Peter Stella is the former Head of the IMF Central Banking division and has researched and written extensively on safe assets, collateral and central bank operations. Peter now hosts a website Central Banking Archeology. Peter joins David on Macro Musings to discuss the role of money and its relationship to inflation as well as its relationship to the payment system. Specifically, David and Peter discuss the fiscal theory of the price level, how rising indebtedness can signal higher inflation in the future, the implications of the fiscal theory for contemporary fiscal and monetary policy going forward, and much 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: Peter, welcome back to the show.
Peter Stella: Thanks, David. It's great to be back with you again.
Beckworth: Yeah, it's great to have you on, and you've been requested multiple times by many listeners, so you're one of the favorites and it's always great to get a favorite back on the show. So I'm going to begin with a paper you've written and it really delves into this big theme of what is the relationship between money and inflation. And do we have a correct understanding the title of your paper is “Interpreting Modern Monetary Reality,” but maybe start us off with kind of the big picture here you're trying to paint for us.
Stella: So that paper is something that I wanted to do for a long time. And it really is my, you could say, I always did a lazy or patient, but in 1995, Robert Lucas gave his Nobel prize at us and it was all about monetary neutrality and so on. And he's a complete theoretical genius and deserving of the prize and all that. But one of the things that bothered me in that address and bothered me in some sense with the quantity theory of money is that some of its advocates looked to empirical data and basically made the claim, this theory works in every country, no matter what the exchange rate regime, no matter what the fiscal situation. It's basically the relationship between money growth and inflation is one to one over the long run. And Lucas in particular refers to some empirical work that he did not do.
Stella: It's not his own work, saying that this is completely the most valid economic theory in history. And so that the work that he cites, in particular, the one that I always wanted to redo, used IMF data actually, cross country data from 1960 to 1990. And it validates the claim that, yeah, there's this very tight relationship in all these countries. And so having been at the IMF and traveled to a lot of countries to have been the fiscal economist in Argentina in the 1980s, early 1990s, and seeing hyperinflation there, been there during hyperinflation. I was in Argentina in May, 1989 when the price level went up 198.5% in one month.
Stella: And I've also been in countries like Liberia, worked in Liberia where they didn't have their own currency, and yet they had quite a bit of inflation. So I was kind of accumulating observational facts that just didn't quite, weren't quite well explained by the theory. And there was kind of a joke in the IMF at that time, I was in the fiscal affairs department at the beginning. And that was that IMF, which stands for international monetary fund, really stood for it's mostly fiscal. So it was kind of a joke, but you know, I saw this in front of myself and I said there's some things going on here. It's not quite so easy to explain them because people in Argentina at the central bank, they weren't trying to create a hyperinflation. It's not like they were controlling the money supply to create hyperinflation, something underneath was driving this.
Stella: So for a long time, I kind of just wanted to do something very easy, which was to say, let me just rerun the same empirical work with the same countries over the last 30 years of data, or the last 20 years. And this is where either the patients or the laziness came in. So with a couple of former colleagues at the IMF, Maman Singh and someone else in his depart, kind of reproduced those results from the period 1990 to 2020. So it took us whatever it took me 20 years to get around to doing this, but we did it. And I was convinced that the numbers would not come out the way that they had come out in the earlier period and true enough, they didn't. So that was kind of that whole point of that IMF working paper, which came out in January. And then it was a bit like saying, "Okay, there's all these facts that we can't quite explain with this theory."
Stella: And that's the end of it. Now, someone got ahold of that paper and thought, "Oh, this is really good, but I have to read it twice." And so it's the editor of the journal of Fly Corporate Finance at Columbia. And he said, "Could you kind of rewrite this paper in a different context and make it straight forward?" So I did that and I thought I did a pretty good job of making it much, much clearer and not getting involved in all the econometrics. And so he said, "Yeah this is great, but I want the right theory. You've explained why that theory is failing. But now I want to know, what's the truth." And I'm thinking what you think I'm like Albert Einstein or something, I'm not the guy to ask me, what's the new theory, but I went ahead and I said, okay he's the one who asked me to write the paper, he's the boss, so to speak.
Stella: So I said, okay, I'll rewrite the paper and I'll work in what I think is the most plausible sort of explanation or the leading contender to advance economic science, which I would say is the fiscal theory of the price level. And I rewrote the paper pretty considerably with that in mind. It wasn't like I just added a paragraph at the end of the paper. And in writing the paper, I kind of convinced myself that we should look at the fiscal theory as a more general theory of the quantity theory. So the quantity theory is a special-
Beckworth: Is a special case of the-
Stella: Is a special case.
Stella: Exactly right. So if you look at sort of the history of science, let's say real science, like physics, it's kind of around the term of the 20th century, the 1900s. People are doing experiments with new technologies and coming up with a lot of data that Newtonian mechanics, Newtonian physics cannot explain. It's very disconcerting, it's not like Newtonian physics is wrong. It's just, we're now generating a lot of data that just cannot be explained by it. And then at about 1905, Einstein comes up with the special theory of relativity and that explains a lot of stuff, but not everything. 10 years by later, he comes up with the general theory. That explains more stuff. But obviously, I don't think Einstein never ever sort of claimed that he had explained everything. Obviously he hadn't explained everything.
Stella: And that's the way kind of science moves ahead or doesn't move ahead. Is that the theories start failing to explain the new data and new theories have to come up, which not kind of reject the old theory, but simply say, "Okay, we need to have a more general explanation of phenomenon."
Beckworth: Well, how would you define the fiscal theory of the price level for our listeners?
Fiscal Theory of the Price Level
Stella: So I would say that, coming back to the quantity theory, so the quantity theory is basically saying a particular type of liability of the state, which is a monetary, we call it a monetary liability, which is bank reserves or physical currency let's say. Changes in that lead one to one to inflation. So that's the variable to keep an eye on. So it's basically the state is issuing liabilities and this is the part where the fiscal theory stresses, without any change in fiscal policy. It's not promising to raise taxes in the future. It's simply issuing more liabilities against itself, and it's not generating any revenue to redeem those liabilities. So kind of law of supply and demand if a company issues more debt, but just burns the money away, it doesn't generate in a project that generates the revenue to pay off that debt.
Stella: You're going to see the value of the firm fall. And if inflation is measured in the share price of the company, then inflation will go up, but our unit of account is the liability of the state, which is these monetary liabilities. So basically in a nutshell, what I think the fiscal theory is saying is it's not just the monetary liabilities of state, it's all the liabilities of the state. So if the state issues a trillion dollars, if the US issues a trillion dollars in Treasury bills, and there is no sort of fiscal future, fiscal effort to redeem those, that debt, it's going to be just as inflationary as if the Fed were issue, just give away a trillion dollars in bank notes.
Stella: So there's nothing... I would say that again, I can't claim to be at all, the author of the ideas behind the fiscal theory, but my simplistic understanding of it is to say, it's a question of not making such a huge deal between the monetary liabilities of the state and the debt liabilities of the state, or the, let's say the future entitlements that the state is committed to pay for like Medicare and so on and so on. So it's a broader concept. Now, if we want to come back to say, "Well, how did the quantity theory of money do such a good job of explaining inflation if I'm saying it's only a small piece of the pie that we should be paying attention to?" Here’s what I kind of come to in terms of the special case.
Stella: If you say, as I think is legitimate to have claimed say in the 1950s or the 1960s, that when the government says, "I'm going to spend more money," and finances it with money creation, that's a pretty clear signal they're not intending to raise taxes in the future. So it's basically a signal when I finance with money that means, "Hey I'm never going to raise taxes or cut expenditure in the future. This is a permanent increase in fiscal spending. And I'm not going to do anything about it." Whereas in the 1950s or 1960s, the government says, "I'm going to spend more money, but I'm going to issue bonds." Even though a bond is nothing more than a promise to pay money in the future. If you take that as a signal, Hey, this government is issuing a bond, they're going to raise taxes to pay the bond off.
Stella: So basically what I'm saying is when you see, let's say in the 1950s, when you see a government financing with money instead bonds, it's a signal that the fiscal effort that we expect in the future is very different than if they're financing with bonds. And of course, we can look back at cases of hyperinflation in those periods. And they're almost always cases where the bond market is not well developed. So there wasn't really a credible option for Argentina to issue domestic bonds in the 1980s. There was no market. So the way they had to finance, let's say a crisis, if they couldn't borrow externally, was to print money. It wasn't because they thought, "Oh, that's a great way to finance a deficit."
Beckworth: Yeah. So couple of elements you've outlined here. So one is fiscal policy and monetary policy are intricately linked in understanding this relationship up. So there is really no island of monetary policy that's truly independent of what fiscal policies is doing. That's the one thing you've mentioned. The other thing, and you highlight this in your paper, is the fiscal theory of the price level treats all these government liabilities as equivalent or substitutes to some degree. So the monetary base and bonds are substitutes in this theory, they're all government liabilities, you got to consider them. Now maybe they're not perfect substitutes on every margin, but they're, in terms of liabilities, you got to count both of them. And then the final thing you mentioned then is it's the expected future fiscal condition of the government that's going to be driving this process.
Beckworth: So if the government is not going to be running primary surpluses in the future, and if there's not a sufficient flow of seigniorage coming in, such that it leads to say permanently higher debt levels, is that the point where we see inflation begin to go up? Is it the concern about the future debt loads, future fiscal health that affects current inflation?
Stella: Yeah, I would say exactly. So that's exactly the notion. And if I could come back, I don't know your audience, how much they are into theory, but Neil Wallace, who's not as well known, but he's the Wallace in Sargent and Wallace, everyone knows Sargent and Wallace and Tom Sargent won the Nobel prize. And people probably don't know Neil Wallace, but he wrote a paper in 1981. It was called kind of, a Modigliani Miller theory of open market operations, which is a long title that if you don't know finance, probably means nothing to you. But basically what he was saying in a very theoretical model, he's saying, "Central bank issues money and blue government debt. I don't see any reason why that should change anything. It's simply one liability swap," and it's similar...
Stella: He didn't talk about it this way, but treasuries all the time, all over the world are making decisions, "Do I issue a Treasury bill? Do I issue a 30 year bond? A 10 year bond? A floating rate note?" And I've never heard anyone say that the choice between initially a Treasury bill versus a 30 year bond is somehow inflationary or terribly meaningful. On the margin, small changes in the government's debt strategy are just not talked about as meaningful economic factors in theory, or I would say in practice. Small changes. So what Wallace in my mind, and again, this is something that when I read the paper in the 1980s, I understood. But now in retrospect, you go back to that paper and say, "Well he's really asking a really hard question," saying, "Well, this is just a little liability management. Why should it matter?"
Stella: And he's saying it doesn't matter. Unless this is changing somehow fiscal policy. So there's a tiny little difference. Back in those days, the Fed is issuing non-interest bearing money and buying back treasuries, which are paying a little interest. So you could say, "Oh, this decreases the government's interest cost by a tiny little bit. So maybe that's what matters." But if there isn't that little fiscal thing going on, then it really shouldn't matter. Which of course causes all kinds of problems for people doing actual central banking because they think they are actually doing something right with the whole-
Beckworth: Yeah. So Wallace neutrality is a very powerful idea. And I remember having lots of conversations about it and it's the application of a Modigliani Miller theorem to public finance. It doesn't matter how you finance. It's the point. And I remember around QE3, maybe a little after that, we had these online conversations about, well, how affect have is QE3? Is it really making a difference> even Michael Woodford and Gauti Eggertsson's in their papers, they come up with something similar called the policy and effectiveness proposition of QE. And they make the same point that QE really doesn't make that much difference outside of market collapsing, market crashes. So maybe 2008, 2009, March, 2020, that's when large scale asset purchases can make a difference, they actually backed up a market. But in terms of normal market functioning periods, there really isn't a whole lot of bank for buck, other than maybe the signaling that you've mentioned, or it's some indication of the future health of fiscal policy. I want to come back to that in a minute, but let me just push this a little bit more with you.
Beckworth: So why would it be the case that even if we think government's going to run these bigger permanent deficits, so higher levels of debt in the future, why should it cause inflation? What is the actual mechanism or step that would imply more debt, higher inflation in the future?
More Debt Implies Higher Future Inflation?
Stella: Right. So the one who has, I think, the most extensive in articulating the fiscal theory is John Cochran. And he's working on a book, it's currently over 700 pages long. So I'll refer you to him for any details. But what troubles me a little bit about this, and it bothers me as well, personally very often, it's the notion that the government will have to monetize the debt eventually. Which is really the story behind Sargent and Wallace. So if you go back and read that paper, which I hadn't read for decades, I reread it and I said, "This is really kind of an artificial model." It's sort of like the government can issue so much debt and then it reaches this wall and then it has to issue all the money. And of course the world isn't like that. You don't just hit a finite well known barrier and then you flip from all bond finance to all money finance. But I think that's one way to think about it. I don't think that's the right way to think about it. But that's one way to think about it.
What troubles me a little bit about this, and it bothers me as well, personally very often, it's the notion that the government will have to monetize the debt eventually. Which is really the story behind Sargent and Wallace. So if you go back and read that paper, which I hadn't read for decades, I reread it and I said, "This is really kind of an artificial model." It's sort of like the government can issue so much debt and then it reaches this wall and then it has to issue all the money. And of course the world isn't like that.
Beckworth: Well, that's how I think about it.
Stella: The government will issue all this debt and it can't credibly raise the taxes to pay it off, it can't cut spending. So it's just going to print money and we get inflation. But I think that is missing something really important that has to do with the progress of financial markets in the last 50 years. And that is, it is more efficient, it is less costly to finance government deficits and debt. Sorry, government deficits with debt, than it is with money.
Stella: Okay. So this comes back to, you didn't raise this, but the whole idea of helicopter money. Okay. So helicopter money, this sort of drives me nuts, but we have to agree. Helicopter money is fiscal policy, and it's a permanent increase in the fiscal deficit. It is not a monetary policy. It is not something that central banks should do, should ever do. This is a political, Treasury, government decision. It is not a central bank policy. Okay. This is a permanent increase in the fiscal deficit. And why do we call it helicopter money? Because traditionally it was financed with money. But I've written several sort of blog posts, which I don't think are on my site anymore, but you can do helicopter money with bonds. You can give away bonds, and I can have a permanent increase in the fiscal deficit by running arrears or, and this is not a hypothetical example, during COVID some countries just had a state owned electricity company for example.
Stella: I think Bahrain, for example, did this. And they just said, during COVID, don't pay your electric bill for six months or three months. And we're never going to recover that money. I mean, we're never going to ask you to pay that electric bill. Permanent forever. And that didn't involve creating a creating money. But to me, that's a permanent increase in fiscal policy financed by a decline in the net equity of the state electricity company. And we never to, I say, we, the government never has to create money, it's just there are different ways to do this. It's not just money. And I think money is an inefficient way to do it because of what we're seeing right now in many, many countries, countries can issue debt, long-term debt at negative, real rights.
Stella: And in many cases, negative nominal rates. So why is that? It's because as my friend and colleague Maman Singh has talked for a long, long time securities have collateral value. They're really important in a modern financial system. Debt is much more important than money in modern financial systems. I mean the monetary base, so to speak. So if you're looking to finance a permanent increase in the fiscal deficit, to me it's much more credible to issue debt and roll over that debt forever.
Helicopter money is fiscal policy, and it's a permanent increase in the fiscal deficit. It is not a monetary policy. It is not something that central banks should do, should ever do. This is a political, Treasury, government decision. It is not a central bank policy.
Stella: At higher interest rates, but there will be higher inflation, you'll have to pay for it, but it's actually less costly than issuing money because who wants $4 trillion in excess reserves at the central?
Beckworth: Well, I guess that's the condition then is you can keep rolling over debt as long as it's cheaper, long as it's more efficient. But in the limit, I'm guessing the argument is that it may not always be that way. That the government may have to force reserves into the system, may have to force currency out into the public. And that's the certain. Let me push back a little bit on, on the fiscal theory of the price level for all my quantity theory fans out there. So I'm trying to imagine what they would say, and I want you to give answers to this. So let me first... So a couple of things. One, some might say, well, I think the quantity theory money is still valid if we expand the definition of money. And that's kind of what you were saying in your paper, and you just said as well, a few minutes ago, but in other words, think of money broadly as including Treasury securities.
Beckworth: And we include Treasury securities because they are highly liquid. Like your colleague, you just mentioned, said they're used as collateral. They have a convenience yield on them. And so for example, William Barnett does this divisia money aggregate measures that include treasuries. He has an M4 measure. And so if you look at M4, for example, during the great financial crisis, it actually collapsed. Once you account for institutional money asset, it collapses during this time. And so you could kind of point to something like that and say, "Hey from a broader perspective of money, quantity theory, still kind of loosely holds." How do you respond to that?
Stella: Sorry, I'm going to use the word that I don't like to use in a general population, but here we have difference between exogenous and endogenous.
Stella: Okay. So when we talk about causality, the quantity theory of money... We can get into the weeds on this, which I don't want to do, but basically the quantity theory of money and all the theoretical models, money comes out of nowhere where it's created by the state. It is exogenous and people take it or leave it, basically that's it. In the real world the money supply that you're talking about, that in other words, all the monetary measures apart from that which is created by the state is endogenous, is determined by demand. So if we see a correlation between inflation and money growth, the correlation in my mind, and I know people hate this, is running the opposite way.
Stella: So money is going up, yes, but that's because banks are lending more, or incomes are going up. The demand for money is going up. The private sector is creating, if I put a number on it before the big blow up in central bank balance sheets, I would say 99.9% of all money was created by the private sector in response to demand or response to credit supply, what have you. So yeah, there's going to be a correlation, but it's running the causation. If there is a causation, it's running the other way. I mean, that's where I'm drawing the distinction between government debt again, is exogenous. So the US Treasury, doesn't say, "Hey come and buy as much debt as you want." No it's setting the amount of debt it's selling, and then the market's determining the price. So basically the sum of money, state liability, monetary liabilities, and other liabilities is determined exogenously. So if we're looking for a driving variable, a causation variable to me, it has to be that one, not the other way around.
Beckworth: Okay. Let me do a second objection from a quantity theory perspective. And it goes something like this, the proper view of the quantity theory would maybe be a little more nuanced than what you outlined earlier. And that is, it would be limited to government liabilities, actual government money. So the monetary base. It wouldn't include necessarily M1, M2. I know Robert Lucas looked at broader measures that included a lot of private money, but it would be limited to the monetary base. And it would be limited as well to injections of the monetary base that are with two conditions. One, it has to be greater than what's demanded by the public. And two, it has to be permanent or expected to be permanent. It's never going to be reversed. And I think probably one answer you're going to give to that objection is, well, what is a permanent increase in the monetary base? It's a helicopter drop.
The quantity theory of money and all the theoretical models, money comes out of nowhere where it's created by the state. It is exogenous and people take it or leave it, basically that's it. In the real world the money supply that you're talking about, that in other words, all the monetary measures apart from that which is created by the state is endogenous, is determined by demand. So if we see a correlation between inflation and money growth, the correlation in my mind, and I know people hate this, is running the opposite way.
Beckworth: Which gets us back to fiscal policy. But I think that's one of the views out there is that it's going to be permanent. It's not going to be reversed by future taxes, and it's going to be greater than what's actually wanted. And therefore it will be spent or put to use on spending. Any thoughts?
Stella: So I completely agree with that. But that's again, what I'm saying, that is the mechanism, but it applies equally to... Let's imagine a little piece of paper and we call that piece of paper, a bond. It applies to the bond. And it's kind of funny, you go back in US history the constitution, it was interpreted saying because of the experience of inflation during the revolutionary war plans, that Congress couldn't issue money, it could just issue bonds. Paper money, couldn't issue paper money. So money was gold or silver. That's what the traditional interpretation of the constitution was. Coin money. That's the base of constitution. So that's very clear. Put a stamp. This is a US dollar. It's a certain weight of gold, certain weight of silver. Now there were times of course, where the government needed money, well needed financing.
Stella: And they issued small denomination, bearer bonds that paid like one mil, which is like one 1000th of a percent, in six months or something. Basically the interest rounded to zero. Okay. So this was paper money, but it was authorized by Congress. It was a big controversy in the Congress about whether this was printing money, which it really was. Okay. It was. But technically legally, "Oh no, no, no. We're just issuing the bond." Okay. So what I like to say is, look, what is the difference between one piece of paper and the other piece of paper that makes it so different? I don't think it's the interest rate. I don't think it's what the bond promises, because the bond promises money. That's all. Promises the paper money. So what I'm saying is, it's all about fiscal policy. And like you say, if you are issuing more bonds or more money and people want in quotation marks, yeah.
Stella: The bond prices are going to go down, price level will go up. I mean, all this. The value of the government bonds will go down. How does that happen? Well, price level goes up and the real value of the debt goes down or the real value of the money goes down, which is what we call inflation. So I completely agree that that's the mechanism. And then I come back to the signaling issue and I want to bring another example from US history. So during the depression, a lot of people had really innovative ideas about, they didn't talk about helicopter money, but printing up greenbacks and giving them out. One was for civil war veterans to give them a bonus and cash, so on. And it's very interesting. If you look at some of the draft proposals and Irving Fisher was involved in some of these. For some of these, what we would call pure paper money, irredeemable paper money was a phrase that Irving Fisher used.
Stella: And I think it's very important that he used to irredeemable. So that's coming back to, okay, that's a permanent thing. This is not paper money that's redeemable and gold or silver. This is irredeemable. There's no promise to ever, ever give you something real for it. So that was the idea, we want to create inflation during the great depression. I think some people did. It's very interesting. Some of the proposals actually called for these money, these greenback issues to be redeemed over time to be redeemed over like 20 years, because people were afraid it would kind of get out of control. It would be too easy. And Roosevelt actually, as you probably know, didn't want, Congress gave him the power to print up greenbacks and give them away. And if you read some of his speeches, he was totally against that. Because it was talking about fiscal policy and he didn't think that's the way you did things.
What I like to say is, look, what is the difference between one piece of paper and the other piece of paper that makes it so different? I don't think it's the interest rate. I don't think it's what the bond promises, because the bond promises money. That's all. Promises the paper money. So what I'm saying is, it's all about fiscal policy.
Stella: In a crisis yes, you can do this, but he didn't want this idea. He talked about the veteran's funds. He vetoed the bill to pay the veterans bonuses in advance. And he said this is really bad, because everybody's going to start asking for this and this doesn't make sense. So I completely agree with the, let's say the validity of the quantity theory of money in the circumstances you described, if it is a verifiable or true signal of the future of fiscal intentions of the state. I absolutely agree with that. But all I'm saying is in, in modern times I would say the idea that you'll issue more government debt and then one day redeem it, I think people would laugh at you today. It's like, "Are you kidding me?" The debt that countries are issuing now, you think we're going to like reduce the nominal stock of bonds-
Beckworth: Let me push back a little bit on that point. I think the way I wrap my mind around this, and this is based off some work by Marcus Brunnermeier, he talks about fiscal theory of the price level with a bubble. So part of what he says supports your point. If you look at discounted present value of future primary surpluses, expected primary surpluses and seigniorage, it doesn't really add up to the real price of bonds today. There's not enough government fiscal resources in the future being collected that to justify, to warrant of value of government liabilities today in real terms. And so what's left? And that's this bubble term, which he and others are arguing comes from this convenience seal, this liquidity that government bonds provide as collateral, as well as the monetary base.
Beckworth: In other words, look at seigniorage broadly beyond just what the monetary base provides, but also what the bonds are providing. So if you look at all of those, in other words, there's a real service being provided by these assets and the government's tapping into the revenue that service provides, that there's a broad seigniorage flow, then there's some, hopefully some primary surpluses. And so from that perspective, there is a future flow of real claims against the economy the US government is taking in that keeps the price level low today.
Stella: Yeah. So I completely, I think I would be sympathetic to that view. But let me come back to something personal where I saw a bit of broader insight into this. And that is, as you might know, I got involved in the issue with central bank capital a long time ago, 30 years ago. And it was a bit controversial people weren't thinking, "Oh, central bank's not capital. Does it matter what the net worth of the central bank is ?" So I did I did a fair amount of work on that. I mean, I wasn't at the IMF, my job wasn't really to do research, but I did it because I'm curious and came across all these things that just didn't make sense to me. And I had to try to figure them out.
Stella: So basically I did a lot of research on central bank net worth and capital to this matter. And, and eventually I wrote a paper with a friend of mine, ex colleague Ulrich Clu at the IMF, and it was kind of an econometric look at this. And you know what, we found is that the, and I'm coming back to this consolidated government debt pricing equation. So where do we get the market value of bonds? And you say, "Gee, it doesn't seem to square with the present discount of value of future primary surpluses," so I'm coming back. I'm going to get to that point. But what we found was central bank negative equity, it's not a linear type problem. In other words, if your central bank net worth goes up by 1% of GDP, and it's 2% of GDP, it doesn't affect anything.
Stella: And if it's 1% of GDP and it goes to zero, that doesn't really affect anything. It's a very non-line relationship. So when it gets very negative, then that's a real problem. So it's a very non-linear thing. And it is very similar, I think to, if you look at Merton Miller, like distance to default with the corporation. So you have equity and you have debt, and basically you can have a lot of movements from AAA to...The probability of default does not rise linearally with distance to default. So you get downgraded from AAA to AAA plus it doesn't change probability by anything. But when you get close, it's very non-linear. So a small shock when you're on the verge, makes a big difference.
Stella: Okay. So that's the same thing with central bank equity in the sense that when your equity is deeply negative and it is in danger of getting more negative, then basically you're going to have a lot of inflation. You've got to finance your, you either have to default as a central bank, or you create a lot of inflation. So small changes are very important in that. So I would come back now, look at the sovereign. Look at the US sovereign. And in my sense, and I did a lot of work on measuring the fiscal deficit within IMF as well. So it's a very complicated thing, and it's much more complicated than any model. You've got two or three variables. In the real world, you've got present discount of value of social security obligations. Medicare is very, very complicated. I like to say in a model, there's one way to default on government debt. In the real world, there are 10,000 ways to default. The government can default in 10,000 ways. Break its promises, right?
I completely agree with the, let's say the validity of the quantity theory of money in the circumstances you described, if it is a verifiable or true signal of the future of fiscal intentions of the state. I absolutely agree with that. But all I'm saying is in, in modern times I would say the idea that you'll issue more government debt and then one day redeem it, I think people would laugh at you today.
Stella: Raise taxes, cuts spending, whatever. So it's very complicated. And I think in countries like the US, or let's call them Switzerland or call them the advance economies. They're not always so advanced. They have this equity or margin or left balance sheet resources, whether it's political determination or sensibility, or just if the US started charging entrance fees to the national parks, like Disneyland charges.
Stella: There are a lot of resources that a country like the US has that aren't on the balance sheet but they could be called upon. And when you look at countries like Argentina or Venezuela, I don't want to pick on them, but there are a lot of countries in that category that I've seen and I've been there. And they just don't have that.
Stella: So they're on that margin. So if they issue, if they have a... In COVID, let's say, they issue something people say no way in the world, these people are going to raise future primary surpluses. There's the track record of that. So if you look at, coming back to John Cochran's book, in terms of the general theory of the fiscal theory he says pretty clearly, or well, it's a presentation. That he says, he's in the US talking to a European and says, "In countries like ours, it seems like when we issue a lot of debt, which we have with COVID, people expect us to kind of generate higher primary surpluses." And he comes up with a general estimate, like two thirds. Two thirds of it will be covered with future primary surpluses. And a third will be covered by basically inflation.
Stella: And when I'm watching that, I'm thinking, "Oh yeah, in Argentina that happens." And basically people expect 0% to be covered. In other words, it's not even a negative correlation. It's like, if an as well starts doing something, you would expect even more, like get worse, for it to get worse, not for them to recover. So coming back to the fiscal theory. Yeah, there's a lot that is not necessarily obvious on the balance sheet. And we don't really... I mean, Larry Summers has talked about this and you had Jason Furman on and they wrote this paper about how to measure the debt. It's very interesting, these kinds of things, where in the US do we get a calculation?
Stella: What is the infinite horizon fiscal policy? Well of course we don't know that. It's uncertain. There would be 30 years of work to figure out, "Oh, how do we know that the US is issuing more that than it can?" It's not so simple. In other words, it's not a linear relationship. We're not on the margin. We can't tell. Now, it depends on all these things we don't know. But I guess what I conclude in the paper is to say, the only thing I'm pretty sure of is we should be looking at the total of the state's monetary plus securities, liabilities, not just at the expansion of the monetary liabilities to interpret modern monetary reality. It's not just money anymore. Maybe in the 1930s, it was just money. It was either money or taxes. And now we have to look at the sum of the two. But even then, it's not a linear relationship. I don't think we're ever going to get one equation that says, "Oh yes, all the time. There's going to be a one to one match."
Beckworth: So Peter, a fascinating conversation. I want to take what we've discussed and apply it to the current developments, what we've seen over the past year with Fed policy, with Treasury and what do you see as the implications of all this discussion we've had about the fiscal theory of the price level and such for the implementation of monetary policy and fiscal policy?
What is the infinite horizon fiscal policy? Well of course we don't know that. It's uncertain...the only thing I'm pretty sure of is we should be looking at the total of the state's monetary plus securities, liabilities, not just at the expansion of the monetary liabilities to interpret modern monetary reality. It's not just money anymore.
Implications for Fiscal and Monetary Policy
Stella: So one of the clear points coming out of the fiscal theory mileu if you will, Chris Sims has a nice address to when he was president of the American Economic Association, basically saying we need to rethink macro economics and look at the relation shift between monetary policy and fiscal policy and the relationship between central banks and treasuries. So to give one example of this, and it's a really big picture thing. And I think it's a very, very important thing from the political economy standpoint. So right now today... Well, let me take a step back. In 2007 in the old days, this is what the Fed was doing. This is how the Fed did monetary operations. It was lending through repo operations, 20 billion and banks were holding. All US banks were holding 16 billion at the Fed. So the Fed was a net lender to the US financial system of four billion, four billion.
Stella: That's nothing. When the Fed raises interest rates, raised interest rates in that scenario, it was successfully doing monetary operations policy, you'd say, "Oh, there was a fiscal impact." So the Fed was lending 20 billion. Let's say it was paying interest on reserves. Well, it wasn't paying interest on reserves. So it increases the rate effectively on its $20 billion loan by 25 basis points. Oh big deal. Who cares? Nobody cares except maybe Neil Wallace, right?
Beckworth: Right, right.
Stella: It's nothing. It has a fiscal impact. Yeah. But it's no fiscal impact. Zero. Now, today, right now, October 7th, the Fed is lending $60 billion to the banking system, which is a lot. And then these days, it's nothing. What is it borrowing? So we have to think of deposits at the Fed as lending to the Fed. Just like when you make a deposit at the bank, you're lending to the bank and the bank is lending your money out. So that's borrowing. So what is the Fed borrowing from the financial system? Well, it's borrowing for 4.2 trillion from US banks in terms of reserves and it's borrowing 1.7 trillion in overnight reverse repos from non-bank. Okay. So that's $6 trillion in borrowing, net borrowing. So it went from a net lender of $4 billion in 2007. It is now a net borrow of $6 trillion.
Stella: It is paying interest on that. That's 25% of GDP. Okay. 25% of GDP. So now let's say the Fed is managing 25% of GDP of the US debt. And now the Fed has to raise interest rates. So Fed raises interest rates by 25 basis points, 50 basis points, 1%, a hundred basis points. That has a real fiscal impact. The Fed is now paying 1% on $6 trillion and it's not doing it in a market determined way. It's doing it, let's say, arbitrarily. It's setting the rate. So this is what we call a floor system versus what the old system was. And I would say, forget about what the operational issues, if the floor system is 20, you're lending, or you're borrowing $20 billion, I don't care. But this is a system where you're borrowing $6 trillion. And imagine the political perspective on that.
Stella: If the US Treasury were to say, "Hey we've got $6 trillion of notes and bonds out there and all you people who are holding them, we're going to pay you another 50 basis points coupon on it. We're just going to give it away." You would say, "You're out of your mind." No Treasury would ever do that just out of nowhere, say, "Oh yeah, here's a gift." But that's essentially the political problem that the Fed could be getting into by having this huge balance sheet just from the optical standpoint. And it's not just me, who's raised this issue. It's come up in the UK where I think the bank of England is much closer to raising rates and people have, respected economists have said, "This might be a problem. Maybe you're not going to raise rates because of this sort of political problem or the fiscal impact of it."
Stella: So I think that's something that people are, are sort of missing because most people were, are still in that world where central banks were very small players in the financial market in terms of magnitudes and now you cannot ignore them.
Beckworth: Yeah. That's fascinating. So $6 trillion the Fed is financing, whereas normally it would be with Treasury. So you could argue financing costs one way or the other would occur, but the political economy, how it looks, the optics are just completely different. And it's arbitrary. As you said, the Fed sets an administrative rate, the Treasury takes it to auction. So there's big differences there. And what I see this leading to, I mean, maybe this is your point too, but is the Fed losing independence. If the Fed gets cut up in a political firestorm, because it's paying bankers and Wall Street, higher interest and Congress steps in and says, "No, you can't." Then the Fed can't respond to say inflationary pressures. And it loses its ability to fulfill its mandate of price stability. So I could see how this political consideration ultimately affects actual monetary policy.
Stella: Right. And as you acutely pointed out, from the US context, we went for decades without paying interest on reserves. So it's not so ingrained and obvious, why are we doing that? And I think when we started paying interest on reserve, there's a lot of confusion about that. And I think there remains a lot of confusion about that. And I might have said this before, but I was at Jackson Hall in 2016 and a board member from the ECB said, "I don't understand why negative rates are so on unpopular." And I was in the audience and I kind of raised my hand and I said, "Well negative rates would be a lot more popular if you were lending at them instead of borrowing at them." In other words, in 2007, yeah lower rates are always, everyone is happy. But you have to realize you're not a net borrower.
Stella: You're not a net lender anymore. You are a huge borrower, so of course people are unhappy at negative rates. People are lending $6 trillion. And if you put the rate negative, it's like taxes. Again, it's kind of a fiscal, it's almost like a tax. For central bankers not to have seen that this, in mathematical terms, the derivative, it's now on a negative number, it's not on a positive number anymore. So the derivative is negative. It's not positive anymore. So I'm really worried that people are underestimating this problem. And when we talk about, I guess the question going forward is, is this the future going forward? And we have to recognize that okay so that means the Treasury's got to take into account the Feds policy for its debt management strategy.
Stella: And we've delegated a huge amount of the sovereign debt to be set by an administrative rate for a different purpose. Except for a monetary policy purpose. And in let's call it the old days, the fiscal implications of that were nothing. Although of course, for decades, every time central banks raised rates, even though direct fiscal impact was not big, finance ministers jumped up and down because they knew they would have to sell in the market, their debt, and roll it over to high rates. So they would always complain. And when you come back to central bank independence, and in my mind, if I define central bank independence in one sentence, it's the ability to raise interest rates when the Treasury doesn't want you to. And the Treasury almost never wants you to, because of the cost of the debt.
Stella: And now we're putting it in the lap of the central bank and saying, "Okay, central bank, raise interest rates." And I think most central bankers would say, "Oh yeah, we're going to make a loss and da, da, da. Don't worry about it," and say, "Okay, yeah, I want to worry about it. If you're talking about like $10 million and your seigniorage is $100 billion." But now you're talking about raising rates on $6 trillion and people are going to start asking questions, what is going on? So to have it on this debts balance sheet, I think is extremely...
Stella: First of all, I don't think it makes sense because I think the Treasury's job is to manage the debt and to fit everything in the budget. It's not the job of the central bank to be doing that. But second, just from the political administrative and even economic aspect it's, let's face it, a floor system. It's an administered rate. It is not auctioning a Treasury bill, where there's an interaction of demand and supply. So I think that is very different than the way we used to think of things before the GFC. The US was trying to conduct monetary policy, which is a little impact on the yield curve as possible.
If the US Treasury were to say, "Hey we've got $6 trillion of notes and bonds out there and all you people who are holding them, we're going to pay you another 50 basis points coupon on it. We're just going to give it away." You would say, "You're out of your mind." No Treasury would ever do that just out of nowhere, say, "Oh yeah, here's a gift." But that's essentially the political problem that the Fed could be getting into by having this huge balance sheet just from the optical standpoint.
Beckworth: Let me ask you this, Peter, in the time we have left, you have a proposal. We talked about it, I believe on your first show, first appearance here on Macro Musings. But you have a proposal that would address this problem. And what it would do is it would transfer all those liabilities from the Feds balance sheet to Treasury's balance sheet. So tell us how that would actually operate. What would be the steps necessary to shrink the Fed’s balance sheet and have Treasury take over the liabilities?
How the Treasury Can Help Shrink the Fed’s Balance Sheet
Stella: Okay. Right. So I started proposing this to finance ministries in 2005. I won't mention the country, but basically this is not a new idea for me. And so basically the proposal is you say to the Treasury, "You've got to issue a ton more debt and keep it at the central bank” and what the Treasury, when I'm there in the office and this gentleman, the guy literally got so agitated that he said, "Excuse me I have to leave the office." He reached into his drawer and he took something out of the drawer and went out of the room. It was like he was going to have a heart attack. So the answer is very simple. But people immediately think I'm crazy. So the answer is US Treasury go out and issue $2 trillion more in debt and deposit that at the Fed.
Stella: And the Treasury is going to flip out. But what the treasuries don't kind of realize is when you've got $6 trillion in excess reserves at the Fed, well, the money is there. It's not like the old days where you have to go out and borrow it. All you're doing is, is doing a debt management operation. You're changing the issuance from issuing debt that you are very good at and retiring this overnight debt. Okay. So in very practical terms, if the US were to do this, I would say something like, look, you've just announced over the next 12 months, 24 months, the Fed is going to restrict the amount of excess reserves that it will pay interest on reserves. To say, "Okay, now we're paying it on all of it. We're going to reduce that by $100 billion dollars a month or a quarter." Well then the Treasury just comes in and scoops that money up. So eventually you get to a point where, and I think politically, it would look good because, oh yeah, we're going back to not paying interest on reserves, isn't that great? Well, that's kind of an illusion because the Treasury is paying interest on the debt.
Beckworth: Right. Banks still get the money.
Stella: But still you're doing it in a much more are sophisticated way. The Treasury is issuing floating rate now. It's for 30 years. It's issuing all these instruments. Instead, the Fed just issuing kind of one instrument, which is interest on reserve or overnight reverse repo. And other people, other countries have done it by the central bank issuing debt. Right. So you absorb all the excess reserve, but debt. I don't like that because it fragments the debt market. The US thought of doing that right at the beginning of the GFC, I think rightfully so there were discussions at the New York Fed but they decided not to do that. And I agree with that. But if you recall at that time, the Treasury went out and issued debt and did exactly what I'm talking about, and it was called the supplementary finance program. Went out and issued about $650 billion of short term cash management bills. Took that money at the Fed, placed that money at the Fed.
If you recall at that time, the Treasury went out and issued debt and did exactly what I'm talking about, and it was called the supplementary finance program. Went out and issued about $650 billion of short term cash management bills. Took that money at the Fed, placed that money at the Fed...It was not to finance the deficit. This was simply a debt management operation...that's the proposal. It's basically shrinking the, either you want to call it the floor, the amount of money that's at the floor or creating the reserve shortages like we had before.
Stella: It was not to finance the deficit. This was simply a debt management operation. And it happened. It's not a theoretical thing. They actually did this for a different reason. It was to keep the Fed funds rate above zero before it had the authority to pay interest on reserves. And before the target went down to zero. But that's the proposal. It's basically shrinking the, either you want to call it the floor, the amount of money that's at the floor or creating the reserve shortages like we had before.
Stella: I'm willing to be completely agnostic about if you want a floor system, fine, but have a floor system where it's $100 billion dollars or $20 billion, not $6 trillion. That's all I'm saying. You don't have to have $6 trillion there. And I really don't think it's efficient. The Treasury can finance at a lower cost. Right now, the Treasury can issue bills at a lower cost than the interest on reserve rate. So it would save the taxpayer money for the Treasury issued bills and deposit, take that reserves out of the system. It would be cheaper. That's just a micro example. So that's trying to bring in this thing, like fiscal theory of the price level saying, "Yeah, QE it doesn't really matter whether it's reserves or it's..."
Stella: From that perspective, yes. But from this other perspective of blurring the boundaries between fiscal responsibilities and central bank responsibilities, between what we understand to be monetary policy and what we understand to be fiscal policy, I think that's a real dilemma or a problem going forward. I don't think perpetuating this system makes sense for me. And as I've written elsewhere, other countries have solved this problem. I mean, they have put the genie back in the box, so to speak. Chile, Mexico, India, Israel.
Beckworth: Well, that is a great proposal. And we'll provide a link to your article where you outline that. But with that, our time is up. Our guest today has been Peter Stella. Peter, thank you so much for coming back on the show.
Photo by Alex Wong via Getty Images