May 11, 2020

Nathan Tankus on Public Finance in the COVID-19 Crisis: A Consolidated Budget Balance View and its Implications for Policy

The creation of a consolidated finance ministry may actually improve Federal Reserve independence.
David Beckworth Senior Research Fellow , Nathan Tankus

Hosted by David Beckworth of the Mercatus Center, Macro Musings is a new podcast which pulls back the curtain on the important macroeconomic issues of the past, present, and future.

Nathan Tankus is the director of research at the Modern Money Network, and a research fellow at the Global Institute for Sustainable Prosperity. Nathan is also the author of a number of articles on the Fed's recent activity at his Substack page titled “Notes on the Crises.” Nathan joins Macro Musings to talk about the post-Keynesian view of money, central bank independence, and the consolidated view of public finance, as well as evaluate the policy responses by the Fed and Congress to COVID-19.

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

David Beckworth: Nathan, welcome to the show.

Nathan Tankus:  Thank you very much for having me. Great to be here.

Beckworth: Great to have you on. So, you're one of these Twitter friends I've gotten to know over Twitter, which has been great, over the past few years. My Twitter world has grown, and I meet interesting people like you and others that probably would not have met otherwise if... for Twitter. And what's interesting, though, is we come from different perspectives, and yet we have a lot of interesting discussions and share a lot of fun debates about what monetary policy is, what fiscal policy is. So why don't you tell our listeners a little bit about yourself and what you're doing now? And we'll go from there.

Tankus: Yeah. I'm a research director for the Modern Money Network. Before everything happened with coronavirus, I was working on a lot of work on monetary policy, specifically targeted to what monetary policy would look like in a Green New Deal world and the different ways that monetary policy would have to adjust to accommodate a large-scale Green New Deal program. I'm still working on that, but for the last month or so... a little more over a month... I've been focusing on covering everything going on with the coronavirus depression. There was a point around mid-March where it became clear that this was the big one, as it were. And it became very clear that the Fed was going to, as ever, do a lot of heavy lifting in terms of this crisis and do some unprecedented things. There wasn't many people I was seeing who were preparing to do in-depth coverage of that. And it felt like the right time to start my own coverage. Without too much bluster, I think I was right in that assessment and that a lot of people have found frequent coverage of the crisis, especially on the monetary policy side, very helpful.

Beckworth: Oh, absolutely. I've enjoyed your articles. I know others have. And I know you've received a lot of attention because of them. How were you so well set-up to do them, coming into this crisis? I mean, you must've done a lot of reading. What shaped your formative experience in macroeconomics that made you prepared to do these articles?

Tankus: No question it's the Great Financial Crisis. I was telling this story a little bit on Odd Lots, the Bloomberg podcast, but I had a couple oddball history teachers, social studies teachers, in a kind of weird, alternative high school in Manhattan. And in January 2009, two of these teachers looked around and said, "Why don't we throw a class together that's studying the financial crisis, and what's happening, and what's happening with the stimulus, and such?" And, basically, they just took selections of the newspaper and had us read and argue about it each week.

Tankus: So we went through arguing about the AIG bonuses, and then arguing about stimulus, and then had broader introductions to monetarism, Keynesianism, that sort of basic high school level that you would get. But it was completely addictive and hooking, and there was something fascinating about the idea that no one seemed to be really sure, confident, in their understanding of what seemed the most important thing going around at the time, the global financial crisis. That was inherently fascinating, and I got completely hooked, read Friedman's Monetary History of the United States over that summer.

Beckworth: Wow.

Tankus: Started reading other things. Picked up a copy of Keynes' General Theory and read it on and off for the next couple years. Discovered Minsky because he was one of the few alternative monetary thinkers that was available in the New York Public Library, because his book Stabilizing an Unstable Economy had just been reprinted. And that just put me on a complete trajectory of getting interested in macroeconomics and monetary economics. When your introduction to economics is crisis, you inherently get very focused, even if outside of that context, or maybe a number of years later, I wouldn't necessarily think crisis is everything. What happens in normal times is important, too. I've made a lot of arguments around monetary policy as it was operating in 2017, 2018, 2019, which, while there were issues, you wouldn't characterize as a crisis. But that introduction always makes you more focused on crises than everything else. When you grow up on crisis, and the crisis being where all your intellectual energy is focused, it makes you very prepared to pick up where things left off, for another crisis.

Beckworth: Yeah, absolutely. And even before this crisis began, as you noted, you were doing some other work. Before you started your Substack page, you had some great threads on Twitter, of all mediums. And you did several on the repo spikes late last year, when we saw repo rates jump real high, what was going on. And you really got into the plumbing of monetary policy and what the Fed was doing. And it was interesting to see people like you and George Selgin converging on a similar understanding, what was going on.

Beckworth: One thing that makes your work really fascinating to read is that you really know the institutional details of the plumbing of monetary policy here in the United States. And I think that's tied to your training in post-Keynesian economics. And I say that because I had another guest on the show, Marc Lavoie, who is a post-Keynesian, a very prominent one. And he had a great paper on the Bank of Canada's corridor system. So he really knew the institutional details, the inner workings, of how a corridor system works versus a floor system, and particularly how it's run in Canada. So, do you think part of your deep knowledge of the plumbing is tied to your connection to post-Keynesian thinking?

Tankus: Absolutely. I think it's very important, both post-Keynesian economics and, specifically, Modern Monetary Theory, is definitely what introduced me the most to that sort of structure of thinking. And very early on I realized that... or relatively early on, a couple years in... that really understanding monetary policy and financial positions was about understanding basic accounting. So I would open up a Google Docs, which is, basically, what I've been doing with writing the Substack, and just draw out examples of transactions, and follow it through a set of balance sheets. And when you do that, you can sort of combine the T accounts approach with sectoral financial balances. That sort of rote effort, which is like doing problem sets, really, really gives you a good grasp of the underlying mechanics. And that combined with legal knowledge can really get you into the institutional details. And I think that was critical for me, absolutely.

Beckworth: So you think it's essential for all economists to take at least a basic course in accounting so they understand balance sheets, income statements, things like that?

Tankus: Yeah, I think it is. But I also think it's important for it to be integrated into economics education at the ground level.

Beckworth: Okay.

Tankus: I think a ground-level understanding of payment systems. Because learning accounting is important, but accountants are doing, basically, really, micro-accounting. They are dotting all the I's and crossing the T's to make sure that you don't violate regulatory rules or tax rules. And there's things that, to really grasp what's going on operationally, that you would draw in T accounts in a macroeconomics class, which isn't necessarily what you would do in a conventional accounting class, which is microeconomic accounting. And that's important to understand, but I think that at another level you need it integrated directly into the macroeconomic education. You need to learn sectoral financial balances. You need to be able to understand surpluses and deficits, and how they're interrelated across sectors. And it's a sanity check to whatever theory you have.

Tankus: People often say, "Accounting isn't economic theory, and you can't simply derive a causal argument from accounting." But there are all sorts of stories that economists have been inclined to tell over many, many decades that don't necessarily stand up to scrutiny or are accounting incoherent. And so that basic accounting check, "Does my story make sense if I apply sectoral financial balances? Does my story have households having a surplus, and corporations having a deficit, and governments having a deficit, and does that add up to zero?" So on and so forth, for however many sectors you're talking about. That fundamental principle is critically important, and it's also just essential to understanding the plumbing of payment systems, which, as I've been starting to write in my Monetary Policy 101 series, I think macroeconomics would do a great job if it was built up from that payments layer foundation. And a lot of confusion would be eliminated if you understood everything from a foundation of the mechanics of payment systems.

Beckworth: I agree with you. I have benefited greatly from learning the sectoral balance sheet approach. It helps you think through things clearly. And, I agree, it's not a theory per se, and you've got to complement it with a certain vision of the world, a certain macroeconomic theory. But it's very useful. Like you said, a check... what you're learning. And I think it's useful at all levels.

Beckworth: For example, prior to this crisis, there were a number of stories about how global debt to GDP has risen to 200% plus, and, "Oh, the humanity!" you hear. But then, the flip side of it is, well, that must mean there's a lot of savings out there, too, right? If there's that much debt out there, the flip side of it is that there's savings somewhere out there as well. And maybe the question should be, "Why is there so much savings?" as opposed to, "Why is there so much debt?" And that takes practice, to think through both sides of that balance sheet, the global consolidated balance sheet. So I think it's a very useful contribution that all macroeconomists could benefit from. I know I did immensely.

Beckworth: And along those lines, I want to get our conversation started on first issue related to this crisis. And I'm going to call this accounting gimmickry. And it's tied to the fact that Congress is putting a lot of the heavy lifting on the Federal Reserve. And before we get into that, great detail, I want to come back to this idea of thinking through the balance sheets, and particularly the consolidated balance sheet of the U.S. government. And to maybe motivate this, I want to read from one of your posts from Notes in the Crisis. And, again, we'll provide links to this. And you say this. You say, "Central bank independence is the crown jewel of mainstream economics. It is seen as critical for proper macroeconomic management of the economy. What's strange about this consensus is that, at the most basic level, the Federal Reserve doesn't have independence."

Beckworth: And I think that's a great starting point in terms of thinking about the consolidated balance sheet and accounting gimmickry. So to start us off, why is it a façade to claim the central bank is independent?

Is Central Bank Independence a Facade?

Tankus: It's a façade at the most basic level that monetary policy, specifically purchase and sale policy, operates through buying one fundamental instrument, the Treasury security. It's the most important instrument in monetary policy. Obviously, that's changing somewhat in this crisis. There's an expansion of the Federal Reserve's operating procedures and what they're considering purchasing and selling, but at the fundamental level, the Treasury security is the basic instrument. It's the atom of monetary policy, specifically purchase and sale policy. And that instrument, that most basic atom, is not issued by the Federal Reserve. It's issued by the Treasury.

At the fundamental level, the Treasury security is the basic instrument. It's the atom of monetary policy, specifically purchase and sale policy. And that instrument, that most basic atom, is not issued by the Federal Reserve. It's issued by the Treasury.

Share this

Tankus: If the Federal Reserve wants there to be, say, $50 billion outstanding in 10-year maturity Treasury securities, it has to call up the Treasury and coordinate with the Treasury's debt management department, and cajole them, ask them, maybe beg them to not issue 10-years this week, or this month, or this year, and focus on issuing five-years or focus on issuing 20-years. At the fundamental level, that distribution is not solely up to them. Now, of course, they can go in and purchase after the fact, but then that, of course, expands their balance sheet. And they don't necessarily have the ability to sell the instrument of the maturity that they want to.

Tankus: So, for example, if they want there to be only... let's say, just for sake of argument... only seven-year Treasury securities outstanding, and they want only seven years, they need the Treasury to issue as sufficient a number of seven-year maturity Treasury securities. Otherwise they're not going to get the monetary policy they want. So that distribution across the yield curve is not solely up to them, and to a large extent it is determined by the issuer. Now, luckily for the Federal Reserve, it has had basically cooperative Treasury for many decades, which cooperates in all sorts of ways in issuing the securities that they want to. Especially, in extreme circumstances, issuing Treasury securities even when the U.S. government doesn't currently need money in its account to spend. The Supplementary Financing Program in 2008. But it won't necessarily always have a cooperative Treasury Department. And, by definition, if you need a cooperative Treasury Department to conduct the policy you want, you're not independent.

Tankus: And so at this basic, germ level, the Federal Reserve does not have operational independency conducting purchase and sale policy. I think it's a very overlooked point in understanding and conceiving of central banking events.

Beckworth: Yeah, I think many people, including myself, a decade ago would've said, "Look, the Fed has its levers. It pulls them. It sets rates. It gets the policy it wants, and it goes from there." But, in fact, like you mentioned, it definitely runs up against what the Treasury's doing. So, even during the last crisis, I believe, the Fed's outstanding stock of Treasury securities was about 20% at its highest. I know it's going to be much higher now, but back then, 20%... That means 80% of Treasuries were held by other parties, outside the Fed. So they weren't a big, big player relative to the overall market.

Beckworth: And that was something that people started pointing out. "Well, how effective is QE? I mean, is QE really making much of a difference if there's far more issuance coming from Treasury, and the Treasury has all this control?" So I think that's a great point, and it's something needs to be really considered carefully when we think about macroeconomic policy. What really drives countercyclical macroeconomic policy.

Beckworth: Here's another way I've thought about it. And tell me what you think of this illustration, if it's a useful one or if it's too confusing. But if we think of a scenario, let's say, where the Fed becomes insolvent... And by that I mean there's a big hole in its balance sheet. So let's say it bought up a bunch of mortgage-backed securities. Let's go to an alternative universe, 2008. Buys a bunch of mortgage-backed securities up. For whatever reason, they all go belly-up. Those homes are underwater. And so the Fed loses value on its balance sheet on the asset side. And it has all these liabilities it created, all these bank reserves. And I believe the traditional concern would be, "Well, you don't have assets on your balance sheet now to retire or pull in as reserves if you want to in the future, which therefore, at some point, it could lead to inflation." And if that far-fetched scenario were ever to happen, what would come to the Fed's rescue? Treasury. Tax-payers would, right? They would bail out the Fed.

Beckworth: And so, implicitly, even today or prior to the crisis, Treasury is always there, backing up the Fed. The Fed's solvency is integrally linked to the Treasury solvency, to what taxpayers are willing to do. And, in my mind, that clearly connects the two together in a way that's important for the determination of the price level. Does that seem reasonable?

Tankus: Well, but where I would disagree is legally the Federal Reserve is responsible for its own accounting rules, so the Federal Reserve can change its accounting treatment of... This gets back to, as you started with this, accounting gimmickry, and this can be very, fundamentally important, in terms of... This might not be true for certain other central banks. You can make a point that you can have a set of laws that say that when a central bank has a negative net worth, for it to continue operating it needs to get support from the Treasury. But that's not the law of the land in the United States. The law of the land in the United States is that the Federal Reserve has control over its own accounting rules. It can simply say that it can book losses as a negative liability... in other words, as an asset towards the Treasury in terms of future remittances... and balance its books in such a way that its net worth isn't affected. And you can say that's an accounting gimmick, but it's an important accounting gimmick because, since it's been granted this latitude, it can operate as before.

Tankus: Now, as you say, they might not have sufficient assets in the coffers. They might not have sufficient Treasury securities to conduct the monetary policy that they want. And in that case, they do need to go to Treasury, but not for a recapitalization, for a bailout, but for more supplementary financing operations. Issue more Treasury securities to drain settlement balances, fill up a Treasury account at the Federal Reserve, and on a consolidated basis gets settlement balances out of the system. Or, on the other hand, as it's been authorized, it can issue term deposits, basically saving deposits, draining those settlement balances and also, because it has the ability to pay interest on settlement balances, and the very least can conduct interest rate policy as it wants, even if purchase and sale policy is affected.

Tankus: I mean, this is a case where I think the quote unquote accounting gimmickry... or, less cynically, just the legal rules around accounting for administrative agencies, especially the Federal Reserve... is very important.

Beckworth: No, I completely agree with all that. And really, this is a question of how much negative equity on the Fed's balance sheet before it becomes an economic problem. So I think one thing I often wrestle with is there's an implicit asset on the Fed's balance sheet. And that is the net present value of future seigniorage flows. And I think what you're getting at, what it can do, is it can tap into that if it does have a hole. So the question is, when the hole gets big enough, that you exhaust that flow of future seigniorage revenues. Then it has to go to Treasuries. And I guess my bigger point is, though, at the end of the day, the Fed in an extreme situation would rely on Treasury, which goes back to show the importance of the taxpayer support for what the Fed is doing, at least indirectly. But I agree with you. In terms of legality, accounting rules, the Fed can defer some of those losses if the losses aren't that big.

Tankus: Yeah. But, I mean, in principle, I don't think that the problem is the degree of losses. I think almost any degree of losses, as long as it's blessed by the current legal structuring in terms of the Federal Reserve's accounting rules... Obviously, the Federal Reserve would probably want some explicit indemnification for a crisis. And we can talk more about that later. But I think that, in principle, any degree of losses is okay. And the problem for them is simply sterilization, quote unquote, is the ability to drain settlement balances out of the system.

Beckworth: Great points, Nathan. Let me apply this question in a different setting, different context. So, George Selgin made a point on the podcast a few weeks ago, and I kind of echoed it on Twitter. But after reading one of your posts, I was given pause in terms of what the actual outcome may be. And this is the argument that was made. And this goes to the heart of whether the CARES Act, and the way the relief effort is going, is really the best way to do it. And so George is making this argument... I've been very sympathetic to it, and I think you are, too... that Congress should do much more direct funding, as opposed to going through the Federal Reserve... is doing kind of accounting gimmickry, kind of off balance sheet. And one of the points he makes is that if you fund most of the relief effort with the Federal Reserve, it's going to be funded through bank reserves. The price of the bank reserve, or the cost of the bank reserve, is interest on excess reserves. That's a short-term overnight rate that can go up.

Beckworth: So let's imagine some point in the future there's a recovery. Those rates might go up. It might increase the cost to funding the relief efforts that was incurred back during the crisis. On the other hand, if Treasury funded this so that Congress was, say, sending out more grants... If Treasury funded it by issuing, say, long-term bonds, you might be able to lock in low interest rates. And I think that makes a lot of sense if you stop the analysis there, but the thought hit me as I was reading your piece. This assumes the Fed is not going to buy up those Treasuries, that the Fed's not indirectly financing it, right? So, we know there's going to be big deficits... going to be run over these next few years. And for that argument to hold, if I'm getting my thinking right here... is that it would require the Fed not to buy up those Treasuries. Because if the Fed did buy up those long-term Treasuries, it would still, ultimately, be funded by reserves. Am I thinking through this correctly?

Tankus: Yeah. I think that's right. I think that's a very critical point from that post that I was making... is that you hear this all time. In the post I specifically referred to Jim Cramer of CNBC, but it's all over the place, the thinking, "Well, you can finance long term, lock in a low interest rate, quote unquote, and then you spend as you want." Whereas if you've relied on the Federal Reserve's balance sheet and such, as you say, it'll be funded by issuing settlement balances and go into a checking account of a bank, at the central bank.

Tankus: But the problem is that, as we were just talking about, the Federal Reserve has a desired distribution of Treasury securities across maturities. And the thing it absolutely wants the least in this crisis is for the Treasury to start filling up its checking account by selling 30-years. That tightens credit conditions. It potentially raises long-term interest rates, right when that's what the Federal Reserve wants the least. And if the Federal Reserve then goes around buying up those 30-year Treasury securities, then we're back to the place that we started in Selgin's example, which is on a consolidated basis of this spending being, quote unquote, funded by issuing settlement balances.

Tankus: So I think that that is a very key point of balancing out and looking through the balance sheets, is that the idea of financing spending in recessions with long-term issuances just isn't consistent with how the Federal Reserve conducts its purchase and sale policy. And so either way you're going to be relying on short-maturity debt. And if you're worried about interest rates and interest payments to the level of GDP in the long term, you're going to have to tackle the Federal Reserve's interest rate policy, and purchase and sale policy, directly rather than trying to find a way around it by issuing a long enough maturity Treasury security, or whatever other structure you come from. This is a key point from monetary theory, that the primary determinant of the level of net interest payments to the private sector, or to the rest of the world, is the level of interest rates and the Federal Reserve's monetary policy more generally, and not specific choices on the fiscal side, that the interest rate on national debt from monetary liabilities is policy variable, and you have to design your policy around it being a policy variable.

The idea of financing spending in recessions with long-term issuances just isn't consistent with how the Federal Reserve conducts its purchase and sale policy.

Share this

Consolidated View of Public Finance

Beckworth: Okay. Well, let's do a thought experiment. Let's take a country, and let's consolidate the Fed and the Treasury, or the central bank and the finance ministry. And in most advanced economies, that's not the case. They're separate, as you mentioned in your post. Inflation targeting is kind of the norm around the world. So these roles are pulled apart as much as possible, even though to some level it's a façade, as we mentioned earlier. But let's say we do away with the façade; we bring them together. And we do see that. And we see that in extreme situations often. So in a case where there's hyperinflation, it's true that central bank, finance ministry become one. But let's think of a scenario where you don't have hyperinflation, and you've consolidated the central bank in a finance ministry. And I can think of some historical cases where this'd be the case.

Beckworth: I have a book here in front of me. It's Richard Timberlake's book “Monetary Policy the United States: An Intellectual Institutional History.” And he actually has a chapter... It's real fascinating. And the title of the chapter is “The Independent Treasury System Before the Civil War.” And he goes into discussing how the U.S. Treasury did open-market operations and did counter-cyclical policy.

Beckworth: So, I know you know this, but for some of our listeners who may not, this is the period after the Second Bank of the United State is shut down by Andrew Jackson, when he was president. He got reelected on a mandate to shut down the awful central bank that we had back then, the Second Bank, as he perceived it. And so you're just left with Treasury, and, ironically, their view was if you go to the central bank, there'll be less government intervention. But it ultimately fell upon Treasury to kind of act as a central bank and as a finance ministry. I want to know, are there any other examples of that, or any ways of thinking how that would actually play out in the real world if we had a country where these operations were consolidated, but we weren't into hyperinflationary environment?

Tankus: Really, U.S. history is the most prominent example, and not just pre-Civil War and post-Civil War, postbellum policy and Civil War policy, of course. But also the American colonies, before the United States was formed, also effectively ran on such policies, where they emitted bills of credit, and they had bills of credit that had different maturities, effectively had these central authorities who were emitting a various range of maturities and then redeeming them through their acceptability in payment of taxes. So the unique feature is, rather than having a 10-year bill that you would then cash in for cash, or gold, or silver, simply you would reach the point where that was currently acceptable in payment of taxes at face value, or even accepted at a discount a year or two before its official maturity date. The U.S. is really the richest history in terms of-

Beckworth: Okay.

Tankus: ... in terms of that consolidated approach. But, of course, the other thing that would be obvious to bring up is the medieval world in general. You can think of that as, essentially, a consolidated approach, where you have money financing through coinage, which isn't necessarily all just metallic coins but can also be tokens where the metallic content is relatively minimal. There's, of course, tally sticks in England before the Bank of England and even at the same time periods. So I think that primarily it's a historical example that I would focus on. I think there's some other examples. I think there's some Brazil 19th century history, but I would have to refresh myself on that history to get into the details. But I think that the ideology of central bank independence, and the innovation of central banking, has been pretty dominant in the world. Though different legal arrangements between a treasury and the central bank, other arrangements where the central bank is pretty subordinate to the treasury and more like a bureau in it. But there's usually, nowadays, a distinct entity.

Beckworth: Yeah, and the motivation for that line of thinking, kind of the mainstream view, is we want to remove the temptation to monetize debts, to fund government operations directly and avoid hyperinflations, which is a understandable concern, but I think what you're advocating... And I want to get to this in just a minute... is a move toward a consolidated financial ministry, which would be like, in our case, the U.S. Treasury and the Fed being one organization, and doing so in a very responsible manner, where you would have price stability as well as an opportunity to do counter-cyclical macroeconomic policy. Is that fair?

Tankus: That's definitely fair, although I would say the initial proposal, from our point of view, isn't subordinating the Federal Reserve to the Treasury, and in fact, as we intro'd with, in a lot of ways it gives the Federal Reserve more independence. For us, it’s about clarifying the division of powers between the central bank and the Treasury. And then, I think, after that there's an argument or debate to be had about the proper role of interest rate policy. But first we want to clarify things, clarify that the debate, in terms of monetary policy, isn't that, "Oh, we need monetary policy to be the Treasury's piggy bank." Instead it's about what's the proper way to tackle our demand management policy and clarify that whatever securities that some government agency is issuing, they're monetary policy tools, or should be monetary policy tools, and shouldn't be treated as financing tools.

The initial proposal, from our point of view, isn't subordinating the Federal Reserve to the Treasury, and in fact, in a lot of ways it gives the Federal Reserve more independence. For us, it’s about clarifying the division of powers between the central bank and the Treasury.

Share this

Beckworth: Yeah. I mean, this discussion is bringing to light that what you're advocating is a more subtle, sophisticated argument than simple, "MMT wants to print lots of money." Because you often hear that caricature, I'm sure, a lot. People make that claim. But you're advocating a consolidated view of government finance. And let's use that to move to your argument, or your case, for the Federal Reserve issuing its own securities. So walk us through that. And how would you see that happen and evolve over time?

Tankus: Yeah. From my point of view, it logically follows from the idea that the Federal Reserve's conduct of monetary policy is interfered with by having to rely on Treasury issue the sort of atom of monetary policy or the atom of interest rate policy, Treasury securities.

Tankus: And the obvious corollary is we need to get the Treasury out of the issuing Treasury securities business. It needs to rely on other ways to finance itself, which we can talk about in a bit. And we need to make that the Federal Reserve's job since Treasury securities are about monetary policy. They're about providing safe assets to global investors. They're about draining settlement balances from the system, so to manage credit conditions, the economy. They're about setting interest rate policy, and affecting investment decisions. And it's the borrowing cost of private firms and municipal governments, you name it. That it's much more proper for the Federal Reserve to be authorized to issue securities in whatever denomination it wants, whatever maturity it wants, and it to be the final determinant of what those are, have the authority to buy and sell checking account balances with commercial banks. And when you put all that together, gives the Federal Reserve full control over how much settlement balances are circulating in the system, control of how many Federal Reserve securities that there are outstanding, and the ability to provide securities as needed.

Tankus: So, for example, there's a whole debate around shortages of safe assets. And the way I really think about this... and I wrote about this on the site is that really what we're talking about is a shortage of large-denomination money. The Federal Reserve ensures that there's ample small-denomination money by selling it at a fixed price. A dollar gets sold for one settlement balance. It gets bought for one settlement balance. And it doesn't buy and sell physical currency based on the Treasury's need to finance itself. It simply supplies whatever amount as needed so that people have the physical currency that they need. And a Federal Reserve that was freed from the need to coordinate with the Treasury in terms of security issuance could issue as many short-maturity securities as it wanted to supply global investors with large-denomination money and end the shortage of large-denomination money, as needed.

Beckworth: Very interesting. So, Nathan, under your ideal arrangement, would the Treasury never issue debt and only the Federal Reserve issue debt?

Tankus: Yes. That's definitely our ideal arrangement, would be, as a previous guest Peter Stella mentioned, you can have some issues where investors are confused about which one would be the best instrument to purchase. So we would prefer that it would be one instrument circulating. The difference is that Stella wants that instrument to be Treasury securities, and we want that instrument to be Federal Reserve securities. Yes. Our preferred policy would be that the Treasury doesn't issue any Treasury securities at all, and that it relies on another form of finance.

Our preferred policy would be that the Treasury doesn't issue any Treasury securities at all, and that it relies on another form of finance.

Share this

Beckworth: And under this framework, the Fed would provide securities to accommodate federal government's financing needs, as well as deal with economic concerns such as countercyclical management.

Tankus: So, when you get to brass tacks, we think that the Treasury should be money financing itself, but not, say, in overdraft from the Federal Reserve. It would directly be issuing its own money. The most obvious, kind of flashy example would be large-denomination platinum coins, which are technically already authorized by 5112(k) in the U.S. code. You can have different authorities. You could authorize it to issue its own digital fiat currency, or you could authorize it to be able to issue settlement balances within the Federal Reserve's payment system. There are a number of ways that you could go about organizing the Treasury money financing itself.

Tankus: The thing is just that, since the Federal Reserve has the ability to issue securities on a consolidated basis, the consolidated federal government isn't necessarily going to be operating with money finance. It depends on what the Federal Reserve's judgment of how many settlement balances that it needs to be in the system. So, if they judge that we need to issue lots of Federal Reserve securities, or we need to sterilize all the emissions that come from the Treasury that they can in a not-consolidated basis, there's no money funding after the fact. Or even the Federal Reserve can choose to issue securities well more than the Treasury is emitting. And, in that case, on a consolidated basis you would get, quote unquote, money destruction or a ton of money sterilization at the same time as the Treasury is operating under money finance.

Beckworth: Okay. Very interesting. Now, one of the hot topics leading up to this crisis was whether the Fed would use yield curve control. And I get the impression that under your proposal, yield curve control would be very easy for the Fed to do with these securities. Is that right?

Tankus: Yeah. That's absolutely true. If you read Kenneth Garbade's recent paper... he was a long-time New York Fed economist. One of the things that you notice when you get into the history of yield curve control in the United States, with the Federal Reserve before 1951 and the Federal Reserve-Treasury Accord, is that a lot of the difficulties aren't about the ability of the Federal Reserve to cap interest rates. But it's about the ability for the Federal Reserve to set floors of interest rates and to fully control the yield curve because they have to rely on the Treasury to issue securities, and the Treasury's interest... or what the Treasury sees, itself, as its interest in terms of trying to reduce as much as possible the interest payments on the national debt... comes into conflict with what the Federal Reserve thinks is proper monetary policy. And so this proposal completely gets rid of that problem.

Tankus: And, in fact, that experience in the 1940s is the primary motivator for the earlier version that we discovered of this proposal, from Albert Hart, an obscure economist now but was a big monetary economist in the Treasury during World War II and then a Columbia economics professor. And he proposed precisely a system where the Federal Reserve would have sole responsibility in terms of issuing securities. And that was the motivation, was seeing the difficulty of yield curve control functioning when you had to rely on Treasury to issue securities.

Beckworth: That's a very interesting paper. We'll make a link to that as well on our show page. Nathan, let me switch gears a little bit here and go to what has actually been done in terms of relief for this crisis. So, we have the CARES Act. We have the Federal Reserve intervening very aggressively. What is your big-picture take on how the government has responded to this crisis?

Evaluating the Policy Response to COVID-19

Tankus: My big-picture take is that the Fed has responded extraordinarily far past that most people would have imagined it would have responded. And, at the same time, the overall macroeconomic policy mix is very underwhelming because the problem isn't primarily access to credit. The primary problem is that you've had this huge loss of income. And in this way I agree with Selgin on the podcast last week, that that's the primary problem, and more debt isn't the solution. The best that the Federal Reserve can do, absent some very creative policies, is make it easier to get into debt. But state and local governments, and the private sector, don't need more debt. They need something that can replace their income, which means grants and such.

Tankus: And, on that scale, policy has been far too low. For example, the accounting gimmick, quote unquote, where $454 billion of the CARES Act goes to capitalizing of Federal Reserve emergency facilities. On a consolidated basis, that doesn't provide any sort of income or other support to the private sector, or state and local governments. Now, you can argue they wouldn't be conducting as much credit policy without that, but there's alternative ways to encourage the Federal Reserve to use its powers, alternative accounting gimmicks, as it were, that don't take away from the headline appropriations number, which costs a lot of political capital. If the headline number had been 1.6 trillion or something like that, or 1.7 trillion, rather than 2.2, there would've been more political energy around increasing the amount of grants and increasing the amount of support to state and local governments, rather than the headline number being 2.2 trillion, where politicians are very open to pat themselves on the back for how big and bold they are going.

Tankus: So I think overall we're doing far too little on the income and grant side. And what we are doing, the forgivable loans running through the Small Business Administration, are underwhelming and have problems that, historically, Small Business Administration programs have, in terms of being administered by banks and being slowed down and filtered through all the biases that banks have.

I think overall we're doing far too little on the income and grant side. And what we are doing, the forgivable loans running through the Small Business Administration, are underwhelming and have problems...

Share this

Beckworth: Yeah. I agree with that view entirely. And I don't blame the Fed. I don't blame Treasury. I mean, it's the fog of war there. There's constraints kind of built in place. But Congress did have some political capital, as you said, and the question is how could it have been spent better? So instead of going down the path they did, could they have had a larger amount of grants going directly through the SBA, or some other facility, to small businesses? Or maybe they could've spent that capital on changing the Federal Reserve Act so the Fed could do more in terms of grants. Any idea of where you would've gone in a perfect world, if you could've waved your magic wand and caused it to happen?

Tankus: Yeah. So, last week I wrote a proposal that was, basically, "Why don't we do emergency basic income for businesses?" That, rather than these complex conditionalities, running it through banks... We have payment processors that are used to dealing with businesses. It's a lot easier to get businesses' bank account information and have businesses come in, to get access to payment information for them, than it is to get it for individuals. And if you just eliminate any sort of conditionality, and just, "We'll give you X amount of dollars per worker you have employed," or some other measure of the size of the business and their needs for support, that would've, I think, been administratively much simpler and would have gone everywhere. And I think this is a time for going big and bold, and it matters a lot more that you miss giving to money to people than you give the money to a business that doesn't necessarily, quote unquote, need it.

Tankus: And if there are people more of my political persuasion, in terms of more liberal or left side of the aisle, then you deal with those concerns by suspending payouts. You deal with it on the distribution to owners side, on the dividends to shareholders, on buybacks. And you freeze those during the crisis, which we shouldn't be getting a whole bunch of payouts anyway, and then we can worry about the distributional concerns after the fact. But the problem now is to immediately get money in the hands of those who need it and make sure that businesses keep up their payrolls. And if we're going to do that, we should just do it, rather than this partial, sloppy way with specific appropriations to the Payroll Protection Program. Instead it should just be distributed universally, and it would've gone out the door much, much quicker.

Beckworth: Yeah. Great thoughts. And indeed, it's been sloppy. And the system is set up for bigger businesses, corporations, because they do have collateral. They do have credit ratings. They do have ways to evaluate them that just makes it easier, and the government doesn't have the infrastructure in place to suddenly set up shop and to reach that last mile into small businesses, whether it's the Fed or the SBA. It's really been a struggle for them to do that. So I think in an ideal world we would've done something more along the lines of what you've suggested.

Beckworth: So this has been a very fascinating conversation, and it's good to think through these issues, and I really do encourage our listeners to engage with the sectoral balance sheet approach if they haven't already. But we're running low on time here, Nathan. And I want to do one other area on this show. And so I'm going to switch gears and ask you a question about your view of money. So, I know the post-Keynesian view of money is a state view of money, that it requires the state to back it up for it to gain value. Or I think you call it a chartalist view. And there's another view, kind of an emerging order view. And my personal view is it could be both of those involved. But I would ask this question for someone from your perspective. What role is there for, say, network effects? So, the state gets the money up and running, according to your view. But can network effects get a life of their own so that the value of money comes from something other than just the state backing it?

Post-Keynesian View of Money

Tankus: Sure. One thing I would say is, actually, post-Keynesians tend to have either a credit view or come to some sort of modified chartalist view. But in terms of state backing, believing that it is taxes or, more generally, the imposition of non-reciprocal obligations, of imposed obligations onto the population that are denominated money... so court judgements, fines, fees, and such... that view is specifically an MMT view, and then there's historical antecedents who you could label chartalist. But in terms of modern economists, it tends to be concentrated among MMTers.

Tankus: That said, that view that taxes and other imposed obligations back money... It definitely doesn't have no role for network effects. And, in fact, I think that sort of relies on network effects. It's just, rather than network effects being the leading cause, there's some spontaneous formation of network effects that build up in an asset and make it money. The chartalist view, the neo-chartalist view, the MMT view, relies on having a public authority of some sort make a certain instrument acceptable in payments to itself, acceptable in payment of taxes, acceptable in ritual payments to the temple.

Tankus: And what network effects do is they get private actors to adopt it. So once everyone knows that a certain IOU, a certain piece of paper, whatever, has value because it's accepted in payment of taxes, court judgements, and such, private merchants who are within the territory of that political authority tend to start... spontaneously, as it were, or just by nature of being part of that system... to adopt selling things in that unit of account to be able to settle their tax obligation or other obligation. So the standard story... as, say, David Graeber writes about in his book “Debt: The First 5,000 Years,” is if you need to provision an army, you impose taxes, and then suddenly there's a big incentive for the local population to start selling goods and services to soldiers who've been given that money. And then it provides a administratively simple way of provisioning your army. So that two-step, where you impose obligations and then private actors start adopting that unit of account, is definitely a story that relies on network effects. And there's a force that builds up there-

Beckworth: Sure.

Tankus: ... where the merchants you have who are now adopting that unit of account, there's suddenly other people who now want that money, not just to settle obligations, although that's still important to them, but also to be able to purchase goods and services of merchants. And those people don't necessarily have to be the initial soldiers. So it spreads like that. And it can also fail. You can have a money that is only backed by its tax and other backing, and you can have people who choose not to adopt it. For example, they politically oppose that regime, and are doing everything that they can avoid participating in it, and only get just as much money as they need to settle their tax obligation, and that will reduce the liquidity premium that develops around that money. So, absolutely, there's a political, social relation element that relies on adoption. And adoption can fail.

Tankus: Now, the chartalist point, the MT point, is, as long as those obligations are in force, that money will still have some value. It still will be, to a certain extent, money. It just won't be as powerful as it would be if everyone starts adopting and using that unit of account. In the United States sense, the United States is the dollar hegemon. Part of it is the imposition of dollar debts globally, and access to the United States, which is large. But partially is definitely private actors who are making choices to adopt the dollar unit of account in their own trading, in their offshore banking activities. And those are choices. And they're choices based on network effects. So, the short form of that is there is a role for network effects, but it falls along from that imposition of obligation starting point, rather than network effects being the whole story.

Beckworth: Fair enough. Well, our time is up today. Our guest today has been Nathan Tankus. Nathan, thank you so much for coming on the show.

Tankus: Thank you very much for having me. It's great to be here.

Photo by wyliepoon

Support Mercatus

Your support allows us to continue bridging the gap between academic ideas and real-world policy solutions.Donate