July 9, 2020

A Fed for Next Time: Ideas for a Crisis‐​Ready Central Bank: Panel 4 Preserving Monetary Autonomy

David Beckworth

Senior Research Fellow

George Selgin

Senior Affiliated Scholar

How Will the Fed Fight the Next Crisis?

In just a dozen years, the Federal Reserve has faced two severe crises. And twice it has responded by leaning heavily on emergency lending powers it seldom used before by improvising temporary lending programs and taking part in fiscal policy.

In the meantime, the Fed’s nonemergency lending facilities have hardly changed, and may well prove insufficient when the Fed faces its next crisis.

The implication of this is both obvious and ominous: while we still count on the Fed to deal with crises, we no longer know how it will deal with them. Instead of being predictable, the Fed’s crisis‐​prevention methods have become unpredictable–and controversial–adding to, instead of allaying, economic scrutiny.

Can we do better? Can we improve the Fed’s systematic response to crises, making that response both more effective and more predictable? Can we thereby limit the Fed’s entanglement in politics? What can the Fed do to promote these ends? What might Congress do?


  • Sebastián Edwards, Henry Ford II Professor of International Economics at UCLA and former chief economist for the Latin America and Caribbean Region of the World Bank.
  • Peter Stella, former head of the Central Banking and Foreign Exchange Operations Divisions at the International Monetary Fund (IMF) and curator of the website Central Bank Archeology.
  • George Selgin, senior fellow and director of the Center for Monetary and Financial Alternatives at the Cato Institute

Moderator: Victoria Guida, financial services reporter covering banking regulations and monetary policy for POLITICO

Introductory Remarks

David Beckworth: Greetings, and welcome back to our conference titled “A Fed for Next Time: Ideas for a Crisis-Ready Central Bank.” In this conference, we are looking at ways to improve the ability of the Fed to better respond to future crises in a way that avoids entanglement with politics. This conference is being hosted by the Mercatus Center and the Cato Institute. George Selgin and I have been the hosts. We've had three great panels so far, ones that have looked at reforming credit policy at the Fed, better defining the boundaries between fiscal and monetary policy, and one on modernizing liquidity facilities. If you haven't already seen them, please go check them out on the Cato website for the conference event.

George and I have been taking turns hosting the conference, and it is now my pleasure to host the fourth and final conference that is titled “Preserving Monetary Autonomy.” This will look at the issues of fiscal dominance and preserving Fed independence. I will now turn our program over to Victoria Guida. Victoria is the Fed reporter for Politico. She also covers bank regulations. Victoria will introduce the panel and the program. Victoria, it's all yours.

Panel Discussion

Victoria Guida: Thanks, everyone for joining us for the final panel of this Cato series. Today, we're going to talk about how the Fed can preserve its monetary autonomy, with discussion of the proper role and relationship of the Fed and the Treasury Department. I'm joined by three esteemed panelists with a range of international experience: Peter Stella, who spent 25 years at the International Monetary Fund, including as head of the Central Banking Division, where he was responsible for overseeing quality control of the fund's advice on central bank operations for over 100 member countries' central banks. He also curates the website Central Bank Archeology. We also have Sebastián Edwards, a former World Bank official who teaches international economics at the University of California at Los Angeles, and is the author of American Default: The Untold Story of FDR, the Supreme Court and the Battle Over Gold, as well as George Selgin, who runs Cato's own Center for Monetary and Financial Alternatives, and has authored several books, including the recent and relevant The Menace of Fiscal QE.

A few logistics on today's panel. You all attendees can submit questions via the Cato webpage, Facebook, Twitter, and YouTube at the hashtag #CatoEcon. Peter, we'll start with you, if you want to take it away. I'm actually told that we need to move to Sebastián first. 

Sebastián Edwards: Thank you, Victoria. I was ready to take notes on what Peter was going to say on his archaeological analysis, and then take it from there, but I guess it's my time now to start. So I've been fascinated by this series of round tables, and I've watched all three of them, and I have learned a lot. And at times, I wanted to jump in and talk at those other meetings. So what I'd like to do today is stay roughly within the topic of our own discussion here, which is how to avoid the entanglement between the fiscal and the monetary side.

I'd like to first start with, what are really the questions that we're asking? Then I'm going to talk a little bit about theory, then I'm going to focus on some debates that we have had, then proposals. And if I have time, I'll talk a little bit about history, picking up from what was said in the previous sessions.

So what are the questions? So the question I think is very clear in David's introduction to today's meeting, and in the title of the series, how do we avoid the intermingling in a not useful or even dangerous way between the Treasury and the Fed? And this is something that we, as undergrad students, when we were in college a long time ago, have all looked into. And one of the first questions, if you do macroeconomic theory, is you learn what monetary policy does, you learn what fiscal policy does, and then you ask yourself, "Why don't we combine them?" And then you have to come up with answers, or your instructor comes up with answers, and there are two types of answers. The first one is the difference, or differential effects that the two types of policies have on the economy, and the second type of answers is the dangers that each of those policies carry with them.

In the very simple theoretical model of an open economy, and I'm underlining here open economy, because I think that much of the discussion that we have on these topics is a little bit restricted to closed economy. Did I say open again? In the simple model of a closed economy. The main difference between the two policies is the impact that they have on interest rates, and the dangers is, on the monetary side, if you are not careful, of course, and you are too lax, you will generate inflation, and we know that inflation has a number of costs, including that the bond market doesn't like inflation. And on the fiscal side, you may run into a fiscal crisis, and some kind of insolvency, and you may become like Lebanon, or like Argentina, and today, the news is Turkey, which is about to be downgraded, and then it will happen the move into the frontier markets, which is not something I think that any country is looking for.

Now, I say that this happens mostly in the closed economy setting. Once one moves still at the undergraduate level, and this has to do with what I wanted to mention regarding theory, I think this is Mickey Mouse theory. Once you open, you move into the open economy realm, the main difference between monetary and fiscal policy is the way it affects exchange rates. And in these three wonderful seminars that we have had during the past two weeks, exchange rate has been paying a very, very modest role, and I think that we have to bring it back into the discussion, and when we think of theory, of course, we think of a Bob Mandel about 50 years ago, or a very long time ago, before Bob went into academia, when he was at the IMF and he developed the open economy version of the workhorse of macroeconomic theory.

And I think that we need to talk about one of the issues that we would need to address, as I said, is exchange rates, the effect that having crisis-based excessive QE has on the value of the currency. Steve Roach has now argued that the dollar has come to an end, and that in the next few months, we're going to see a total collapse in the value of our currency. I happen to disagree quite strongly with that view, but we have to bring it in, and we have to bring in the exchange rate issue on the one hand. And related to that, of course, the pass through question, if the dollar does lose value and depreciates in a significant way in the next year or so, what's going to be the pass-through? And whether it is still the case that the pass-through is very, very small, almost nonexistent, in the US. So those are, I think, the questions, and I wanted now to move to ... And the point that I want to make here is the importance of bringing back the exchange rate and the value of the currency into the analysis.

Let me now move to what I think is a key issue, and that has to do with inquiries. I've spent much of my career looking at cases where there is fiscal dominance in monetary policy, and those cases have to do mostly, but not exclusively, with Latin America. Not too long ago, at least for people of my generation, Israel was a Latin American type of country, until Michael Bruno and then Jacob Franco, and Stan Fischer came in and put some order on the inflationary front in Israel. But when you have fiscal dominance, of course the anchor in the economy is lost. And the question is, how does one bring it back? Even when you have some kind of coordination, or when you need, because of crisis type of situation, a coordination between the Treasury and the Fed.

And I think that here, we are a little bit at a loss. What is the anchor, from the fiscal point of view? So I get asked this question very often by investors from around world, and the question at the current time is, "Well, the US had 107% of GDP in terms of public sector debt before the coronavirus crisis. It's going to move to 120, 130, or God knows what. Is that too much?" And then the question is, "Well, what is too much?" And if we remember, not too long ago, in 1998, we collectively, the policy community, academics and so on, decided that the number was 60%. Why 60%? Well, the Maastricht Treaty. And there were only two countries in the Maastricht Treaty that did not comply by the 60%, and grow a maximum debt to GDP ratio. One was Greece, and the other one was Belgium. Belgium fixed the problem. Greece did not, and we know what happened.

Now, after the financial crisis we moved, although with some controversy, 60% to 90%, and that is the famous Rogoff-Reinhart number, the controversy between Ken and Carmen on the one hand, and Paul Krugman, and then the discovery that there was a mistake in the Excel. Now, we're not going to move to 92. We don't know what number. It's going to be 120, 115, and here comes, of course, the question, again, of the rating agencies. What are the rating agencies going to do? Are we going to stand in between? What's going to really anchor the economy going forward?

The second point that I want to make within this debate has to do with the role of velocity. I think that much of the discussion tends to underplay the importance of velocity in determining the final outcome in a macro-economy. And if one looks at the international evidence on the effects of fiscal dominance, one of the big problems is that once fiscal dominance starts doing harm, in the sense that prices start going up very quickly, velocity increases very, very dramatically, and that feeds back into the problem, and we see it in country, after country, after country. I'm not saying that that's going to necessarily happen in the US, but I think that it is an area that we need to look at with greater care.

Let me try to touch on two additional points. In terms of practical proposals, at some point, George mentioned that we've had a great discussion, but in terms of practical questions, we're not having a lot of options put on the table. I think it's really very hard. I think that the notion that Elga brought up in one of the other meetings, a paper that I think she did with Stan Fischer on having a contingency plan prepared, makes some sense, but that brings in, How do you put it in place? Who's going to participate? Whether the possible beneficiaries of such a contingency plan, say banks, do they have to enroll ahead of time? That brings to the fore the stigma issue.

Now, the mother of all contingency plans, if I may say so, is the flexible credit line of the IMF, which was, I think, the brainchild of Stan Fischer. I've mentioned Stan many times during my talk, and that until the coronavirus crisis, only three countries had signed onto. One was Mexico, and he'd signed onto because Agustin Carstens was the number two or number three person at the IMF, and he made sure that that happened. The other one was Poland, and the third one was Columbia. So there were no takers. Even though it was very generous, it made a lot of sense, it was much better from our cost benefit analysis than holding very large stock of international reserves. Stigma played a very, very important role there.

Second point that I want to make in terms of proposals, and I'm coming to the end of my 15 minutes, has to do with if one has contingency plans, I think one has to have a crisis committee that is ready to come in and start planning and implementing the emergency plans for the future. And the question of governments there is a very important one. All I'm going to do is throw in the question of the makeup of the board of governors. And whether, as some people have proposed, we should move more into the agency model, where we have a system where even if a party stays in power save for two administrations where the president is reelected, cannot actually appoint all of or a great majority of the board of governors and where we will have that one political party cannot have more than a slight majority.

The RFC, and now I'm going to move to my history final minute or so. There was a lot of talk in one of the sessions about the Reconstruction Finance Corporation, which of course was one of the first institutions that we had to deal with crises and it was founded in 1932. I thought that it was fascinating, the conversation of how difficult it was to unwind it, it took all the way until at 1957. The RFC was run like an agency and it couldn't have more than three members from one political party in the governing body and that was very, very important.

The other point that I want to make in terms of history, the RFC played a key role in the year 1933. And it was at the core of the beginning of the last wave of banking crisis when it refused to lend money to Henry Ford's Guardian, banking holding corporation in Detroit. It then of course played a key role when it saved all the Class B banks after the banking holiday and the Emergency Banking Act of March of '33.

And finally it played a very strange role in the gold buying program that George Warren, the economist from Cornell convinced President Roosevelt to put in place in October of 1933. And I think that once you have an institution like the RFC, there are good things, there are questionable things, and there are some dangerous things, at least that history tells us that such institutions can go through that can create a lot of problems. So, Victoria let me stop now and I'm looking forward, I will take notes to what Peter has to say. Thanks.

I think that once you have an institution like the RFC, there are good things, there are questionable things, and there are some dangerous things, at least that history tells us that such institutions can go through that can create a lot of problems.

Guida: Thanks so much, Sebastián. And it looks like we have Peter here. So Peter, if you want to take it away, it's all yours now.

Peter Stella: Thank you, Victoria and thank you to Cato and Mercatus for inviting me to be on this distinguished panel. Today I'd like to divide my remarks into three sections. One is to talk about what are central bank quasi-fiscal operations and why they're problematic. Then I'm going to talk about how some countries have resolved those problems or avoided those problems. And lastly, what sort of approaches might be relevant for the Fed.

Central bank quasi-fiscal activities comprise policy actions executed by the central bank, although that could be replicated by a government expenditure attacks. Quasi-fiscal operations impact the central bank profit and loss account and balance sheet, but they're not incorporated in the government budget. Significant quasi-fiscal expenditures include subsidize and lending to particular industry sectors, such as the domestic automobile industry, or to banks with political influence, including state-owned banks who have political objectives.

Significant quasi-fiscal taxes include high unremunerated reserve requirements imposed on banks, which is effectively tax, and bank intermediation, raising lending rates to borrowers and lowering rates that depositors receive. And of course, most of us have heard of the inflation tax which is an implicit tax levied by central banks through excess money printing.

So what management problems are created by quasi-fiscal operations? Well, if some expenditures are off budget, then the budget cannot reflect national priorities properly. How can we compare the priority of subsidizing school lunches versus subsidizing the automobile industry? If we can't compare them side by side, even if we think subsidies to the auto industry are proper, how can we contemplate the correct way to deliver the subsidy, if it's not compared with other ways the Treasury provide subsidies? Is subsidized credit the best way? Or should we have tax preferences for the industry? Should we give them accelerated depreciation allowances? Or should we just give every automobile worker $20 on their way out home every single day? It's not clear unless we can compare them side by side.

If quasi-fiscal taxes are not in the budget, then how can we determine whether relying on those taxes is superior to any of the thousand ways the Treasury can tax? Taxes on alcohol, tobacco, carbon, personal corporate income taxes. You really can't compare the inflation tax with those taxes unless they're somehow incorporate in the budget.

Furthermore, just like treasury operations, central bank quasi-fiscal operations tend to result in losses or deficits. And if we just look at the conventional treasury outcome, when there are significant quasi-fiscal activities going on with the central bank, we're not getting a good idea of what is actually the national deficit, nor the increase in the national debt or the composition of the national debt because many central banks issue debt in different funds and maturities.

Since central banks tend to be highly profitable and pay dividends to the government, small quasi-fiscal losses really just mean a reduction in the dividend to the government. Now that might be in the following fiscal year, but the pain will be paid, the chickens will come home to roost eventually. When losses from quasi-fiscal operations become very large, central bank equity becomes negative. Well there's a problem with most central bank laws and that is they don't require the Treasury to compensate the central bank losses or for negative equity. And even when the law does, treasuries rarely actually do it promptly.

What happens when central bank equity becomes very negative? Is that the Treasury is realized, during our administration, we're probably never going to see a dime again from that central bank, we're never going to see any dividends. So they begin to think, "Well, what other kind of expenditures can we put as a responsibility to the central bank? Because we're not going to feel the fiscal cost of it." Now that might seem politically palatable, but obviously it's no way to run a country, fiscal management.

I've been to a country, a central bank where the legislature attempted to get the central bank to pay the expenses of the country's Olympic team. It sounds like a joke, but that really happened. That central bank governor told me they had to hire a full-time lobbyist to sit in the Congress 24/7, just to fend off crazy ideas that were coming from the legislature. Who had realized, it's free game, sort of free money and central bank can print it up.

Now let's look at some country examples of how countries have dealt with this in the past. I want to start off by talking about Peru. Peru 30 years ago, was basically undergoing an economic collapse. The hyperinflation, virtually complete dollarization. There was a Maoist insurgency, ‘Sendero Luminoso’ at the gates of Lima. Massive quasi-fiscal operations. What happened? There was a change in regime. There was a new central bank law. They restated the central bank accounts and introduced the recapitalization program for the central bank. What was more important was that the new law basically identify all the quasi-fiscal operations that the central bank had been doing.

Subsidized credit to the government, to the sectors, multiple exchange rate practices and the new law specifically prohibited them. It's quite interesting if you look at the law, it's like they said, "You did A, B, C and D in the past, now you can't do A, B, C and D." There was also a significant fiscal adjustment and the development of a domestic debt market. Both of those factors led to a reduced need by the Treasury to rely on monetary finance. And in fact, the Treasury built up large balances at the central bank. So the Treasury actually wound up lending to the central bank rather than borrowing from the central bank.

This robust system survived many populist presidencies, including the second round of Allende Garcia who presided over the country during the hyperinflation. Chile, which Sebastián certainly can talk about, is an example were an extraordinary strong central bank separation from fiscal policy was introduced in the constitution of the country, in a new constitution. So the central bank autonomy is in the country's constitution. Prohibition of government financing is in the constitution of the country.

There was also a significant fiscal adjustment, development of a domestic debt market, significant buildup in treasury balances of the central bank, because it knew it couldn't rely on borrowing from the central bank.

So the Treasury effectively again, was supporting the central bank’s financial position rather than weakening. Now that law is really the strongest law, we think that a lot of central bank laws are dealing with. And I'd have to say, I put that number one in terms of central bank autonomy. Even maybe a little bit extreme for my taste. But it really has led to or facilitated a lot of significant improvements in Chile, including the development by the government of a sovereign wealth fund which it can use to intervene in a crisis.

Let's turn to a couple of countries that were able to avoid getting involved in this situation from the beginning. So Canada is a good example. During the global financial crisis, Canada generated a very small increase, a temporary increase in the central bank balance sheet. That was reversed. And although we could see 10 years later from those interventions during the crisis, it's an account which has a name called the Government's Prudential Liquidity Management Account. And that was established in the budget to pre-fund the acquisition of cash that could subsequently be used to continue government spending in the face of a potential inability to raise market funding for a period of at least 30 days.

So essentially the only expansion of the central bank balance sheet of the Bank of Canada was an increase in government deposits that essentially pre-funded spending in the event of the next crisis the government wouldn't be able to issue domestic debt for a period of 30 days.

Does that mean Canada didn't support the financial system at all during the crisis? No, but it was done on the government's balance sheet and this is very important. So assistance was provided through the Insured Mortgage Purchase Program, which authorized a government-owned housing corporation to provide longer term funding up to five years to Canadian banks by purchasing up to $125 billion in mortgage-backed securities. So $69 billion in those securities were purchased through competitive option process and this program was wound down. That had to be through the issuance of Canadian government debt and it was wound down by March of 2015. So that's no longer on the government's balance sheet. Had that been on the central bank balance sheet, the Bank of Canada balance sheet would have more than doubled from 2008.

Let me move on to transition to talk about the United States and the Fed. I want to start with the Housing and Economic Recovery Act of 2008. It was passed in July of that year during the Bush Administration. That law gave the Treasury the authority to purchase and sell obligations of Fannie Mae and Freddie Mac. But in order to assume the powers under this law, the Treasury secretary had to determine an emergency exists, and that action is necessary to provide stability to financial markets. And that's a ding to any politician's political capital to admit, "Okay, there was a crisis on my watch." So that's important.

The authority to purchase the assets was extended only through 2019. So it was time limited. Periodic reports to Congress on the program were required. Funding the purchase was subject to US Code Title 31 Chapter 31, better known as the Public Debt Limit. In other words, the Treasury had to finance the purchases of those assets by selling treasury debt.

So what happened? Well by end 2009, the Treasury owned $192 billion in GSC debt. But by March, 2012, it had all been sold back to the market. The Treasury ended it, why? Well, it was the debt limit. The Treasury couldn't raise any more debt. It needs to fund expenditures, so in order to raise cash, it hired a firms to help it auction off the securities so that it could get cash and continuous spending under the debt limit. It's very important to see the different governance structure of that program than what the Fed did.

So at the same time, in November of 2008, the Fed announced a program to buy direct obligations of housing related GSCs and GSC mortgage-backed securities. There was no time limit on that program, the Fed could finance it by selling its holding of treasury securities or with money creation. Obviously with money creation, there's no debt limit in post. There's no limit on that. And once we accept that zero interest rates are our policy, then there's really monetary finances unlimited. So there was no time limitation, no quantity limitation and never any real pressure issue.

Did it make sense for the Fed and the Treasury to be running two virtually identical policies under completely different constraints, governance structures and balance sheets? I don't think so. So what to do? Well, it would be fairly easy to rewrite the Federal Reserve Act to define the Fed's activities to financial market supervision and regulation, maintaining a first-class payment system and to conventional monetary policy, meaning providing lender of last resort facilities to illiquid and insolvent banks.

It's much less clear on which side of the monetary fiscal line you would put what I call market intervention policy. We could simply say, "Just don't do it." But I think that's unrealistic, particularly given the importance of capital markets in the US and the global economy. So what I propose is a third entity, in between pure monetary and fiscal policy. Let's call it a financial market intervention authority. It would have its own governance structure, balance sheet, and assigned objectives set out in all.

What I propose is a third entity, in between pure monetary and fiscal policy. Let's call it a financial market intervention authority. It would have its own governance structure, balance sheet, and assigned objectives set out in all.

This third governance structure, the board would have representation from the central bank regulators in the Treasury, whatever the weighting of the power on the board, at least the Treasury should be required to register an on the record vote on policies to ensure democratic voice and accountability. The entity would be owned by the Treasury with paid in and callable capital, which Congress would pre-authorize. It would operate within specific risk parameters in a crisis. It would have to call in additional capital to allow the expansion of the balance sheet. The entity should report quarterly according to international financial reporting standards. This is essential. I am very familiar with both central bank and government fiscal accounting methodologies, and either can adequately cope with the task of providing the information needed, to ensure transparent reporting and accountability. And of course, we'd need an independent audit.

The first thing this entity would do is take over the Fed's existing holdings of private securities and get the Fed back to a much smaller balance sheet on both the asset and liabilities side. That concludes my remarks. Back to you, Victoria.

Guida: Great. Thanks so much. And last but not least, we have George.

George Selgin: Thank you, Victoria. And I shan't thank the hosts of this event since I'm one of them, but I would like to thank all of the participants, including this panel, for a lot of really useful suggestions. Just to mention a few, we have ideas about modifying the Fed's 13(3) authority, the idea of a standing repo facility to serve several purposes, the idea of emergency and emergency fiscal facility, Peter's idea of a financial market intervention authority, and also, the suggestion that Sebastián mentioned, that we might try another RFC. I think though that the general lesson in all of this is that it's very difficult and perhaps ultimately it's impossible to completely keep the Fed's hands clean as it were of a fiscal or credit policy during any major crisis. And particularly during a crisis when interest rates hit their effective, if not zero, lower bound.

What I'd like to do in my remarks is to address a somewhat less daunting challenge, but still a very important one, in my opinion, which is that of keeping the Fed out of fiscal policy in non-crisis times and doing so in a way that it does not impair the Fed's ability to combat crises when they happen. And I'd like to, by the way, point out that Peter's remarks make for an ideal introduction to some of mine. So, perhaps having him come second was ideal after all.

Now, some of the points I'm going to be making here are ones that I elaborate in my book, which I hope is less fuzzy on your screens than it is on mine. Of course, it's backwards, isn't it? But it's called The Menace of Fiscal QE. Please buy it. The basic problem that I'm addressing is one that did not exist until 2008. That year marked the occurrence of what David Beckworth has called the Great Divorce. That is a divorce of the size of the Fed's balance sheet on one hand from the stance of monetary policy on the other. And this divorce has very important repercussions for the possibility of Fed fiscal actions, the extent to which these are possible, the extent to which exploitation of the Fed's balance sheet is possible, even in non-crisis times.

To explain, let me go back to the old days before 2008. In those days, there was no Great Divorce. So, the stance of policy depended on the size of the Fed's balance sheet. To make a long story short, if the Fed bought a lot of assets, if it expanded its balance sheet, that would tend to lower interest rates and, other things equal, raise the inflation rate. Which meant there was only so much asset purchasing the Fed could do before it would risk failing to meet its inflation target, exceeding its inflation target. And that's why it was relatively easy, under those circumstances, for the Fed to fend off any attempts to try to get it to buy assets to support this program or that program while bypassing the usual appropriations process. Because it could say, look, we can't control inflation if we do that sort of thing.

What happened in 2008, as I know most of you know, is that quite inadvertently actually the Fed switched to a so-called floor operating system. That involved two things, actually. First, the fact that the Fed did begin paying interest on reserves, which it hadn't done before. And second, the fact that the banking system was flooded with excess reserves. And that left the system in the state that it's been in ever since where the way monetary control has been exercised was not by regulating the ballot. Marginal changes in the size of the balance sheet don't necessarily have any consequences for inflation by themselves. Because the reserves don't have an opportunity cost, banks will hold as many as you can throw their way. And that can be true even when there's no zero lower bound problem.

Instead, the way we control the stance of monetary policy, that is the way the Fed does it, is by regulating the interest rate on excess reserves. What this means, basically, is that now it's possible, even in non-crisis times when the policy interest rate has to be above zero, it's possible for the Fed to gobble up assets to engage in quantitative easing and still maintain control of inflation by appropriate adjustments of the interest rate on excess reserves. This invites, of course, a greater temptation than ever for fiscal abuse of the Fed. For movements, it gives power to movements like People's Quantitative Easing and that sort of thing, where the Fed is seen as an alternative to the ordinary fiscal appropriations process. The Fed's balance sheet is seen as a device that can fund projects independently of congressional funding through ordinary channels.

And I think it's especially naive now, in light of all that's happened, to assume that such proposals will remain on the policy fringe where, I think it is safe to say, they were for some time. Indeed, Congress has already shown its willingness to treat the Fed's balance sheet as a source of backdoor funding. It did that when it passed the Fast Act back in, I think it was 2015. Pardon me if I'm off by a year or so. That's the act that was designed to fund transportation infrastructure improvement. It was funded by expropriating what was then the Fed's surplus capital, which is now a minuscule amount.

Of course, central banks don't need a lot of capital. They can still function without it, though not having much is dangerous. But the point is we have a precedent for treating the Fed as a piggy bank. But that precedent was very limited. The central bank only has so much capital to expropriate. What the new floor system does, what the Great Divorce does, is to create vast opportunities for use of the Fed's balance sheet, not by appropriating its capital, but by taking advantage of its power to expand its liabilities.

So, what should we do about this? Well, I think there's really only one very basic solution. I've considered a number of other solutions in my book, but the most reliable way to deprive the Fed and Congress, I should say, to deprive the Fed power to accommodate Congress's requests for off-balance sheet funding using quantitative easing, even in non-crisis times, while still allowing the Fed to deal with crises, even if that involves some fiscal activity, is to replace the Fed's present floor operating system with a proper corridor operating system of the sort used by many central banks, including by the way, the Bank of Canada, to which Peter referred in his discussion.

In a corridor operating system, you can still have interest on reserves. You don't have to abolish that. But the crucial difference is that the rate of interest on reserves is set typically below the policy rate, not during crisis when you're at the effect of lower bound. In that case, they're the same. In that case, you're in a floor system no matter whether you want to be or not. But, at other times, when the policy rate is above zero, in a corridor system, the interest rate on reserves is somewhere below that. Therefore, you're back in an arrangement where, although it reserves an interest, there is an opportunity cost to banks of accumulating reserves. And therefore, you're back to a world where marginal additions to the quantity of reserves has happened, when the central bank is buying assets, will lower interest rates, the short term policy rate, and will lead to more inflation, other things equal.

The most reliable way to deprive the Fed and Congress, I should say, to deprive the Fed power to accommodate Congress's requests for off-balance sheet funding using quantitative easing, even in non-crisis times, while still allowing the Fed to deal with crises, even if that involves some fiscal activity, is to replace the Fed's present floor operating system with a proper corridor operating system.

Now, of course, there are several questions that should arise in connection with this proposal. I'd like to just anticipate some of them. First, there are some purported advantages to a floor system. And the question is whether sacrificing those would be a problem. Well, two advantages have been ballyhooed most by Fed officials in favor of this system. The first one, which they don't talk about much anymore, is that it would simplify the business of regulating a monetary policy, specifically regulating interest rates. You set the IOER rate in a floor system. And, in theory, you're done. That'll be the policy rate. And that's the rate you're going to see overnight in overnight markets.

Well, for those of you who have been paying any attention at all to what's happened since 2008, you know that it hasn't worked out that way at all. The Fed has had to come up with all kinds of Rube Goldberg contraptions by which to try to make the system work the way it was supposed to in the first place. And the bottom line is it's not simpler to operate. There have been more interventions. It's easier to have a proper corridor system. So, that advantage is no advantage at all. The other advantage that's claimed for a floor system is that it means banks have more liquidity, which should make them safer.

The first thing to be said about that, and I mean, they have more liquidity because they have all these extra excess reserves they're willing to hold. This is really a very doubtful advantage, first of all, because it's redundant. Today, unlike in the past, we have the liquidity coverage ratio. We also, in the United States, have ordinary bank reserve requirements. Well, you don't need a suspender and a belt, let alone a suspender and two belts. So, how many provisions for keeping banks liquid do we need?

More importantly, there is a compromise solution that can give us the advantages of a floor system for ruling out People's QE and that sort of thing, while still maintaining high levels of reserves in the banking system. And this is a so-called tiered reserve system. Several countries have them now. And that's where you have at least two layers of reserves or excess reserves in the system. One layer earns an interest rate that may be above or at the policy rate that's desired, as in a floor system. But then, above that amount, a fixed amount, the interest earned on reserves can be below the policy rate. It could even be negative, that is at a negative rate below the desired policy rate.

In this way, you can have as much liquidity in the system, as you want by having a set amount of reserves that pay more than the overnight policy objective rate. But then, at the margin additions to reserves, that is growth in the balance sheet, has the usual effect of leading to a marginal lowering of overnight rates and to consequent easing of money and higher inflation. Which means you are once again back in a world where a central bank that importunes to expand its balance sheet for strictly fiscal purposes, because people think that's a good way to finance things where you don't have to mess around with pesky appropriations processes and other kinds of obnoxious democratic controls. The Fed can say, no, I'm sorry. We can't do that because we can't achieve our mandate if we do. That's it. Thanks very much.

Guida: Thank you. So, we have a ton of great questions coming in. So, I'm just going to ask one question and then start turning to audience Q&A. But I did have one question following up on what you were just talking about, George. Obviously, that the Fed's balance sheet is ballooning a lot right now. And, at some point we're going to reach a point where people start calling for it to shrink again. So, given that the Fed has said that it's going to stick to a floor system, was there any purpose last time in the Fed shrinking its balance sheet to the extent that it did, given that it kind of just ended in repo chaos? Did it serve any kind of political signal? Or, if you're doing a floor system, does it even really serve any purpose to shrink it a little bit?

Selgin: I think it does serve a purpose. Of course, a floor system requires more excess reserves than a corridor system does. But what the Fed was trying to do then was to determine just how few excess reserves it could have and still be running a floor system. And it found out the hard way that it needed more than it thought it would.

But why would it even bother trying to get as low as possible? I think the answer is that it's desirable for the Fed to have a small credit footprint, as small as possible. And this was long a principle that the Federal Reserve operated on, that they should be as uninvolved in credit markets as possible. They should be a lever for other types of intermediation. They should control them, but they should not have a huge credit footprint. And so, what they were trying to do was minimize that footprint and keep private markets doing as much of the credit intermediation as possible.

So, what the Fed has learned is that, if you have to operate a floor system, if you want to operate it, that footprint turns out to have to be very, very large. Now, just how unfortunate you think that is, is a question we can debate.

But, obviously, one thing can be said, that the Fed seems to have decided at some point that, no matter how big the footprint is, a floor system is still ideal. In fact, if we've learned anything since 2008, it's that the Fed doesn't seem to think anything will prove that its decision to have a floor system was a bad one. I'd love to know what they would consider a reason for going back because every argument they ever gave for having such a system has proven to be false. So, I can't really answer for them. If I were Jay Powell and I were asked to defend a floor system, I would come up dry. I really would. Except for being able to argue that getting to or corridor system, to a floor system is tricky, which it certainly is.

Guida: Yeah. That's really interesting. So, I'm going to turn now to some of the questions that are coming in. Here's one for Sebastián. Based on your work, what is the typical journey into fiscal dominance? And are we in danger of it happening here in the United States?

Edwards: Well, that's a great question. I think it's at the very core of our discussion today. I don't think that we are in danger of getting into a situation of serious fiscal dominance. But I think that we are slowly moving in that direction. And one of the things that has surprised me a little bit in this series of seminars is that we haven't spent a lot of time debunking or talking about modern monetary theory, which is of course at the core of fiscal dominance in advanced countries. And I think that that is the danger that we're seeing. The progressives of course love it. And there is, in my view, a very important difference between helicopter money, which is the once and for all type of policy that you undertake during a crisis and a modern monetary theory type of approach, which is a recurrent ongoing fiscal dominance. Which as I pointed out, once you bring in changes in the laws it can really result in a jump in inflation. So, no, I don't think that we are in danger right now in the US but in terms of the thinking among politicians slowly, slowly, slowly people, or the Democratic Party, which at some point was very prudent in terms of monetary policy, is beginning to like MMT. And I think that we have to be very clear in pointing out all the dangers involved.

Guida: Thanks. And I think Peter also has some thoughts on that question.

Stella: The typical way you fall into fiscal dominance, in my experience, it's basically been when the central bank has intervened in a financial crisis in a massive way. Usually that's at the time of a fiscal crisis. So the fiscal authorities can't afford the bank rescue and the central bank creates a lot of money and takes on a lot of collateral that turns out to be bad and worthless. So I've known central banks that have wound up owning pig farms, hotels, resorts, even television stations. And my point on this is what happens is it's a massive intervention in a big hurry. The central bank acquires a lot of extremely illiquid assets, by definition your assets that are basically based on bad loans. So the banks would go bankrupt and you can't liquidate the collateral. So you wind up sitting on this for a long, long time.

So that's the excuse, let us say, for not shrinking the balance sheet. For those countries in Latin America and in Asia, took quite a long time, maybe 20 years, to get themselves out of it. Now, the Fed doesn't really have that problem because most of the assets the Fed has taken are pretty solid and liquid. So I really don't think there's a very good excuse for the Fed not to have shrunk the balance sheet faster and that it couldn't shrink the balance sheet again. But based on revealed preference, it doesn't seem like they will do it. But I just want to point out that they don't have the excuse that many, many, many other countries had that the assets they had acquired were liquid and it was going to take a really long time to dispose of them. Thanks.

Guida: Thanks. It's hard to imagine the Fed buying a TV station, but maybe we'll get there. So here's one for Peter. Can you explain in more detail how to move assets off the Fed's balance sheet onto the Treasury's?

Stella: Sure. By the way, it didn't buy a television station it took it as collateral from some large investor and they wound up owning it when the investor went bankrupt. So how do you take assets off the Fed balance sheet? Well, you buy them just like anyone else. So basically the Treasury or the Market Intervention Authority would issue debt, acquire cash and buy them from the Fed. So basically that would shrink the Fed's balance sheet by the amount of the bank reserves and then by the assets that you have. So this has been done in different countries in different ways.

It's quite common after one of these financial crisis that I've spoken about for the Treasury, eventually, to do a bond swap with the central bank, sometimes directly issuing the bonds to the central bank and taking off the bad assets and then liquidating them. Unfortunately though, many times the Treasury gives the central bank 100 year bonds paying 0.1% interest, which doesn't really help the central bank very much. There was a very big such bond in Indonesia and it was called State Bond number 007, and they called it the James Bond. Thanks.

Guida: All right. Here's another one, which it's a question for anyone but I'll go ahead and to direct it to George. Should Congress give the Fed the authority to purchase any kind of financial asset during a crisis or not?

Selgin: Thanks. Well, I think that it is not wise for Congress to give the Fed power to buy just any old thing. The question is whether there are going to be occasions when the Fed has to buy certain assets or rather when Congress thinks certain assets need to be bought to save the economy from a crisis or something like that. And in that case, there are really two questions. One is, should anybody buy those assets? Is that going to help? And the other is whether it should be the Fed's responsibility or Congress's responsibility through some fiscal authority, which of course is a question we've been addressing. My own feeling is that if Congress decides that such assets and anything besides Treasury securities and agencies have to be purchased in large amounts for emergency reasons, then I think it should be done on the fiscal policy books.

I think if at most the Fed should be involved as part of a committee, part of a special emergency fiscal facility or the sort of facility that Peter Stella described so that there are very strict limitations both on when the facility can be used, on the decisions regarding how much it purchases, on how much the Fed is itself on the hook versus how much is pre-funded by the Treasury and how quickly the facility will be unwound. I think all of this has to be very clearly predetermined so that we both know what to expect in a major crisis and we can avoid a lot of ‘hanky-panky’ in a major crisis and not end up as some of the countries that Sebastián was talking about and that Peter has dealt with so often, have ended up in the past.

Guida: This next one is for Sebastián. Can you elaborate on your suggestion of moving the Fed Board of Governors to an agency type organization?

Edwards: Yes. Agency type organizations, of course, a particular political party cannot have more than, I think, five members in the governance body, more than three members from one political party. And there is the notion that if we have a particular party that has a run at the White House that is 16 years, given that members of the board don't stay for their lengthy term, whether we can have a very large majority by one political party. So the question of a political equilibrium is something that has come up a number of times. And I think that there are countries, some of the ones that Peter mentioned in his presentation that at least implicitly have taken that view where members of the board of the central bank have to have some kind of equally room with respect to the political forces in the country. And I think that it's something to explore.

When I read back, again, two days ago, the Reconstruction Finance Corporation Act of 1932, one of the things that caught my attention is that it was operating as an agency. And even during FDR's time when Jesse Jones was the chair of the RFC, it had two members of the Republican Party on the board and that they made difficult all the quite controversial decisions and policies that the RFC put in place, including the fact that it bought gold with debentures of the own RFC that it gave it at a huge discount to gold producers in order to artificially move the exchange rate above the $20 and 67 cents price of gold at the time. So I think it's something to explore in terms of governance.

Guida: Yeah. And then there's also a question one of Peter's ideas. It says question for Peter, do you think moral hazard is a problem if there's constantly a reserved fund for handling emergencies?

Stella: So basically that's where I came up with the MIA. I shouldn't say MIA. The Financial Market Intervention Authority. You can draw a line here, monetary policy, fiscal policy, but what do you do with this stuff in the middle? And one can pretend that, well, we could just say we're not going to do it, but I think revealed preference shows that we are going to do that. So if there's moral hazard, there is moral hazard. So I'm not saying that we should do it, I'm just saying if we are going to do it, this is a better way to do it. And in fact, I actually believe the current seeming obsession with eliminating all volatility in financial markets is a big mistake. I would allow more volatility. I wouldn't be intervening all the time.

It's a bit like not ever allowing yourself to be exposed to any illness or take a vaccination. You need a bit of a live vaccine to develop defenses against it. If you take all the volatility out of the system, the market will stop hedging, will stop developing instruments to deal with volatility and then you're just going to have one really, really big crisis when it happens. So again, I'm not saying we should be doing this, I'm just saying, I think we're going to be doing it. This is a better way to do it in my opinion. And I agree with Sebastián about the agency structure. For this entity, I think that's a great idea. But I really think the fetish of market volatility has to be addressed and discussed in the legislation for this entity. ‘When do we intervene?’ as George was asking.

Guida: Great. So we're almost out of time here. Before we wrap up I just want to give Sebastián and other minutes to talk about fiscal rules as anchors.

Edwards: Thanks, Victoria. So I was not planning to talk much about Latin America, but Peter brought up a number of cases which are very interesting and I happen to know a little bit about some of them. So let me very briefly tell you what Chile did after its very traumatic experience of 1000% inflation in 1973, 1974. An inflation that, as I point out in my article on populism in the Cato Journal, came from fiscal dominance. The overall fiscal deficit went up to almost 30% of GDP in 1973. And inflation went up to 1000. So Chile forbade the central bank from a financing the government. That's sort of standard in many central banks, but also forbade the central bank from purchasing government paper even in the secondary market. And what the central bank did, then it issued its own paper and monetary policy and open market operations were carried on the basis of central bank paper which had been issued.

Well, what happened during the coronavirus? It ran out of its own paper. And now there is a constitutional amendment being discussed in the Chilean Congress. It will allow the central bank to purchase Treasury paper only in the secondary market but with a super majority of the members of the board, four out of five, and there is even talk of five out of five. So it would really make it an emergency situation. In addition to this, and this is the point that I want to make is that Chile has a fiscal rule. And the idea although it's being discussed how to change it is that over the cycle, the primary balance of the Treasury has to be such that a stable debt to GDP ratio is maintained over the cycle.

It's not extremely rigid, but it has the distinction in structural deficit, the current deficit, and then they have to be corrections. And I think that this is something that needs to be thought if we want to maintain the separation between the Treasury and the Fed going forward. You need an anchor and a fiscal rule is a way of providing an anchor. And that may be that George, that we have suspenders and a belt, but maybe that's the way to go if you don't want to be caught with your pants down. Thank you.

Guida: Sorry. I almost forgot to unmute myself. So that should wrap our event for today. Thanks to everyone for joining. We had a lot of really great questions. Sorry I couldn't get to all of them. We had a lot of concrete ideas, certainly you all given me plenty to think about. And with that I'll turn it over to David.

Closing Remarks

David Beckworth: That was a great panel and a perfect way to round out the conference. Thank you to Victoria for doing a great job moderating this final panel, and thank you to the participants, Peter, Sebastián and George, for thoughtful discussions and comments. Also, a big thank you to all the viewers who took part in participating in the Q and A and who have been with us these past two weeks. This conference would not have been a success without you. So thank you for participating. I also want to give a big shout out to all the events staff, the organizers, IT people at Cato and at the Mercatus Center who have played a big role in bringing this conference to fruition.

We really appreciate all the hard work behind the scenes. So thank you everyone who has helped out. Finally, I want to encourage you to continue this conversation we've started at this conference. This is an important conversation. We want the Fed to be ready, robust, but also constrained and prepared for the next crisis. And the conversation we've started here today is an important one that will continue. We're hoping that it leads to great improvements moving forward. Until we meet again, take care.