Eric Leeper on *A Fiscal Accounting of COVID Inflation*

Eric Leeper is a professor of economics at the University of Virginia, a former advisor to central banks around the world, and a distinguished visiting scholar at the Mercatus Center. Eric is also a returning guest to the podcast, and he rejoins Macro Musings to talk about his work on the fiscal accounting of the COVID inflation surge. Specifically, David and Eric discuss fiscal dominance during the pandemic period, how the fiscal theory of the price level explains inflationary trends, the backward and forward-looking fiscal accounting exercises, and more.

Read the full episode transcript:

Note: While transcripts are lightly edited, they are not rigorously proofed for accuracy. If you notice an error, please reach out to [email protected].

David Beckworth: Eric, welcome back to the show.

Eric Leeper: Thank you. It's good to be here, especially in person.

Beckworth: It is nice to be in person in the studio. Since the last time we were together on the show, you have become a colleague of mine, which is a pretty cool thing to say, that Eric Leeper is my colleague. And you have jumped right in. You've done a number of papers, policy briefs, congressional testimony. You participated in the Bennett McCallum Conference, so you've been a busy person. On top of that, you have all of your responsibilities back at UVA as well. You are a busy person, and I want to draw from some of this work that you've done recently. In particular, I want to look at your paper that's titled, *A Fiscal Accounting of COVID Inflation.*

Beckworth: I really enjoyed it, and I think our listeners will as well. We'll provide a link to it in the show notes. It's a great reflection and accounting of the inflation we had during the COVID period, the pandemic period. Along the way, as we're working through that, I will draw upon another note that you've written on fiscal dominance. The title is, *Fiscal Dominance— What It Is, and How It Threatens Inflation Control.* I want to begin by reading the first few sentences in your paper, and then we'll go from there. You state, “When American inflation began its upward march in 2021, economic analysts lined up the usual suspects.”

Beckworth: “Among the suspects for the sources of inflation were overheated markets, supply chain disruptions, shifts in consumer demand from services to goods, food and energy price rises, excessive corporate profits, and the perennial favorite, price gouging. Many of these candidates affected the evolution of inflation, and none caused it. We focus on the single cause, a large increase in federal COVID-related spending financed by new government borrowing, with little to no discussion of how, ultimately, to pay for the spending.”

Beckworth: That's your lead-off paragraph or two, and you go on to talk about roughly $5 trillion in unbacked spending, and then you account for it, two different exercises we'll walk through. I want to just park there for a minute, because you talk about how these other things may have affected the evolution of it, but they didn't cause it. This is, to me, the perennial issue in macroeconomics, identification, what really causes inflation, proximate cause versus a deeper primary cause. I think that's the point you're making here, right? We have got to look past the surface. Yes, we see supply chain disruptions, we see shifts from services to goods, but all of those things presuppose households have the ability to spend, and that's where fiscal policy came in. Is that right?

*A Fiscal Accounting of COVID Inflation*

Leeper: Yes, that is exactly where it came in. I think the first thing, though, that I want to say is that this was co-authored with Joe Anderson, who is a graduate student at University of Virginia, an excellent graduate student who will be looking for a job in a year. So, what I was trying to get at was the distinction between changes in relative prices and changes in overall prices. Inflation, by definition, is a steady increase in overall prices. We often look at things that are happening at high frequency, various shocks that hit the economy, that can move relative prices around. Then, it's convenient to say, "Oh, they must have caused inflation."

Leeper: And it's certainly true that if the price of goods suddenly goes up relative to services, that over some period of time, where people haven't fully adjusted their buying habits, they will end up spending more overall, and so inflation can pick up. But that can't be sustained. Eventually, they will adjust their habits to accommodate what's happened to the relative prices. And so, those kinds of shocks to the economy are inherently going to have transitory effects on inflation. Then, the other side of this, which you already alluded to, is that people still have to have the income to buy the stuff when the relative prices have gone up. Where is that coming from? To me, that's really getting at what the fundamental cause of the overall inflation is. As you suggested, I think it was because we handed out a lot of transfer payments to businesses and individuals in the economy.

Beckworth: If someone asked me, what is my prima facie evidence, or my go-to fact that suggests aggregate demand was a big part of the story, it is nominal GDP, the dollar size of the economy. If you look, I think it's [during] Q2, the economy had fallen about $3 trillion, almost $3 trillion if you round up, below its trend path. Today, it's about $2 trillion above that trend path. So you can think of, there was this contraction as we shut down the economy, and then we opened up, and quickly it jumps back up to where it should have been, and now we've gone above that. Where did that additional $2 trillion come from? It just didn't magically appear. It had to be injected somewhere.

Beckworth: And the recovery as well, the $3 trillion to close the gap. And so, fiscal policy played a huge role in that. And those numbers nicely line up with the $5 trillion or so that the government injected into the economy that we'll get to in your paper. You go on in that first page to talk about [how] this is a great natural experiment. We rarely get these in macro. Macro is really hard to get good empirical data. And you talk about this in the context of the fiscal theory of the price level, which you and others have authored and promoted. Is it also a great natural experiment to think about the macroeconomic regimes that you have talked about in your other papers, a monetary dominant regime versus a fiscal dominant regime?

Leeper: Yes, it definitely is, because one of the themes of the fiscal theory of the price level is that inflation is always and everywhere a monetary and fiscal phenomenon. What happened in COVID was that COVID hit, the Fed immediately reduced interest rates to the zero lower bound and pledged to keep them there for as long as they needed to. So, you had this automatic monetary accommodation going on, or what gets called passive monetary policy. Then, at the same time, you had the fiscal stimulus that we alluded to from the various COVID bills that were introduced.

Leeper: Critically, the atmosphere around fiscal policy was different than it often is. A lot of the standard procedures that Congress follows were suspended. In fact, the initial CARES Act was approved by a voice vote, unrecorded voice vote, in the House. That was over a trillion dollars. That's an awful lot of spending to be done anonymously like that. And on top of that, the communication coming from policymakers was, "Hey, we're in a crisis," which admittedly we were. And, because we're in a crisis, we're not going to do things as we normally would.

Leeper: And, if you lay on top of that the idea that President Trump had his name on some of the checks, it was pretty clear that these transfer payments were meant to be gifts. They weren't meant to be loans that would have to be paid back with interest in higher taxes in the future. You don't typically put your name on a check when it has attached an IOU for future taxes. That communicated to people that their permanent income had gone up. Well, they're going to want to translate that permanent income into consumption. That's what standard economic theory tells you.

Leeper: This is where some of the relative price phenomena came into play, because they couldn't spend it much on services. I couldn't get my hair cut when the barber was closed, and so I bought new furniture, and I renovated my house, and actually, in my case, I built a gym at home. That then drove up the price of goods relative to the price of services. So, I think that there were supply chain issues, but there was also a relative shift in demand at the same time.

Beckworth: So, you note in the paper how this transpired, the voice vote, as you just mentioned, but also the way it was approved, that there wouldn't be any back, in the S-PAYGO. Maybe walk through the PAYGO, S-PAYGO laws and how they operate, and why they were important at this time.

The Basics and Importance of PAYGO and S-PAYGO

Leeper: The idea of PAYGO, which came about in the 90s, was that if you propose new spending, then you had to also propose either offsetting declines in some other form of spending or offsetting increases in revenues. That got suspended permanently and they moved on to this thing called the sequestration PAYGO, which, in practice, is supposed to mean that you make a proposal for new spending, and the spending cuts would automatically occur across the board to offset it. But, to some degree, even that was suspended during the COVID period, certainly on the initial CARES Act that was done. And so, yes, the whole tenor of how spending decisions were being made was different under COVID. And that, to some degree, is to be expected because it really was an emergency.

Beckworth: Yes, so you can think of this as fighting a war, a public health war during World War II, similar mindset. The question comes, well, when was it too much? Was CARES enough and ARP too much? And going back to this distinction that you make, you note in the paper that the CARES Act was completely exempted from S-PAYGO, so clearly no expectation of raising taxes in the future to recover this. Then ARP, the 2021 [law] under Biden, was not exempted, but the sequestration was pushed to the future, and it sounds like they keep kicking the can down the road on that. So, effectively, two big programs with no backing as far as the eye can see, which means somehow it has to get paid, which is in the paper.

Beckworth: You note that a lot of it gets paid through inflation, the inflation tax. But just to go back to the macro regime, so you have a monetary policy-dominant regime and a fiscal-dominant regime. And just to make this clear, because I think it's a really nice, clean way of looking at it— and, again, I'll tell listeners to go check out your paper, *Fiscal Dominance— What It Is, and How It Threatens Inflation Control.* We'll have a link to that as well. But we can think of these two regimes in the following way. When you have monetary dominance, which is what we typically think we have, and it's certainly what Fed officials think we have, Treasury manages fiscal solvency and the Fed manages the price level.

Beckworth: Those would be the two tasks. Think of those two tasks, who manages the price level and who manages fiscal solvency. Under monetary dominance, Treasury is keeping the US government solvent and the Fed's keeping inflation. In fiscal dominance, it flips. The Treasury begins to manage the price level effectively, because it's issuing all of this debt and driving up prices. And the Fed is working hard to keep the US government solvent. The roles completely flip. And I guess the natural experiment is that we saw that happen, at least for a season, during this period. But if you ask Fed officials, in fact, they were asked. They were like, "No, no, no." Governor Waller said, "No, there's no fiscal dominance." Jay Powell [said] late last year, "Oh, no fiscal dominance." But I think we have good reason to believe this was a period of fiscal dominance. And we've seen fiscal dominance in other places, right? Hyperinflation clearly is a case of fiscal dominance, World War II, the US government, fiscal dominance. But, in your mind, that was a nice, clean illustration of fiscal dominance.

Fiscal Dominance During the COVID Period

Leeper: Yes, I do. If you look at how inflation has behaved, it behaves very much like what you would expect from a theoretical model under fiscal dominance. What happens is that the government issued $5 trillion in new debt. There was no expectation that primary surpluses were going to rise in the future to pay off that debt. So that debt has to be revalued, has to be devalued, which happens through a combination of lower bond prices and a higher price level. That's because, basically, the current value of the goods that will support those new debt issuances hasn't changed, and so the real value of debt can't change.

Leeper: One piece of corroborating evidence, which isn't exactly in that paper, is if you look at the market value of marketable government debt as a share of GDP, and you compare what it was at the end of 2019 to what it is today, that ratio has fallen, despite the fact that, over that period, we now have upwards of $7 trillion more government debt outstanding. The only way that can happen is if you have had some combination of growth in nominal GDP— and we've had both real and price growth— and lower bond prices.

Leeper: And so, that's exactly what the fiscal theory would predict. Unfortunately, what has happened is that Congress did what we call an unbacked fiscal expansion, and then turned to the Fed and said, "Okay, now you mop it up." But what the theory tells us is that, in the absence of some kind of fiscal consolidation that ultimately raises primary surpluses to soak up that debt, there's nothing the Fed can do to permanently offset the inflation. They can change the timing of it. And so what the theory tells us is that the Fed's increase in interest rates serves to reduce inflation, at the time, by pushing it into the future. The way we've seen that show up is that bond prices have fallen, and so that is consistent with an expectation that there's going to be higher inflation in the future.

Beckworth: So, we see evidence from the bond market of this idea, fiscal theory of the price level, fiscal dominance. I've heard someone else make this case as evidence for fiscal theory of the price level, that what we've seen with actual inflation also fits it, that we have this one-time helicopter drop, so you'd expect a period of inflation going up, and then eventually it comes back down and returns to some kind of baseline trend. And we see that. Is that also, you think, consistent with the fiscal theory?

Leeper: Yes, except there's one proviso, because that thought experiment that you just described typically maintains that the Fed doesn't do anything with the interest rate. Basically, the inflation just burns itself out. Once the revaluation has occurred, inflation goes up, [and then] it comes back down. But, that isn't what has happened. Instead, the Fed has raised the interest rate. And so, what we're now seeing— another way of thinking about what's going on right now— is that interest payments on the debt are exploding. Then, the question becomes, how are those interest payments going to get financed?

Leeper: So far, they've been financed by just issuing new debt. If that continues, you can expect more inflation. If, on the other hand, what happens is that Congress starts to see that they've got to pay the bills from borrowing, and therefore they can't spend money in other ways that they would like to spend it, then they do something to either raise revenue or cut spending. How those interest payments get financed is, I think, the critical question for thinking about inflation going forward.

Beckworth: So, the way it gets transferred to the future, the inflation that would have otherwise occurred, is through interest rates and through the interest payments that the government's making to the public. That includes not just Treasury holders, but the Fed making interest payments to banks, to repo market participants. Would that also affect households in any way? Are households maybe sitting on excess liquidity, waiting to spend? [Are] there any implications there?

Leeper: It's certainly shown up in mortgage rates, and so that has had consequences for the housing market. Interest rates across the board have risen relative to what they were before COVID.

Beckworth: Well, that's a nice segue to the next question I wanted to touch on before we get into the evidence in your paper. You do two exercises to illustrate the effect on inflation. But, the idea that if it's not backed, these government expenditures, it will lead to inflation, and the question is, why? You've kind of already answered it, but maybe go through it again in terms of the fiscal theory of the price level. Maybe restate the fiscal theory of the price level and why it is, if it's not backed, that we would expect inflation.

Inflation as Explained by the Fiscal Theory of the Price Level

Leeper: Okay, I'll try to do this in a couple of different ways. One way is, as I was saying before, that if it gets communicated to people that the transfers they're receiving today from the government are not going to be offset by taxes in the future, then individuals feel wealthier. And so they try to convert that wealth into consumption, because that's what makes them better off, and that act increases aggregate demand. That's thinking about this purely as a wealth effect. From the consumer's perspective, I think that's probably the right way to think about it. There's also thinking about it from the perspective of financial markets.

Leeper: The government pumped a bunch of dollar-denominated debt into the economy. That debt got absorbed, but at what price? The demand for government debt is not perfectly elastic. So what we saw was that people were willing to buy that debt, as long as the price of that debt went low enough. And the reason that the price had to go down was because the expectation that people had was that when that debt finally matures and gets redeemed, the dollars that they're going to get back are going to be worth less in terms of the goods and services that they can buy than they would otherwise have been. That tells people, "Okay, I'm willing to hold this debt as long as it doesn't cost me as much, because my expected real payoffs have gone down."

Beckworth: So, you're telling a wealth effect story for households, a portfolio story for markets, and the return they can expect. Could we also frame this in terms of a simple, maybe even crude, quantity-theoretic approach, that you permanently increase the liabilities of government, there has to be a price adjustment somewhere, just kind of a broad view of a quantity theory of government liabilities?

Leeper: Yes, actually, I would add that. I do often think about this in that context, because nobody disputes that Treasury bonds are a liability of the government. Nobody disputes that the bank reserves that the Fed created to buy government bonds are also a liability of the government. Both of these liabilities these days pay interest. Then, you've got to think, what are the offsetting assets? If the government's going further into debt, there have to be assets that offset that. That's where the primary surpluses come into play. And those assets don't have to be present today. There has to be some assurance that, as these bonds mature, those assets will be present. And so, that's the other way to think about it.

Leeper: Now, here's the difference. We think about those assets as being denominated in units of goods. The liabilities, though, are denominated in dollars. So, the price level can adjust to equate the real value of those liabilities to the real value of the assets that back them. Now, whether you want to call that a quantity theory or not, I don't know, because the quantity theory doesn't really bring the asset component into the picture.

Beckworth: What the quantity theory does… and I'm, of course, thinking of the old monetarist quantity theory here, and I think you can do it in terms of rational expectations. If you expect a permanent increase in the monetary base, there would be some kind of proportional increase in the price level. And there's a lot of work that was done on that by the old monetarists. But, I think what's interesting is that I think of your work, and I think of the New Keynesians, particularly the people who worked on Japan, and I've collected a bunch of quotes for a paper I did, but they all stress a permanent increase in the monetary base.

Beckworth: Now, this is before 2008, when the monetary base was mainly currency. It wasn't reserves with earning interest. But even Paul Krugman's 1998 paper, *It's Baaack: Japan and the Return of the Liquidity Trap,* it was all about the expectation of a permanent increase in Fed liabilities. Michael Woodford makes a similar argument. He points to Japan— and I've mentioned this on the show before— He mentioned the original QE, 2001, the Bank of Japan's balance sheet expanded. Then, it returned to that trajectory as if there had been no increase. He said that it’s as if the public expected this to be a transitory increase and then a decline, so you've got to have this expectation that it’s permanent. That will lead to the price level effect, and more importantly, the aggregate demand effect, which would then get them out of the zero lower bound. So, all of this discussion around the zero lower bound, the need for permanence, I think it ties back into what you're talking about, and particularly if you think of it in terms of like a government budget constraint, right?

Leeper: Yes, and I think, in some ways, you can think about the fiscal theory as generalizing what you were just talking about, generalizing the quantity theory, because even if what is getting pumped into the economy is dollar bills, green pieces of paper, which don't pay any interest, well, if taxes adjust in the future and are used to retire the new money that gets pumped in, then those liabilities are backed by assets. And so, what the fiscal theory says is that what you have to always be thinking about is, what are the assets that are backing the liabilities of the government, whether those liabilities are currency, or reserves, or Treasury bills, or notes, or bonds. And so, the example that you talked about with Krugman… he was basically maintaining the assumption that they weren't going to be backed by future taxes. But, they could have been, right? Then, you wouldn't get the aggregate demand stimulus.

Beckworth: Yes, and I think that that's such a key insight. In graduate school, I learned time series, we did vector autoregressions. I know you used to do a lot of them, too. And I remember one exercise we did in William Lastrapes’ class, if he's listening out there. We had a fun time. We would do the long run—

Leeper: Hi, Bill.

Beckworth: We did the long run restrictions, a bunch of quad restrictions. And we'd put money in, and you'd have this long-run restriction, and eventually, it would peter out. But what was really cool is you put the monetary base in, and it would have a temporary effect on real GDP, but what would pop out was this permanent effect on the price level. You're like, "Oh, it's the quantity theory. It's beautiful." But, when you step back, and you think, "Okay, but what's going on behind the scenes in terms of that government budget accounting?" I mean, I think if I redid that VAR, I maybe could take out the monetary base and put in government liabilities and get the same result.

Beckworth: The Fed is effectively doing little helicopter drops, buying up those assets permanently. In fact, you can think, I believe, as we go through time and this demand for currency just naturally grows, the Fed is doing a bunch of little helicopter drops. It's buying up Treasuries, injecting currency into the economy. So, I like your point. The fiscal theory of the price level is a generalized theory that captures all of these other insights. We've all been poking at the elephant, and you're telling us that the elephant is the fiscal theory of the price level. Alright, fun stuff. Let's go to your actual exercises using this framework, and you do two. You do one where you take the accounting approach, looking backwards, and then a forward-looking one. Let's start with the backward-looking exercise. How do you do it, and what do you find?

Fiscal Accounting: The Backward-Looking Exercise

Leeper: What we did there was we just mimicked this terrific work by George Hall and Tom Sargent, who have looked at US government debt since the inception of the country. They've done an enormous number of different accounting exercises. The idea is basically that you take debt, say, in 2019, debt as a share of GDP, market value of debt as a share of GDP, and you can take it at the end of 2023. And you say, okay, that debt-GDP ratio went up by, say, I'm making numbers up, [but] 20%. Now, what drove that increase in the debt-GDP ratio? To answer that question, you look at the different components of the government budget identity.

Leeper: Obviously, nominal bonds [are] part of that, interest payments on nominal bonds are part of it, inflation is part of it, growth in bank reserves is part of it, and real economic growth is part of it. And so, what we did was we picked two dates, one right at the beginning or just before the start of COVID, and one at the end of our sample period, which was maybe the third quarter of 2023, and we said, "Okay, how much of that change in the debt-GDP ratio was due to each of these components?" Not surprisingly, what we found was that economic growth did contribute some.

Leeper: Oh, I forgot one element, kind of an important one, the primary deficit. Sorry about that. So, what we found, not too surprisingly, was that what drove debt more than anything was this massive increase in the primary deficit as a share of the economy. But, two other very important components were that there were negative yields on government bonds and there was higher inflation. Those taken together generated very large negative real returns on the Treasury portfolio. And Hall and Sargent have a paper specifically looking at that issue during COVID, and they do a much better job than we did, going into a lot more detail. That's a paper that needs to be read by people, especially policymakers, to see who is actually paying for COVID.

Beckworth: Right, and I think that they have several versions of that paper, but I think one of the titles is, *[Three World Wars: Fiscal-Monetary Consequences].* I actually had George Hall on, so we'll encourage listeners to go back and check that episode out as well. So, very fascinating work. One of the key insights I got from that, and from what you did, is the fact that the debt-to-GDP ratio actually went up and then it declined, due in part to the inflation. It actually brought us back down to a level that was surprising given how much we actually injected via fiscal policy.

Leeper: The important difference that they highlight is that after both World War I and World War II, the government ran primary surpluses. That really contributed, also, in an important way, to bringing the debt-to-GDP ratio down. If you look at Congressional Budget Office projections now, we have primary deficits as far as the eye can see. So, it's not obvious that the third war, COVID, is going to look like the first two wars. Let me just add one more important point, which is that what they're really looking at is how close was fiscal policy to what an optimal fiscal financing theory would tell you that you should do. And what that theory, due to Barro, says is that if you have a temporary increase in spending due to, say, a war, then the optimal thing to do is to finance that through debt, but then adjust spending and taxes so that you run primary surpluses afterward.

Beckworth: And, again, that's what makes this period different is that after those war times, we tended to see more emphasis on fiscal consolidation, whereas if you look at the CBO projections, which if we get to, you'll criticize… but what they do show is that, no, that's not happening. As far as the eye can see, we’ve got growing primary deficits, growing interest expenses, and the threat of fiscal dominance in the future if things don't change. And so, circling back to this point, though, about these fiscal costs, a topic that we have visited several times in the show with previous guests, is who's bearing the cost of this interest rate risk that we now have seen?

Beckworth: So, we saw banking turmoil in March of last year. We see the big losses on the Fed's balance sheet due to interest rate risk, both of these things due to interest rate risk. If the Fed's holding bonds, long-term bonds would be 1%, they're paying out 5%. They're bleeding. They're literally losing money, and that's less money going to the Treasury, which then contributes to the deficit. I've had people on that have discussed [this] and [said] this should be getting more attention. The Fed's losing money, taking losses.

Beckworth: I've had other people along say, "It's not that big of a deal, because if the Fed hadn't been holding them, the public would have." They would argue that, either way, someone's going to bear this cost. If the Fed's taking a hit, it's passed on to the Treasury, and the Treasury's passed on to the taxpayer. If the public were holding these bonds, they would be taking the loss as well. Either way, the taxpayer is going to take the loss. Does who holds it, its distribution, matter that much? Should we care about the losses, or is it just a big macro accounting exercise, and we just put the losses somewhere on the map?

Should We Care About Fed Losses?

Leeper: I think there are reasons to be concerned about the losses, at least two reasons. One is that there are distributional consequences. Typically, we like to think that when government takes decisions that have distributional consequences, those decisions are deliberate. Well, it's not so obvious that there's any deliberation going on that led to this outcome. At this point, it looks very much like a lot of the losses— this is what comes from Hall and Sargent— have been incurred by bondholders. And who are they? They're people like me and you, and people who have pensions and 401(k)s, as opposed to the taxpayers as a whole, which might get more spread through the population.

Leeper: The second reason regarding the Fed's balance sheet is more of a political economy concern, that back in 2017, I testified before the House, and at that point, they were just giddy about how large the remits were from the Fed, even though the gist of the hearing was, should the Fed balance sheet be as big as it is? There was at least one representative who said, "Look at the size of these remits, they should have an infinite balance sheet." Well, now those remits are zero. And I'm not saying that the Fed even brings this into-- whether it even enters the back of their minds. But it certainly is entering the minds of Congress, especially at a time when we're running chronic deficits, and they're not getting any revenue from the Fed. So, it could lead to some uncomfortable interactions between the Fed and Congress. And I think that threats to the independence of the Fed are overblown, typically, but somebody who worries about that could--

Beckworth: --That'd be an issue for them. Yes, I agree. And I do think it's interesting that in the US—while there's some attention given to the Fed's balance sheet— currently, there's many other issues that supersede that, at least compared to discussions overseas. At the ECB, they're actually reconsidering how they implement monetary policy, the size of their balance sheet, the floor versus corridor system, ample reserve versus scarce reserve [system]. In the UK, if they lose money, the Treasury over there explicitly recapitalizes it. Here, we do some accounting tricks. And so, It's more of an issue overseas than it is here.

Beckworth: And It's interesting that that's the dynamic right now, maybe it will become more politicized. Who knows? Maybe Trump will bring it up as a presidential issue. So, it's interesting to see that take place. And regarding the comment that this person gave at the congressional hearing, an infinite balance sheet, there's other issues, too. I really don't want the Fed to have such a big footprint on the financial system. I mean, it's true that the Fed did make a lot of money for the government by basically stepping in and becoming a hedge fund and stealing some of the market activity that maybe the private sector would have otherwise done.

Beckworth: And you can imagine, in the limit, it could take all of that, that term premium spread, away from other financial firms. And that's a question that I think should be debated. Do we want that or not? And this, in general— going back to your point about deliberating on how this happens— the Fed is effectively becoming a manager of public debt to some extent. It's not the full… And is that something we delegate to the Fed or do we delegate that to Treasury? And, again, it’s something worth talking about and why I think we don't want to continue to have an enlarged balance sheet. And so, I think there's lots of good questions surrounding that issue. But going back to your paper and the cost of the pandem, so we've already looked at the rearview perspective, looking at the accounting exercise. Let's look forward. And in that case, you used the fiscal theory of the price level framework. Walk us through that part.

Fiscal Accounting: The Forward-Looking Exercise

Leeper: Okay, well, we've alluded to this a couple of times, but I'll be more explicit. What the fiscal theory of the price level says is— let's say we want to think about debt as a share of GDP— it says that the market value of outstanding debt as a share of GDP has to equal the expected discounted present value of future primary surpluses as a share of GDP. And so, if you take that as given, then what we can tease out from the data— as we see that market value of debt as a share of GDP change over time— what we can tease out from that is, well, what must have been happening to people's expectations about future surpluses or discount rates? And so, that's a different accounting exercise, and I want to emphasize why it's important to think about both exercises.

Leeper: The backward accounting is telling us, how did we get to where we are today? That's always good to know, but it isn't going to really have much to say to policymakers other than, well, this is what you did before. The forward perspective is really something that can inform policy, because they can change what people expect about future surpluses by changing legislation. And so, if what we see is, well, we're getting a lot of increases in the price level because more and more nominal debt is getting issued and there are no plans for raising surpluses in the future, [then] you can say to Congress, “Look, this is where this inflation is coming from, and we have a fix for that,” which is, do something about future surpluses in a credible way. That's why I think that the forward-looking perspective, in some ways, is more useful for practical policy questions.

Beckworth: What is it showing right now, if you look forward?

Leeper: Well, we're continuing to see that— I use this shorthand of the price of the government bond portfolio, which I'll calculate as, you take the market value and divide it by the par value. This is a very crude measure, but it is now at levels that we haven't seen since the 1970s. On top of that, in the '70s, when that price of the bond portfolio bottomed out, it was always at the beginning of a recession. Now, we're seeing it at those levels despite the fact that the economy is growing well, and so it doesn't look like anything we've ever experienced. Then, you have to go to the next step and say, well, what does this tell us? And there's also a puzzle here, because the overall value of government bonds— meaning the whole portfolio across all maturities, et cetera— is declining, but why aren't we seeing the 10-year bond yield go up?

Beckworth: It went down for a little bit at the end of last year, yes.

Leeper: Exactly, and so that's a bit of a puzzle, and I don't have an answer for that, but I think there are just some very troubling signs coming from the bond market. The other issue that I've commented on in testimony was that the last two auctions, the Treasury auctions, didn't go well. One of them was a tailing auction, which means that the final bond prices were lower than people expected they would be. And then— I think it was the December one— they couldn't issue as many long bonds as they wanted to, because there just wasn't any demand. Lay on top of that the fact that foreign demand for Treasuries has been— well, Chinese demand in particular has dropped quite a lot over the last several years. Japanese demand has been flat, and we have about $9 trillion in debt that comes due in 2024. Add on to that the $1 trillion-plus that we'll have in the deficit that has to be financed and the fact that the Fed is undertaking QT and putting more bonds into the system. Easily, we have to finance about $10 trillion, which is roughly a third of the stock, in one year.

Beckworth: That's a very grim picture you just painted for us.

Leeper: Thank you.

Beckworth: But let me push back just a little bit on that grim picture. I mean, it is grim. I can't push back on that reality, but what I can look at is that inflation is coming down. We talked earlier about how you have a one-time helicopter drop. Inflation goes up, it comes down, a permanent price level effect. Why not see more changes in the trend inflation rate? Maybe it's just too soon, I don't know, but PCE is now under 3%. If you do the last six months, on an annualized basis, we're pretty close to the 2% target. That's another mystery, it seems.

Leeper: One thing to keep in mind is that our theories typically embed rational expectations, and one of the maintained assumptions is that, whatever policy regime is in place, it will remain in place. I think that, right now, there's a serious question about, what is the policy regime? Where are we going? Congress can't even pass a budget. So, I think there's an enormous amount of uncertainty right now.

Beckworth: Right, the increased polarization has led to increased dysfunction in Congress, and it does seem unlikely that Congress could ever get together and work on some long-term structural fiscal plan, which is what we sorely need. And we've had other guests on the show that say, look, and if you really want to do that, you can't just tinker at the edges. You can't play with discretionary spending. You've got to actually touch Social Security. You've got to touch Medicare, Medicaid. You've got to touch the big unfunded liabilities, and that just seems off the table, both on the left and the right.

Leeper: Yes, and so we talked a lot about how what the fiscal theory says is that people are thinking about what future primary surpluses are going to be when they decide how valuable government bonds are. Well, where are you going to get that information from? That's hard. I think that this is a very unusual time, and one thing that we know about our theories is that they break down during unusual times, in part because people don't know what the policy regime is.

Beckworth: So, you're suggesting that, maybe, people have some kind of adaptive expectations. They're still hung up on the past regime and they haven't quite caught up to the reality of all of these grim facts you just outlined for us.

Leeper: Well, that's one possible view. I think that there are probably many possible views out there about this.

Beckworth: Yes, well, it is striking. Let me add to your grim picture, another interesting observation, a striking observation. If you go to FRED or any data website and plot two series, plot the unemployment rate against the budget deficit. They typically are inversely related, as when you have a big recession, unemployment goes up, the deficit increases, it turns negative, and so it makes sense. The business cycle drove the big changes in the deficit, but since the mid-2000s or so, that relationship is breaking down and is projected to increasingly break down, so it's this peacetime increase in spending.

Beckworth: In fact, even last year, we saw an increase in the deficit from $1 trillion to roughly $2 trillion, if you include, I believe, student aid, student relief. Let me throw one other mystery out there that has been covered by some, I believe, in this field. And this is pre-pandemic, I believe, but if you go out and measure the market value of all of these US Treasury securities, as you mentioned earlier, and then you go and look at, say, CBO forecasts of primary surpluses, it doesn't add up. The discounted present value of the primary surpluses simply doesn't get you far enough to what the market seems to value. Maybe the reason is that the market hasn't caught up to reality yet. Maybe that's one explanation.

Beckworth: Another explanation— and I'm curious to see how far you think this goes— is it's the liquidity services that the Treasury provides. So, Markus Brunnermeier and some others have a paper [titled], * [The] Fiscal Theory of the Price Level With a Bubble.* I believe that even Governor Chris Waller had a paper earlier where they look at the liquidity premium, and so you add an extra term on that, to that theory. But do you think that's enough, both empirically to cover that, or just in practice? Can we depend on the goodness of foreigners buying up our debt to keep this from blowing up?

Leeper: No, I don't think we can, and that's part of what we're seeing happening. In the absence of some global crisis, okay, I certainly believe that there are all manner of services that government debt provides that the simplest version of the fiscal theory doesn't take account of. We have very bad models of those services, assuming that people just derive utility from government debt, which is what is often done. It gives us a little wedge term that we can carry around in our model, but it doesn't really tell us very much. And you have to ask yourself, well, are those services going to persist regardless of what the state of fiscal policy is?

Leeper: I mean, one reason that they serve those purposes is because government bonds have had a reliable, stable value over time. But once you're in an environment where that may not be true, the private sector will innovate. They'll come up with substitutes that are safer relative to this highly uncertain return that you might end up getting with Treasuries. I agree completely that there's a lot of things that Treasuries do that we should be modeling, but I don't think that we really have a very good fix on how those services get valued when we're in a situation where fiscal policy seems to be just evolving autonomously.

Beckworth: So, you made an interesting comment there that you think that, at some point, the private sector will step in and provide alternative stores of value, safe assets, maybe, even. And I'm curious as to what you think those would be. And I ask that just because the scale of the US global dollar market, Treasuries, but also repo markets, even private bonds… globally, it's just so much larger than anything else. There's almost like this first mover advantage, network effect. Let me put it this way: If there were to be a serious run on the dollar, where would investors run to? If you're all in one bank, it assumes that there's a second bank you can run into. There'd have to be either some big price adjustment on their assets or an increase in quantity in other assets.

Leeper: I think it'd be both. But, you know, Apple bonds? I would guess that those are pretty safe.

Beckworth: Enough of them to--?

Leeper: Well, no, no, I'm not saying that there are enough of them, but you've got to remember that there hasn't been any reason to have more, because Treasuries were reliable. And so, we just don't know how they would innovate. Now, the reason I said in, “the absence of some global crisis,” was part of what you have now brought back into the conversation, because you still think, “okay, yes, maybe fiscal policy in the US is a mess, but where is the world going to go?” And the US has been milking that cow for a long time.

Beckworth: That's part of the problem. It's been there. That cow's been there, and it's created, I think, bad habits.

Leeper: And to be honest with you, I did paint a grim picture, but deep down inside, I actually believe that, when interest payments get high enough, even a highly dysfunctional Congress will do the right thing.

Beckworth: Wow. You're optimistic.

Leeper: I am. Now, the question is, what has to happen before that occurs?

Beckworth: Right, right. So, I hope you're right, and therefore, maybe we should have a nice run with higher rates for a while. But, I had Brian Riedl on, and he does budgeting calculation, more of accounting exercises. And his view is this: that because the US is so dysfunctional on this issue, we're never going to get there. Now, maybe your story is the salvation story here of how we get through it all. But he worries that we're going to get to some crisis, and we're going to usher in a new set of taxes, so either we cut spending or increase taxes to solve this problem.

Beckworth: He thinks that we're going to become like Europe. We're going to have a whole new spate of value-added taxes, and it will be borne largely by the middle class, because they're the ones that really will be able to pay the bill. He does not think that taxing the rich is going to solve it or other changes on the margin. You've got to actually go after the big tax base, and that's the middle class. His fear is that we'll end up like Europe, but without the benefits of being Europe. We'll have all of these value-added taxes, but all of the benefits will flow to senior citizens. They won't flow to free education or health care. So, it's going to be the worst of all worlds, is his very dire outlook. But you've painted a more positive one, that we'll get our act together when we begin to feel the pain of higher interest payments.

Leeper: Yes, I mean, I don't have a crystal ball. But, I'm just looking back… after the Reagan tax cuts, there was a series of revenue enhancements, because Reagan couldn't raise taxes, but he could enhance revenues. Then, George H.W. Bush, “read my lips,” did a tax increase. Then, Clinton took over and did a major reform on the spending side. And that occurred at a time when— at least back then— we thought, “My God, the political system is so polarized.” We had Newt Gingrich running the House and a liberal Democrat in the White House, and yet they managed to do something.

Leeper: And so, I do believe that, eventually, Congress will do something. And I think that they all recognize that you can't just make marginal changes, that something fundamental has to change. And I'm sure that entitlements would be part of that, and maybe there'd be some other taxes, maybe a consumption tax. But the resistance to taxes in the United States is so fundamental to our nature. It was never fundamental in Europe like it is here. I don't see us going down that path. I think that's overly pessimistic.

Beckworth: Okay, well, thank you for the optimism and the cheer that you bring to the audience here. Alright, in the time we have left, let's transition into the testimony that you gave recently to Congress, on the CBO's projections and the issues it faces, and how it could do better. So, how could the CBO do better?

How Can the CBO Do Better?

Leeper: Let me start by explaining what they do when they construct their long-term projections. They come up with projections for the economy. So, that would include things like GDP growth, inflation, interest rates, a whole slew of things. It's quite a large spreadsheet. They then take those as given, and from them, they have a very detailed analysis that then produces what the path of expenditures and revenues will be, taking current law as given. Sometimes that current law means a tax cut has a sunset provision, sometimes not. So now, we've got those economic variables, and we have a path, say, for the primary deficit. Then, they basically stick it all into a spreadsheet to get the level of debt-GDP that is required— given those economic assumptions and the surpluses— to satisfy the government budget constraint.

Leeper: And so, what's missing from this is any feedback from the fiscal variables to the economic variables. And so, when it comes to the actual fiscal projections, it is really a pure accounting exercise. It's not an economic exercise. And that's why what we see… you know, back when I first criticized this in 2010, they were going out 75 years, and they had some projections of the debt-GDP ratio getting to 300%, 400%. You just see this exponential growth. And my argument back then, and my argument today, is that these accounting exercises aren't useful for policymakers. There are two possible reactions you could have.

Leeper: You could say, “Oh my God, look at those deficits, that debt level, we have got to do something.” Or you could say, “You know, debt gets to 300% of GDP, inflation is 2%, economic growth is 1.8%. What's the problem with exponentially growing debt?” And so, what we really need is to put all of this into a coherent economic model that ensures that there actually is an equilibrium, so that the debt path that comes out of that model is one that can logically occur, unlike what the CBO now produces. And then, that would open up the door to the CBO offering a menu of policy options.

Leeper: Let's say you want to stabilize debt-GDP at 80%. Well, here are a bunch of different ways that you could do that through changes in taxes and spending, and here are what the macroeconomic consequences of those policies would be. You could put this into a model with heterogeneity, like an overlapping generations model. You could then report on some distributional consequences of those policies. But, the bottom line is that, then, what you're giving Congress is basically what the staff at central banks give their governors all the time, which is, if you do X, then our model says this is what the likely consequences will be. And we know there are all kinds of shortcomings with that, but at least the logic hangs together. That's what's missing right now.

Beckworth: That's a great insight. Has the CBO responded, or [has] anyone in Congress shown interest in moving in this direction?

Leeper: No.

Beckworth: So, you keep fighting the good fight and hope that one day they will, but yes, it's true. You look at these forecasts and you're like, man, 200% debt-to-GDP, and everything's just going to go along fine, ho-hum, hunky-dory. There's just no way, right? At some point, people will either respond politically, or there'll be economic consequences to this, endogenous to the higher debt growth path. We'll never get that high, probably, in the real world.

Leeper: Let me say one more thing about this, because what these kinds of projections encourage is for people to stand up and pontificate about [how] fiscal policy is unsustainable. That's a completely meaningless statement, because people are still buying debt. Well, they wouldn't be buying debt if they thought that was possible, right? So, obviously, policy is sustainable, we just may not know exactly how it's going to get sustained. We talked about that already. 

Leeper: But, there are so many logical disconnects in that, that I just don't see how, if I were a policymaker, I wouldn't learn anything at all about what alternative policies might actually stabilize debt and what the consequences of those policies would be. That, to me, is what is useful policy analysis, and I don't see that. Now, let me just say one more thing about the CBO, because the CBO does great work at the microeconomic level. I think that a lot of their studies are very carefully done, and they do the accounting beautifully. There are lots of things that the CBO does very well, but I think that these long-term macroeconomic projections are not among those things.

Beckworth: Well, maybe one day you'll be nominated to lead the CBO. Go in, bring in the GE models, and roll with it. Well, with that, our time is up. Our guest today has been Eric Leeper. His paper is titled, *A Fiscal Accounting of COVID Inflation.* We'll have a link to it in the show notes as well as his other papers. Thank you, Eric, once again, for coming back on the show.

Leeper: Thanks, David. It's always fun to be with you.