Kris Mitchener on What Actually Anchors the Price Level

Kris Mitchener is a professor of economics at Santa Clara University and is an economic and monetary historian. In Kris’s first appearance on the show, he discusses how he fell in love with building data sets out of old dusty archives, the origins and fall of bimetallism, the pros and cons of the gold standard, the problem of operating losses on the Fed’s balance sheet, what truly anchors the price level, and much more.  

Subscribe to David's new Substack: Macroeconomic Policy Nexus

Read the full episode transcript:

This episode was recorded on March 4th, 2026

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

David Beckworth: Welcome to Macro Musings, where each week we pull back the curtain and take a closer look at the most important macroeconomic issues of the past, present, and future. I am your host, David Beckworth, a senior research fellow with the Mercatus Center at George Mason University, and I’m glad you decided to join us. 

Our guest today is Kris Mitchener. Kris is a professor of economics at Santa Clara University and is an economic and monetary historian. He joins us today to help us think about what determines the price level. Kris, welcome to the show.

Kris Mitchener: Thank you so much. It’s a pleasure to be here. Longtime listener and happy to finally join you on your show. Thanks for the opportunity.

Beckworth: It’s great to have you on. As I mentioned, you’re an economic and monetary economist and historian. You’ve done a lot of interesting work on the Great Depression. You’ve done work on bimetallism, the silver standard. These are all things we want to touch on briefly today. You also had a paper that was really fascinating about central bank operating losses. Listeners and watchers of the show know I’ve spent a lot of time thinking about these losses recently, not just at the Fed, but at other central banks.

What I love about this paper we’re going to discuss later today, you actually have a look back at 10 advanced economy central banks, and you show they also went through a period of raising their rates and what happened to their balance sheets during this time to help shed light on what happened more recently here. 

Kris’ Career Path

Before we get into all of that, tell us a little bit more about yourself. How did you get into this career path of being an economist and focusing on monetary and economic history?

Mitchener: I think it probably started with a passion for interdisciplinary approaches to the social sciences. As an undergraduate, I liked political science, I liked history, I liked economics, and I found the combination of history and economics super powerful for giving us insights into behavior, both at the micro and macro level. I think once then I found that I enjoyed sleuthing around archives and pulling out musty old government documents and constructing datasets—something that maybe will be increasingly a thing of the past with modern technologies, we’ll have more efficient ways of doing it. I like that detective story approach to things.

I also think a lot of topics then, once you get more focused into understanding theory, what really fascinates me both about history and economics is how crisis periods leave such a deep impression on both our institutions and our economy. I was drawn to those episodes, and those, by their nature, we hope at least, are rare events. Maybe we could argue they’re more common increasingly. In the last 30 to 40 years, we’ve seen more financial crises. Nevertheless, we still think of these as rare events, which then makes you shift your lens to longer-run perspectives.

One part of me said it was the interest in understanding these inflection points in society, and another side of me was probably driven by folks like Robert Lucas, who said, paraphrasing, “Once you start thinking about growth, you can’t stop thinking about growth.” Some of my early research was actually on thinking through institutions and growth. I think that’s how I ended up in this field.

Beckworth: Very interesting. You mentioned how important it is to look at a broader historical timeline to really get a rich perspective. I remember during the Great Financial Crisis, Niall Ferguson, the historian, mentioned that many people who were working on Wall Street at the time were young, they were in their 20s, and they had no perspective. For them, it was the Great Moderation. Everything was stable, everything was normal. To them, they had no concept, for example, of a national housing crisis. Last time that happened was during the Great Depression.

It is important that we look back to history, and people like you help us see the bigger picture and lessons from the past. We had Ben Bernanke leading the Fed, and fortunately, he had looked at the Great Depression and some of the lessons. Who’s to say another version of history could have been one as severe as the Great Depression had we not had someone who had a historical perspective like he did? Again, that’s why it’s useful to speak with folks like you.

Mitchener: I think that’s right. I think his perspective was very useful. We can always argue about the specifics, but I think having that perspective that we didn’t want to repeat the Great Depression was no doubt important. He’s the one who ultimately issued the mea culpa for the Fed after he had been serving as the Fed chair. He issues this mea culpa that they got things wrong and then apologized. It’s not for his period, but more so for historically how the Fed made these famous sins of omission and commission during the Great Depression, as Friedman famously called them.

Beckworth: I want to be fair to Bernanke, to the Fed, and just to remind our listeners and watchers that it’s easy to criticize the Fed, no doubt about it, they make mistakes, but it’s also easy to forget the true counterfactuals. What would have happened had they not made the choices that they did? In my mind, it’s not inconceivable we could have had another Great Depression had they [not] stepped in. Yes, there’s a lot of intervention, a lot of things that I may not be comfortable with, but in the heat of battle, decisions are made. Now, the job of folks like you is to help us look back and assess, “Okay, this is how we thread the needle better next time that you step in.

Mitchener: I think you make an important point there, not to digress from our broader topics, but some of my research is really about those kind of real-time reactions and how short those windows are. If you’re trying to make judgments about saving banks or not saving banks, your policy windows are quite short. Some of my research, again, in different contexts, going back to the Depression, really shows how quickly those deposit outflows take place.

If you don’t have some sort of response, whatever your lender-of-last-resort optimal response would be, if you don’t have some sort of response to stop those outflows when there are runs taking place in the financial system, that can really cascade and spill over to other financial institutions and really create an economic calamity, starting and emerging from the banking system and spiraling out further into the real economy.

What Is Bimetallism?

Beckworth: Yes. Let’s tap into this knowledge, this history that you’ve built up and you’ve published. I want to go back and use it, both your work on the historical past as well as your more current work, to really get at a fundamental question that I think I often take for granted and I think we do on this podcast, and that is what ultimately determines or drives the price level. It’s easy, I think, if you get in your own camp to say, oh, it’s monetary aggregates if you’re a monetarist, if you’re maybe a fiscal theory of the price level person, it’s fiscal policy or it’s the expected future primary deficits. Maybe if you’re a new Keynesian, it’s output gaps and Phillips curves.

I think history really provides an interesting way to look at this because we can look at monetary standards like the gold standard, the silver standard, bimetallism, and you’ve written on that. I want to go there, and then I want to tie that in into your current research on central bank operating systems and balance sheets, and what happened to the losses we saw at major central banks recently compared to the 1980s. Do balance sheets matter for the price level?

We’re going to look at these different pieces of the puzzle, put them together today. Even though there’ll be very different historical periods, I think we can paint a picture, bring together, have that puzzle complete, and help us make sense of what really drives the price level. Again, is it the monetary standard? Is it fiscal backing? Is it money? What else is going on here? Let’s begin by going back to a topic that I don’t think we’ve ever talked about on the show other than in the passing, and that is bimetallism. Help us understand what is bimetallism, when was it used, and why did it even emerge. Why did we have bimetallism historically?

Mitchener: The macro teacher in me would say, we know for good reason that money has many uses. It has use functions. We saw these metallic standards. When we think of bimetallism, what we have in mind is a system where you could just be using silver and gold themselves, minting those and having those circulate, or you could be using paper currency with some sort of backing rule backed by silver and gold. Bimetallism is going to say the amount of money circulating is going to be determined by that ratio between gold and silver.

These systems essentially emerge out of convenience. We want some sort of specie that’s going to be convenient and durable. We want to use it as a unit of account, as a medium of exchange, as a store of value. As other social scientists have written, money is a social construct. We choose things of value, and silver and gold were precious metals that had value. They had durability, and they’re fungible. You can melt them down and change them. Of course, the issues came up over centuries. Could they be counterfeited? How should we set that ratio famously?

Isaac Newton was the Master of the Mint, gets the ratio wrong. England tips off of a bimetallic standard onto a gold standard, and that starts a chain of events we’ll talk about in just a moment. It starts with this idea that you need something useful for money. In most societies, be it cowrie shells, be it the giant stones that they use in the Yap Islands, whatever it is, there was something of cultural significance that they could designate as money, even if they then used some quasi-equivalent to blockchain in the case of the Yap Island currencies. They’re not moving those around. They’re keeping some kind of unit of account system.

The predominant forms of something precious that began to circulate are gold and silver. They had different intrinsic values to them, and so they naturally allowed monetary authorities, governments, or private issuers to rank the value of silver relative to gold. Of course, you would have a market price for these things where people could just transact in these based, again, on just the relative scarcities. Then you could have a mint that actually minted coins.

The tension when you ran these metallic systems was always between what they called the mint ratio and the mint price versus the market prices. The challenge was to, of course, keep those things relatively close, that the Master of the Mint had to keep the price relatively close to that market price. Otherwise, if those things deviate, you’re going to get an arbitrage opportunity. People, if you set it too high, they’re going to start melting down or taking in their metals and converting them and getting gold out of that, and then taking it to the market.

The market pins down what the monetary authority can actually set as the price. Then the hope is that then whatever is issued then is pinned down, and that, in fact, is going to pin down the price level. The easiest way to think about, since you want to cast it in terms of the price level, imagine a simple world where we didn’t have bimetallism. Just imagine just one of these, silver or gold. You issue it, and you issue it in a fixed quantity, and so there’s no backing rule. You’re just circulating the coins themselves. That is your money.

In a monetarist interpretation, if we just think of a quantity equation, we would have the money on the one side and the price level on the other and your GDP being determined by long-run factors like labor force and capital and technology. Assuming velocity is constant, you’ve got this nice relationship. You put more of these coins in circulation, it’s going to affect the price level. Then what the system did was the markets pinned down the ability of the monetary authority to issue more.

Even once you have a backing rule and you issue notes, you could only issue notes up to whatever your backing rule was. Without a backing rule, it just meant whatever gold you have, you mint into coins. You can also see how mercantilism, the early form of trade policy, came about. This is a function of, I need more coins to trade more. That’s how those notions of early trade policy also emerged. They go hand in hand in some sense.

The 19th century really represents the watershed for both those. I think you had a previous colleague of mine on a previous podcast who was talking about the emergence of the global economy. Of course, Chris Meissner would have talked about the importance of the 19th century and this first era of what economic historians would call globalization, the first era of globalization. This is changing trade relationships. As I’m sure he talked about, early trade agreements between the French and English set the table for free trade.

With that comes the need to have currencies that are going to be global. What emerges in the 19th century, over time, is this slow shift, again, with Isaac Newton and the Master of the Mint, initially by accident, that England goes on to gold, and it’s the first one to tip a big superpower, trade and geopolitical superpower of the 19th century. It tips onto gold accidentally, and then it starts pulling its most important trade partners onto gold through these kind of network effects, these network externalities.

This transition that you wanted to ask me about, this transition from bimetallism to gold, gets underway in the second half of the 19th century. To some extent, bimetallism received a lot of attention from monetary theorists because when they were writing, that was an interesting way of thinking about how prices relate to this kind of complicated system. The gold system, I think, for many reasons, actually, turned out to be a better system to support this whole new apparatus of global trade and finance that was emerging, because in some sense, it was simpler. You don’t have to keep track of two backings, especially two market prices. You just keep track of one.

The Gold Standard

Beckworth: Going back to what you mentioned earlier about Isaac Newton setting the ratio at the mint between gold and silver, and he did that in a manner that led to gold being the medium of exchange as the form of money, that was somewhat of an accident. He didn’t mean for that. That wasn’t the intention.

Then, of course, Great Britain is this global power, and other countries follow along. Not entirely an accident, but some part of this was some path dependence set up early. Then later, it becomes norms and countries want to follow and be civilized; nations want to go on the gold standard. Then there’s bimetallism in the US for some time, but eventually, the US settles on gold by the end of the 1800s. Is that correct?

Mitchener: Yes, that’s correct. Again, harkening back to our earlier discussion, there’s a strong coalition among farmers and among miners in the West that really don’t want to go to the gold standard. Even as the US comes off of its experiment with greenbacks and issuing unbacked notes from the Civil War, and it’s called resumption in 1879, the Silverites are still quite strong in politics. They have vested regional groups that are trying to normalize the system that prevailed prior to the Civil War.

They’re able to win some Pyrrhic victories, but I think their last stand was really Bryan in the election of 1896. By that point, it was pretty clear, as it was in other places around the world, such as India, which began to demonetize silver and other holdouts. Throughout the, I’d say, 1870s and 1880s, you see a bunch of countries shift from bimetallism to gold. Some of these are accidents. I know you were asking about that. We could think kind of accidents.

For example, how does Japan get its gold to go onto the gold standard? They fight a war, and at the last minute, the war indemnity, essentially, the victory money they get, they demand it to be paid in gold. This is very similar to how Germany gets the money to go onto gold. They demand it in their war, in the Franco-Prussian War. You could call those accidents or negotiations, and then, “Oh, we want payment in gold,” and now you have enough gold to back your currency, to back your note issuance.

You could call it policy plus accident, or something like that. The other thing going on is that scholars have talked about, and I think this is most important for countries at the core of the system, that there was sort of a “good housekeeping seal of approval.” This is the term Michael Bordo and Hugh Rockoff gave it, that at the center of the system, these economies, if you joined up to the gold, you got to borrow at a lower rate.

From a sovereign borrowing perspective, maybe there was a benefit to joining gold. What was that all about? I think that’ll anchor us back on our discussion about price anchors because their argument was essentially about sending a signal to markets that you’re going to have fiscal probity, and you’re going to have fiscal probity in part because you’re going to have monetary probity through the fact that you’re going to commit to a rule.

All these things we’re going to talk about today, whether it’s a Friedman rule or whether it’s inflation targeting, these are all types of rules. These rules-based systems really begin with bimetallism, hence Milton Friedman’s interest in this. Then they evolve into the gold standard, which has rules. By accident or by design, countries climb onto this gold standard, and some of them then do benefit and see a lower cost of borrowing when they join. This is interpreted by those authors as a signal that they’re going to commit to monetary policies that are going to lead to not inflating away the bonds you’re issuing.

Now, the tricky part of this, and just to digress a bit, is that really only held true at the core. Some work I did said, “Let’s look at emerging market countries.” I will just say, “Whereas some people like to put a lot of emphasis on, maybe we should go back to the gold standard, look at all its stability.” You’re not going to find me in that camp because if you’re in an emerging market economy, it was very difficult to stay on gold.

You can see this because we collected literally millions of observations of bond data. We collected paper bonds and gold bonds. Essentially, you can back out what economists call currency risk. It should be that differential. If the gold standard is completely credible, then your paper bond should be as good as your gold bond, and so there should be no premium in the market for holding a paper bond.

What we actually find is that even after gold standard adoption, these risk premiums were quite large, meaning markets never fully thought that these commitments to gold were credible. That’s in part because they turned out not to be credible. Many of these countries abandoned gold because it was much more difficult to be on the periphery. Then there’s a large literature on this. You weren’t the conductor of the international orchestra, as Keynes called it.

That was happening at the center of the system so that when a country like England or France changed its rates, in order to stay on the gold standard, other countries would have to change their discount rates, those were the policy rates at the time, in order to prevent gold from flowing across borders and from you being forced to leave your gold standard, meaning that’s going to transmit stocks from the center to the periphery that might be asymmetric or unaligned with your own business cycle. 

Beckworth: That credibility that you bring up, I want to use it to segue into this bigger point. The markets back then had questions about the periphery. Could they credibly commit to the gold standard? They were saying no. They don’t have the fiscal capacity, the state capacity to make those tough choices to raise rates when Great Britain would raise their rates. I guess this is the question I want to ask you, tying together all the threads from this discussion we’ve been having here. Was it the metallic standard, whether it’s bimetallism or gold, that really anchored the price level, or was it the fiscal commitment behind the gold standard or the commodity standard?

In other words, during the Civil War, the US goes off the gold standard, and then it commits this journey back, as you mentioned, to 1879, this journey back. It intentionally undertakes deflation to get back to a place where it can go back on a sustainable basis, the gold standard. It was a signal, it was believed as credible, but not every country could do that. I guess my question is, is it really the gold standard, or is it a deeper institutional commitment to price stability?

Mitchener: That’s a great point. It’s going to have to be both. I think there’s a third dimension that I would encourage listeners to think about, which is the policy environment, and also the conceptualization of what we knew about the economy at that point in time. To frame those, let’s think about modern international economic policy.

I think that the way to think about this is to use Obstfeld and Taylor’s framework of international policy trilemma. It’s often called the unholy trinity because you can’t have these three points of a triangle simultaneously, which are pegged exchange rates, fixed exchange rates—that’s your gold standard, that’s your bimetallism—you can’t have that combined with capital mobility and what you would call independent decision-making monetary policy or autonomous monetary policy.

In the case of the gold standard, what are you sacrificing? This was an era of capital mobility, high capital mobility, very low capital control, so they didn’t sacrifice that. It was an era of pegged exchange rates, so they didn’t sacrifice that. What they sacrificed was the ability to make independent monetary decisions. What’s the other defining characteristic of the periphery that even impacted core countries in, say, the early 1890s or the early 1870s? That is financial crises.

I think we have a tendency as monetary economists to overemphasize monetary policy. I see this all the time in policy shops like central banks, but in their state admissions, and many of them, they also are charged with financial stability. There was a lot of financial instability on the periphery in this period. These choices become very challenging in part because what do you do if you have a banking crisis? Can you act as a lender of last resort? Are you going to create an institution as a modern central bank that actually does that?

This is an era where they’re sorting through that. We’re going to see a crisis in England called the Overend Gurney crisis, which is really going to define the Bank of England and have it rethink its position as should it act as a lender of last resort. We’re going to see other European central banks begin to think about this role as a lender of last resort. Some of them don’t even really get this resolved—and maybe we’ll have time to talk about this—up until the 1930s, because it actually is going to feed back into this discussion we’re going to have on central bank losses.

There’s a very interesting episode of the French case, which I’ve been working on now for about five years with Éric Monnet, a French scholar. He and I have been researching this fascinating episode of losses and lender-of-last-resort behavior. To keep it on target, there are tradeoffs. Whatever tradeoff you want to take, the tradeoff of having pegged rates and capital mobility was you can’t use your autonomous monetary policy either as a lender of last resort or maybe just to stimulate your economy.

That’s where the macroeconomic models and conceptualization come into play because if you don’t view unemployment as a social concern, if you don’t view it as something that policymakers should be concerned with, you view it as a personal shortcoming as they did in the 19th century, then it’s easy to focus with laser-like precision on what we call external balance, maintaining balance of payments stability. That’s what the gold standard was really good at.

If you just say, put zero weight, so it’s a different kind of policy function—you were talking about new Keynesian models with weights on unemployment or the output gap and inflation. There, the weight is like what kind of weight did we put on internal balance versus external balance? The weights in the 19th century were zero and one. You could do that because the franchise was limited.

To answer this shortly, your best example to get to your question, is it the standard itself that delivers this credibility? It’s not independent of politics. The reason it’s not independent of politics is because they try to put Humpty Dumpty back on the wall after World War I. They create this kind of system called the gold exchange standard. Prices have risen because of the war, people went off the gold standard, they have to fight a war. Some of that war is financed through bonds. A lot of it’s financed through just money creation.

They’re faced with this world where they don’t have enough gold, possibly, to go back on their pre-war gold-backing ratios, so they pyramid it. They say, “Countries like France and the US, in particular, which never went off gold during the war, they’ve got a lot of gold. We’ll just say, if we hold their currency, it’s as good as holding gold.” That’s how they take advantage of this scarce resource, the backing thing. They basically come up with this modern concept that a lot of EMEs use today, which is hold global currencies as your reserve currencies.

They do this whole system, but of course, the world’s changed. In order to get people to fight in the war, they made political promises, and those promises included the vote. Now you’re in England, and you’re thinking about monetary policy, and you can’t have that laser-like focus on external balance because now if you don’t think about high unemployment after World War I, which of course England has before in spades, it has its depression before the Great Depression, what do you do? Do you respond to that, or do you get run out of office?

I have this view and I think others hold this view, particularly Barry Eichengreen, that you could maintain this singular focus on price stability when the franchise was limited. When that changes, that changes the ability to have this anchor really work in the way you want it to. We can’t go back in time and just say, put the gold standard in place because we also have democracies where the franchise is much more extensive than it was historically. We might say these are great anchors, but in a very narrow sense, in a world where we didn’t have full macroeconomic models and understand how the spillovers to unemployment and incomes and production work. Those models compare.

Beckworth: Yes. If we did go back, we would probably get the interwar gold standard, which was a big hot mess as opposed to the classical gold standard, which worked relatively well compared to that. The key point, I guess, you’re making is, yes, it is ultimately credibility, but politics play into that credibility. What are you actually able to commit to doing? 

Let’s segue to the present, and you mentioned inflation targets. I often have this conversation with people. They’ll say, “Oh, the Fed liabilities are no longer backed by gold. If you return $1, you just get another $1.” I would argue there’s still an implicit backing. It’s the taxpayers. It’s the future taxpayers. There’s still real resources backing that, just like gold. You go off gold, you’ve got to have real resources to go back on gold or to support the gold standard. I would say fundamentally that it’s the same principle, even if we have a different monetary order or an approach.

Disinflation Policies and Central Bank Finances

I think this is a nice segue into your work on central bank balance sheets and operating losses because, as you note in this paper, and the paper’s title is “Do Disinflation Policies Ravage Central Bank Finances?” You’re speaking to what we just went through, a lot of operating losses in advanced economies. There’s been other work that has also shown that central banks can have balance sheets with negative equity, and guess what? They can still do their job.

Mitchener: Yes, they still do their job.

Beckworth: It doesn’t impair. Maybe at some point, I imagine you get enough negative equity, you’re in trouble, but I think this speaks to credibility and to the belief that there’s enough fiscal backing to make sure the central bank continues to do its job. Let’s talk about this paper, its motivation. We touched on it, but give us the motivation behind it, and what period does it cover?

Mitchener: The motivation behind it is pretty clear, as you pointed out. If we look around the world today, it’s not just one we could chalk up to weak institutions and underdeveloped economies that have problems with central banks that are profligate and poorly managed. It’s really a problem that has infected most advanced economies’ balance sheets in the early 2020s, first half of the 2020s. It’s an underappreciated area of research.

Monetary authorities would normally say, “Who cares?” We’ll get into the why you should care. There’s a good reason to say you shouldn’t care. Just to point out the facts first, many of these advanced economies were experiencing losses. In the case of the US, the losses amounted to about $192 billion for 2023 and 2024 combined. You can go to the ECB, it’s on the order of around $9 billion for those two years. Just to remind the listeners, that Banque de France, Bundesbank, these are subunits of the ECB, but they maintain their own balance sheets, and we still see losses there in those institutions.

As you mentioned, we collect data on 10 advanced economies, many of which are having these losses. These have gained, I think, attention in part because they’re occurring alongside fiscal policy, which we know, between the Global Financial Crisis and present, have been persistently running large fiscal deficits. When central banks are no longer remitting profits to state treasuries, that’s going to, in some cases, just a limited few cases, that might even drain reserves from those treasuries.

At a minimum, it’s giving policymakers maybe slightly, on the fiscal side, less fiscal space because they’re not getting those. That’s where we can really say it becomes policy relevant, because when you run these negative profits, these losses, people wake up and they start asking questions. That’s a real political economy issue more than anything else. I think the best way to segue into this is there’s a very interesting set of political economy issues. Then there’s the issue that we really set out to tackle, which is, are the losses themselves directly a result of disinflationary policies?

That was our motivation to say, we can see the balance sheets and we can see why the losses are occurring. Is this something about institutional evolution of a central bank, or is it something about the shocks themselves? The motivation for this paper was to say, let’s go back, not quite as far in time as we just discussed, more than a hundred years, but let’s just go back 40 years or something when I was a child and try to think through the great inflation period. Then because macroeconomists love to call everything a shock, the Volcker shock, the dramatic raise in the—

Beckworth: Named after a great central banker, Paul Volcker.

Mitchener: That’s right. Named after Paul Volcker, who comes in 1979, appointed by President Carter, and then is facing high inflation. It had been, I think, around just under 12% in ’79, and then peaks that are over 14% in 1980. This is true in England, in France, these countries, Italy, they have similarly double-digit inflation rates. Germany, Switzerland, economies like that, they have much higher inflation rates than they were historically having, but lower than these other sets of countries.

We have a mix of countries here, but this is a great period to go back to and say, well, we can separate out two things. We can try to figure out, is it the way that balance sheets have evolved that’s leading to these things, or is it the actual tightening process itself? The process of hiking up interest rates, in the US case, short-term rates peak at around 20%. Does that end up affecting the balance sheets themselves, or is it other factors on the way that monetary policy is being conducted today that lead to these losses?

We go back to that period in part because it was a big inflation shock. It was the last big inflation shock. It’s also better than earlier ones, like the ’20s or ’40s, in part because we know other things were going on, like price controls and factors that would make it difficult to identify these two channels. That was the motivation, just to answer your first question.

Beckworth: We are in a period that is similar to what we went through in the following sense that we have a surge of inflation. Now, in the more recent period, it was quick, it was sudden, but we had rapid inflation in ’21, ’22. The Fed had to respond aggressively. What you’re saying is, okay, let’s go to another period where something similar happened. We had a big inflation surge. Now, this took longer to build up, but Paul Volcker comes in and aggressively addresses it, just like the Fed aggressively addressed it under Chair Powell the past few years.

We got two Fed chairs. We got two inflation episodes. We got two rate hikes, aggressive rate hikes. Then, what does that tell us about the differences in operating losses and profits? Now, we know we had significant operating losses over the past few years. What happened to operating profits or losses during Volcker’s period?

Mitchener: Yes, thank you for starting with that, because one set of key takeaways is in some sense, an accounting exercise. You might say that accounting is about as exciting as watching paint dry on a wall.

Beckworth: Oh, but it is exciting.

Mitchener: This is interesting because, in fact, just the stylized facts, just the summary statistics show that our measures of profitability are things like—it’s not a firm, it’s a central bank—so we use measures like return on assets and then equity-to-asset ratio. The return on assets, pre-Volcker, it averaged across these 10 economies, advanced economies, about 2%. Then it goes up to about 3.4%. That’s return on assets. Your equity-to-asset ratio almost triples. It goes from around 6% to just under 18%.

That suggests right there maybe it has nothing to do with disinflation in and of itself that makes it. Maybe it’s the types of operations you have in place that are going to matter for this. Our approach to this was twofold. First, these balance sheets sound simple. Just pluck them off the internet and use them. Not so simple. Our study period is 1970 to 1990, these two decades.

The balance sheets aren’t harmonized. That harmonization started in 1990. Other studies that have worked with looking at central bank losses and profits begin in 1990 because the IMF started encouraging harmonization of central bank balance sheets. We had to do all that harmonization ourselves to come up with the right balance sheet category.

Contribution number one was just a pure data contribution, which is to say, let’s try to treat apples like apples and oranges like oranges, and figure out where do central banks get their revenues and what are their assets and what are the liabilities and how do we manage these. That’s contribution number one. Now, related to this, what we try to then do is say, in addition to this, not only did the equity asset ratio and return to assets go up, central banks actually made slightly higher transfers as a share of GDP back to the governments.

If you think about obligations like the Fed has a system where first they pay dividends out to the shareholders of the Fed system—we can talk about that if your listeners and viewers want to know more about that—they pay 6% or the highest rate on Treasuries depending on the last auction, depending on what kind of bank you are. Then they top off their surplus and then they remit the rest to the Treasury. What I’m saying is, those remittances to Treasuries, plural across these blanks, went up slightly. That part also went up.

Beckworth: That’s a good sign.

Mitchener: Everything seems to have gone in one direction. Then the question, the puzzle to us is, okay, let’s play counterfactual history. What if we take a modern-day central bank and give a bank, historically, the characteristics of a modern-day central bank? We do this on three margins, the three margins that your avid listeners would think about first. It’s like, okay, let’s go through these. The first is, okay, well, what about the way the Fed conducts monetary policy today? That was clearly different from the 1970s.

To remind our audience, back then we had a corridor system. The corridor system consisted of a reserve requirement, where essentially commercial banks are mandated to hold reserves, and then a policy rate like a discount rate. The Fed is going to be the outlier in our sample period. They’re the only one that’s really using open market operations. All these other central banks, they’re still primarily targeting the discount window rate. That discount rate is, again, the borrowed reserve rate that commercial banks can come and borrow from.

Today we’ve got the floor system. I’m sure you’ve had many podcasts and videos on this. The floor system means we’re paying interest to these banks. The first thing we’re going to do is we’re going to allow central banks in the ’70s and ’80s to have remunerated reserves. We’re going to put in a floor type of system. The other thing we’re going to do is we’re going to allow their balance sheets to be much larger. We’re going to let the reserves of the central bank grow. 

Then the third thing we’re going to do is allow the legacy of things like quantitative easing to affect the balance sheets. By there, what we have in mind from a balance sheet perspective is we’re going to change the maturity of the assets, the duration of the assets, as well as the rate of return. We know that the QE assets and GFC assets, Global Financial Crisis assets, were acquired in a period of very low interest rates with longer maturities on average.

Comparing maturities today, Fed’s portfolio has average maturity of about six years. Back then, it was about three years. Then importantly, the rates of return on these assets is going to look a lot different. The assets today average around 1% and are locked in for longer periods of time. Back then, shorter-term assets, so not locked in as long, and also higher rates of return. Starting in the range of 5% to 6%, and then when the Fed starts changing rates and the portfolio starts turning over, you’re getting returns on Treasuries like 9%, 10%.

Those are the margins we’re going to think through. These are the counterfactuals, the what-if scenarios we’re going to use to pin down what’s going on. Why is it that losses occur today and didn’t happen, given we have these two disinflationary episodes?

Beckworth: Once you do this counterfactual exercise, what is it that you find that explains why we have the losses today that we do?

Mitchener: It’s a triple threat. It takes all three. As it turns out, you can’t kill off the profits of the banks back in the late ’70s through the Volcker shock. You can’t kill it off just through reserve remuneration. We can take that parameter in our model, and we can raise that rate all the way up to, essentially, the ceiling rate, the discount rate, the average discount rate. You can reduce the ROA by about 1 percentage point off of an average of about 4.5% to 5% across our sample.

It’s not that it doesn’t matter, it matters, of course; it’s just not enough to kill it all off. Then you add in that balance sheet size. We actually double the balance sheet. Why do we double it? Because the Fed was at around 16%, and now it’s around 42% assets on its balance sheets coming from the reserves.

Beckworth: As a percent of GDP?

Mitchener: No, as a percent of its assets.

Beckworth: Of its assets. Okay.

Mitchener: Yes. Sorry, I should be more clear there. We’re thinking about the reserves. Remember, the Fed today is on the flat part of the demand function. It’s in an ample reserves environment, so it’s way out there. We’re thinking about, in a world in which we have ample reserves, as opposed to being on the vertical section of the demand curve, in that world out here, that’s about 42% of all assets, as opposed to back when we were on the vertical thing, we’re around 16% of all assets on the Fed’s balance sheet, not relative to GDP.

That’s essentially like a doubling on average, and we plug that in. That’ll cut it down. That’ll double the size of that, so it’ll take another percentage point off. To get it to go negative, you’ve really got to also give it this legacy of the GFC, of QE, and of the COVID, what we would call almost like a maturity mismatch on the central bank’s balance sheet. It looks like that.

You get this picture of central banks, if they’re sailing ships, they’ve got giant sails now. The sails are the size of their balance sheets. Then those sails can be very useful because those are the things that allow us to deal with these shocks that come along. Like, “Oh, we got to have emergency operations to deal with all these crises that hit our economy.” Then those sails, because they’re so large, if a big shift in wind comes, and that’s what I would like you to think of as a rate change, that sail’s going to push your ship in a new direction.

Living here in the Bay Area and running and riding my bike along the bay, I like to watch these sailboats and watch the wind push them. That’s essentially what you have when you have a bigger sail. In a sense, you get this legacy balance sheet because of all the action that’s gone on recently. That’s going to make you subject to these, as you pointed out, it’s a very rapid change in rates. That’s, essentially, in a nutshell, what we find, that it is really a legacy of balance sheets and it’s a legacy of an evolution of how the central banks around the world have moved to an ample reserves framework and to paying interest on reserves. That’s driving these losses. Not the distillation itself, per se.

Beckworth: Right. This raises several questions. One observation to start off with, before we get to the questions, this tells me that we can keep interest on reserves. Even if we went back to a corridor system, where you’d have your target rate in the middle and you had interest on reserves below that, you could still have a positive or negative, depending where you move your rates. Based on a Taylor rule, in pre-2008, that bottom was set at zero. Effectively, interest on reserves is zero. What you’re telling me is, even if it’s positive, you can still keep it and it should not lead to losses for the Fed in terms of operating.

Mitchener: I think from a policy debate—this is a really contentious area of policy and I don’t have work on this particular thing, so I’ll just state it. There’s a lot of debate on whether we should retreat to the corridor system. That’s complicated. You might have even had guests on to speak about this. One of the arguments you hear made is like, “Well, this is like an implicit subsidy to the banks. It’s a large subsidy.” There’s a paper that was presented at Brookings on this by Chris Hughes and co-author  Josh Younger. They come up with this calculation, but there was quite a bit of pushback on the panel from folks like Bill English, who would say like, “Well, look, we need to separate QE from the interest on reserves.”

Beckworth: Exactly.

Mitchener: One way we could approach this sensibly is to say, some of what we really can show very clearly is balance sheets are large partly because of QE. Part of it is also this paying interest on reserves distorts things. Those are two separate things. In a sense, we could remove the QE part of the balance sheet, is maybe a different way of answering your question, and still practice interest on reserves, with a much more constrained balance sheet than what we presently have.

Beckworth: Yes, that’s the point I’m making, is that I think it would be important to keep interest on reserves as a flexible lever that we can move up and down, even with the scarce reserve system, because most central banks around the world, whether it’s an ample reserve or scarce reserve or even a ceiling system, they have a ceiling and they got a floor and they’re moving those things around based on, again, like a Taylor rule. I would not want us to go back to a pre-2008 where we just don’t pay interest on reserves. It’s effectively set at zero. I’d want one that allows us to have a symmetric movement around some target.

Yes, the point you raised about the paper that was given at Brookings, I would also just note that if banks weren’t holding all these reserves, they would probably be holding Treasuries, which would be money funded going from the government to the banks as well. The only, I think, argument one can make is to the extent interest on reserves is higher than like a T-bill. There is some evidence, but it’s not much. It’s really a small drop in the bucket. 

Mitchener: You would think arbitrage conditions should keep that from happening. That’s the other argument that’s often made. It’s like, oh, this is distorting the rates somehow. We are losing market signals, to some extent, if banks—

Beckworth: Oh, no, I am sympathetic on that point.

Mitchener: —are under-utilizing the overnight lending purpose of it. Yes, we’re losing market signals there. I think that’s true, but I think also, on the other hand, you can say, “Well, Friedman would have said that you’re eliminating shoe leather costs, in a way.”

Beckworth: Yes, the Friedman rule.

Mitchener: The Friedman rule is eliminating the opportunity cost for banks to hold the most liquid-safe assets. Then this again spills back over into other objectives of the Fed. If you care about financial stability, then maybe what you’re doing here is you’re actually encouraging reserve management. Given that we have things like liquidity cover ratios and stuff, banks have to think about these things all in the stew of all regulations that they’re confronted with.

Thinking about these things in isolation, as we like to do in our models, can lead us to forget that there’s a variety of things that commercial banks are having to optimize around. I actually think there’s maybe some benefits, as Lorie Logan at the Dallas Fed has pointed out. There’s maybe some benefits here to having this system that encourages banks to think about what the optimal amount of liquidity you want to hold is, given you have these liquidity cover ratios.

Beckworth: Yes. Lorie Logan will be very happy to hear you say that since I know she loves the ample reserve system.

Mitchener: Yes, no, I’m just saying there’s different sides of this.

Beckworth: For sure.

Mitchener: Like I said, I have no horse in the race, so I think it’s useful just to kind of think through these. 

Beckworth: We’ll save that conversation for some other podcast. We have had people debate this on the podcast. Where I want to land the plane, though, is that I do think it’s important to recognize that ceiling facilities, the discount window, the sending repo operations, that’s important. The floor facilities are important. The deposit rate at the central bank is important. We want to make both of those very functional. Right now, we struggle more with our ceiling facilities. There’s a lot of stigma about using that. 

Mitchener: That stigma wasn’t there in the past, too, right?

What Anchors the Price Level

Beckworth: Yes. I think it is useful to have these conversations. I would encourage the Fed to have a review of its operating system. We’ve talked about this again on previous shows. I think this is all really important and fascinating. 

We will save for another podcast this conversation about what is the optimal operating system, the optimal size of the balance sheet, but at the end of the day, again, the framing of this podcast was what really anchors the price level. We talked about earlier credibility. Whether you went off the gold standard or not, it was this credibility. Could you commit and get back on it? Did the market believe you could actually garner the real resources as a government?

Part of your paper, you noted some examples of foreign exchange during this period where there were some losses there, yet they also did not seem to impair. Maybe in closing, Kris, if you can tie this back into what really anchors the price level and why this credibility component is so important.

Mitchener: Yes, I think that’s ultimately what we see in broader studies that have looked at recent losses in central banks across a broader set of countries, that central banks like to report positive profits or nonnegative profits in part because they are ultimately very concerned about the political perceptions of running losses. In part, that then fundamentally relates to questions around independence. Once you get to questions about independence, that raises the specter that monetary policy will have some other purpose other than just to achieve a mandate like the Fed has or many central banks have to maintain price stability.

Ultimately, I think central bankers see this issue as much as about politics as anything. I think that running losses is optically challenging because it could then cause their overseers, usually Congress or legislative branch or the executive branch, to sniff around and say, “Are you doing your job properly?” and to raise questions about, if you’re making losses, then maybe that means you also aren’t doing monetary policy correctly.

We’ve seen that rules-based monetary policy works best when it’s transparent and credible. If we want to compare another lesson outside my paper, but one where I’m sure many of your listeners and audience is familiar with, is that when we just think about the degree to which the Fed had to raise rates recently given the inflation shock versus what I was talking about earlier today with the rate going up to 20%, just look at the changes.

A lot of the reason the Fed could raise rates so little is because Volcker existed. The counterfactual of not having Volcker was we didn’t have the credibility revolution in central banking yet. The credibility in central banking revolution in this country really starts with Volcker and being tough on inflation. What do we know happened recently? We know with the COVID shock, markets started anticipating what the Fed was going to do, and they started doing the work for the Fed already. They started raising long-term rates even before the Fed. We could argue whether the Fed was too slow, but that’s a debate for another day, which I’m sure you’ve had.

The point is that markets started doing the work for the Fed because they believed the Fed would do it, because the monetary policy is credible. Things that undermine the credibility are things like running losses on your central bank balance sheet. Independent of these issues about interest on reserves and other things, I think why these losses to my co-authors and me are so interesting is because they really are about institutional design.

If we had another hour, I would try to convince you and your audience that it’s not only about central bank independence matters for monetary policy; it also matters for lender-of-last-resort policy. You’ll have to wait for my new book on that, that Éric Monnet and I are working on, that’s going to talk about this issue in the context of France in the 1930s, where central banks lost that independence through their voting structure and how that led to preferential lending and led to the largest banking crisis in French history as well as the first-ever bank to be bailed out in French history.

I’ll put it to you, David, as a great student of history, what bank was that? It’s a leading question. The central bank. The first bank bailed out in French history of any notoriety was the Banque de France.

Beckworth: Napoleon’s bank, huh?

Mitchener: Our book is going to explain why these issues around independence really aren’t just limited to operational functions that economists who write down simple monetary models like to study; they’re really related to all the functions of a central bank. This is what got us interested in this in some sense. Now we have two projects, one exploring France, and one now collecting all the losses back through time, and to try to think through why these losses occur and how we can categorize the losses.

We’re going back to the earliest central banks and really trying to catalog these, because we think there are important institutional design issues, just as in our earlier comments on the accidents around bimetallism and the gold standard point out, institutional design or the gold exchange standard in the interwar period, they point out how important the way we shape our monetary institutions is.

Beckworth: Well, on that high note, our time is up. Our guest today has been Kris Mitchener. Kris, thank you so much for coming on the program.

Mitchener: My pleasure. Thank you so much. It’s been a pleasure and I look forward to staying in touch.

Beckworth: Macro Musings is produced by the Mercatus Center at George Mason University. Dive deeper into our research at mercatus.org/monetarypolicy. You can subscribe to the show on Apple Podcasts, Spotify, or your favorite podcast app. If you like this podcast, please consider giving us a rating and leaving a review. This helps other thoughtful people like you find the show. Find me on Twitter @DavidBeckworth, and follow the show @Macro_Musings.