapidly than healthcare spending in the private sector. But basically, it’s an economy-wide phenomenon that’s been going on for years now.

Karen Dynan: And if I can just jump in, I think a natural question one gets after one explains what Doug just explained is, “Well, what about the rest of the budget? Isn’t there room there to make some cuts?” And I think the important thing to keep in mind there is that the rest of the budget is where we do a lot of important spending, and in particular investments in our economy and our society. So the rest of the budget is where we spend to educate and train people. It’s where we spend to build better roads and better airports. It’s where we spend to support research. So that category of spending as a whole, it’s low now by historical terms, and squeezing it further would hardly be a pro-growth strategy.

Jonathan Shaw: Right. Are demographic forces also the reason for falling revenues?

Douglas Elmendorf: Federal revenues are a smaller percentage of GDP today than at many points in our past few decades primarily because we’ve put through a number of tax cuts. Individual income taxes are the biggest part of federal revenues. They’ve gone up and down over time relative to the size of the economy. Corporate profits’ tax revenues are down over time relative to the size of the economy, and the big tax cut that was put through a couple of years ago has been an important factor behind that. Payroll taxes, which most people feel and see more directly, are actually up a little bit relative to the size of the economy, and I think that’s partly why people don’t feel that federal revenues are down, but they are. But they’re in some areas like corporate profits’ taxes, where people don't see them so directly.

Jonathan Shaw: Since the debt relative to GDP can’t grow forever, does the solution to this problem lie in cutting spending or raising revenues, including taxes?

Douglas Elmendorf: I think that most of deficit narrowing that we’ll have to do eventually will happen on the revenue side, and should happen on the revenue side. That’s not a super popular position. I pay taxes too. I’d rather pay fewer, all else equal, but the fact is that we have an aging population. We have people who are dependent on the federal support for their healthcare in old age, and if they don't have enough income to pay for healthcare themselves, those are important programs to support. Karen emphasized the value of the investments that the federal government does. Another piece I should emphasize is support for people of working age who aren’t making enough money or don’t have a job that can pay for their basic expenses. The safety net that we have and the programs that constitute the safety net are quite important in helping people who have not had the opportunities that some of us have had in our economy.

Moreover, there’s a growing body of evidence that those programs help the children in these families later in their lives. There had been concerns that the safety net might be discouraging work, because it provides some support if you’re not working. But in fact, this body of evidence that’s emerged in the last half dozen years has shown that the children in families that receive benefits often do better in their own economic success later in life than children in families where they did not receive those benefits. So there’s actually an investment in poor children through the safety net programs, in addition to the support for those children and their parents today. So all these elements of spending I think are quite important, and American people mostly like them. When you poll people about whether they want to raise taxes or cut spending on particular programs, not just in the abstract but program by program, the answers tend to come back people would rather raise taxes.

Karen Dynan: Another important point about the safety net is that it plays an important role stabilizing the economy when we see a downturn. Because of course people who lose their jobs, people whose incomes go down, they rely on these safety net programs, whether it’s unemployment insurance or it’s food stamps. But it’s not only an argument about the fact that it’s going to be good for these people, it’s going to be good for the economy to have these programs in place.

Jonathan Shaw: So it stabilizes the economy now, and it’s an investment in the future. That leads to my next question, which is your take on when changes in the trajectory of debt growth should be tackled. Why would sharply curbing deficits now pose a risk to continued economic expansion?

Douglas Elmendorf: Yeah. This is an especially interesting question today. For most of Karen’s and my professional lives, the established wisdom among economists, among financial market participants, and others, has been that it is important to reduce projected budget deficits and to do so quickly and vigorously. And the logic was clear, which is that you run budget deficits, the government is absorbing funds that could otherwise be used for private investments, and we’d want to have investment in the economy. If you see an unsustainable trajectory, shouldn’t you get on that right away and shift the trajectory? That’s a pretty compelling logic, but I think it has run up against some surprising evidence. The surprising evidence is that interest rates on Treasury debt have been declining now for 30 years, and it’s not just interest rates on U.S. government debt, it is interest rates on government debt in many countries in the world, and interest rates on non-government debt in the United States and many places in the world. This decline in interest rates is a sea change, I think, in the backdrop for fiscal policy decisions.

Karen Dynan: Yeah. Just to elaborate on that further, I would agree with Doug. It has been something that has in some sense it’s only recently received a lot of attention from economists, even though we think this downtrend in interest rates—we now think it’s been underway for a couple of decades—but really it wasn’t really recognized until we’d been in this terrible recession. And interest rates tend to be low in recessions, so that seemed natural, but then as we were coming out of the recession and the economy kept strengthening and strengthening and strengthening to where it is today, we didn’t see interest rates rise.

That’s when people started to say, “Maybe this is a long-term trend.” And now there’s been a lot of research that explains that this decline in interest rates is actually the result of forces that we know have been out there. It’s partly the result of population aging, and it’s partly the result of growing income inequality. So we think we understand it, and we think it’s a result of slow moving forces, and forces that are not going to reverse anytime soon. So it seems like it’s something that’s here to stay.

Jonathan Shaw: Okay. Now, the classic argument against government borrowing is that it has both social and budgetary costs. What are those costs?

Douglas Elmendorf: Well, the budgetary costs of borrowing is that the government has a lot of outstanding debt and it has to pay interest on that every year. And in fact, interest payments will be a growing part of federal spending in the years ahead. The economic costs, the ones I referred to, which is the borrowing by the federal government, can crowd out borrowing by private investors, by companies, by households that are looking to invest in their futures. And the interesting question now is, well, if interest rates are down, what does that mean about both those sets of costs? Should I answer that question?

Jonathan Shaw: Yes, you should. 

Douglas Elmendorf: So the direct effect of lower interest rates is that the financing burden on the federal government of any given amount of outstanding debt is lower. So interest payments today are a lot lower than most people projected half a dozen years ago, because interest rates are lower. I think that point is pretty well understood. The part that economists are just really digging into now, is how did the low interest rates change our view of the social costs, the broader economic costs of government borrowing? And the research that’s emerging has offered us a somewhat different view than we’ve had in the past.

Economists are still trying to work out what this new low level of interest rates implies about the social costs of government borrowing, but the research we’ve seen so far suggests that the social costs of government borrowing are a lot lower when interest rates are a lot lower, and the intuition I think is fairly straightforward. The cost of the government borrowing socially has been the crowding out of private investment, but when interest rates are low, that means there are a lot of funds available for private investment. Businesses that want to borrow money for investment today can get money at a lower cost than they've had in many years in the past. Households trying to borrow to invest in their housing can do so at a lower cost than in the past in many cases. And so the cost to the economy of the government borrowing are lower when interest rates were lower.

That doesn't mean that the government can then borrow an infinite amount forever. Economists widely agree that is not possible. But it does mean that the urgency of addressing federal deficits and debt is a lot lower than it was before. And that matters, because we can’t solve every problem at once. There’s only a certain amount of attention that policymakers can pay to some problems without skimping in their attention on other problems. And so if this problem is less urgent than in the past, then we should give it less attention in our policy making.

Karen Dynan: So Doug’s making the case for why it is that we can be more selective as we make policy, given these high deficits, and we don’t need to kind of dive in and cut back now. I think the other important point that’s relevant to these low interest rates is that we need to be thinking about the next recession. The economy is in a good place in a lot of ways. Over the past year, economic forecasters have gone back and forth about the odds of recession and where that stands. But certainly six months ago, people were quite worried about fallout from the trade war. So they didn't think a recession was the most likely outcome, but they certainly thought the odds were elevated. Concern about the trade war or has dissipated, but I think just the events the last few weeks with the coronavirus coming out of China has led to new concerns about new vulnerabilities related to supply chain disruptions that could destabilize the economy.

So it’s all a useful lesson that we will, at some point, see a shock that actually does put us into a recession. And where the debt comes in is that the fact that we are vulnerable in that way means that we shouldn’t pile on and basically have a fiscal contraction that would depress demand, particularly when our other policy tool for stabilizing the economy, monetary policy, is limited by these low interest rates. So you know, this was a problem in the last recession, which is the Fed’s traditional tool for stimulating the economy is cutting the policy rate; the policy rate last recession hit zero. It’s not that far above zero now, so they just have limited room to use that traditional tool. They have some untraditional tools which they will deploy, but those tools are not as well understood, and the risks around them are not as well understood, as their traditional tools.

Jonathan Shaw: Now, you both think that interest rates will remain low, but you also have a suggestion about how the government could hedge against the possibility that interest rates might rise. What is that?

Douglas Elmendorf: Yeah, so I think a natural thing for the federal government to do under current circumstances is to extend the maturity of the outstanding federal debt. The government can borrow at lower-than-usual rates, not only at short time horizons today, but in long time horizons, and the federal government should take advantage of that. I would also emphasize that Karen and I don’t say the interest rates will never go up at all, but they’re so low today, they could rise a fair amount from here and still be low by our historical standards. That’s actually what we expect to have happen, is some increase but still remain low by historical standards. But again, the federal government can hedge against some of the risk of rates rising more than we expect by locking them in to some degree through a lengthening of maturity.

Jonathan Shaw: To conclude, even if now is not the time to be raising taxes or cutting spending because of the low interest rate environment, looking ahead, Karen, aging populations are going to present an additional challenge to governments around the world in the United States. What strategies should federal policymakers be considering for the future in anticipation of this demographic shift?

Karen Dynan: Sure. Well, so the first thing we need to think about is these programs that we have to support older households, Social Security and Medicare, and we know those programs are, over the longer run, that they have coming financial problems. So we’re going to have to put in place policies that will fix the finances of those programs. That is something that policy makers, they need to be working on this question. In some sense, these questions are not that hard analytically. We know with social security it’s going to be a matter of raising taxes or cutting benefits, or probably some combination of that. Medicare is a harder problem, because the policies needed to curb healthcare spending are not as clear, but there are economists who have good ideas that would be a start on that problem.

I would say more broadly we should be thinking about what we can do to enhance the productivity of the working-age population, because when you think about it, who is going to be paying for the income support for the older population? Who is going to be paying for their healthcare? It’s at the end of the day it’s going to be the working-age population, so we should be making those investments that we talked about earlier, particularly in poor children and their parents, that would basically enhance not only their productivity as adults but also just the likelihood that they will be engaged in the labor force and working. We need to improve our infrastructure. We need to put money into research that would spur innovation, which would raise productivity growth. So those are also an important part of what we need to be doing.

Jonathan Shaw: Do you want to add anything about household savings and policies to encourage household savings?

Karen Dynan: Yeah, yeah, yeah. That actually is an area that I worked on when I was at the Treasury Department. So if you dive into the data and you look at the financial health of households that are approaching retirement age, so people in their forties and fifties, the data is a bit concerning. Much of the population appears to be struggling to save. It looks like, on average, households in the middle part of the income distribution, they’re doing some saving, but not as much as you would expect, and given that retirement is going to be upon them soon and they’re going to need something to supplement the money they’re getting from the government. I think that’s particularly true in light of the fact that retirement benefits have changed in this country.

If you went back a couple of decades, many middle-income families were expecting what’s called a defined benefit pension. They were expecting that their employer, whether it’s a private company or the government, would be making payments to them in retirement. And those plans, outside of the government, those plans have largely gone away, so that change has actually just strengthened the needs for households to be doing their own saving. So it’s a concern that so many households don’t seem to be able to save, and I think that the policy answer is that we need to find ways that make it easier for households to save. Economists have some good ideas there. Making access to employer-provided defined contribution plans, so like your 401k where the money comes automatically out of your paycheck every week or every month, seems to work well for a lot of households. But the unfortunate truth is only about half of employees working in the private sector have access to those sorts of plans. So, you know, one thing that government can do is nudge employers towards providing those sorts of plans.

Jonathan Shaw: I see. So things to prepare for the future. Thank you both for joining us today.

Douglas Elmendorf: Thank you very much.

Karen Dynan: Thank you.


Ask a Harvard professor is hosted by Jonathan Shaw and Marina Bolotnikova, and produced by Jacob Sweet. Our theme music was created by Louis Weeks. This second season was sponsored by the Harvard University Employees Credit Union and supported by voluntary donations from listeners like you. To support the podcast, visit If you enjoyed this episode, please consider rating and reviewing us on iTunes. Contact us with questions at

Sub topics

You might also like

Former Women’s Hockey Coach Sues Harvard

Katey Stone alleges gender bias in handling of abuse allegations that led to her retirement.

Remembering Helen Keller and Anne Sullivan

On a Radcliffe-Harvard memorial to remarkable figures

Harvard Confers 11 Undergraduate Degrees

Protestors now found in “good standing.”

Most popular

Former Women’s Hockey Coach Sues Harvard

Katey Stone alleges gender bias in handling of abuse allegations that led to her retirement.

Mechanical Intelligence and Counterfeit Humanity

Reflections on six decades of relations with computers

Remembering Helen Keller and Anne Sullivan

On a Radcliffe-Harvard memorial to remarkable figures

More to explore

Broadway Director from Harvard Adapting Disney

Broadway music director Madeline Benson on art and collaboration

How Political Tension on Campus Creates Risk Aversion

How overheated political attention warps campus life

Harvard Professor on Social Psychology for Understanding War

Two scholars’ extracurricular efforts in the Middle East