Related Links

• Money Under the Mattress: Inflation and Lending of Last Resort
• The Fed's Balance Sheet

Transcript

Tim Sablik: Hello, I'm your host, Tim Sablik, a senior economics writer at the Richmond Fed. Just a quick note before we start today's episode. I recorded my conversation with Gadi Barlevy about the Fed's goals of price stability and financial market stability on February 8, 2023, before the bank failures that occurred in March 2023. Now, on to the show.

My guest today is Gadi Barlevy, a senior economist and economic adviser at the Chicago Fed. Gadi, welcome to the show.

Gadi Barlevy: Thank you, Tim.

Sablik: I'm excited to talk with you today about a working paper that you're presenting here at the Richmond Fed's CORE Week entitled "Money Under the Mattress: Inflation and Lending of Last Resort." We'll include a link to that paper on the show notes for listeners.

In the paper, you and your co-authors are examining whether there might be a conflict between a central bank's goals of price stability and financial stability. This topic caught my eye in light of the recent turmoil in the UK debt market last year. The Bank of England stepped in to restore stability to that market. But there were some concerns at the time that this action could jeopardize its efforts to get inflation under control. Could you explain how these two goals might conflict and give some history of this debate?

Barlevy: I'll start with the UK experience, and then I'll talk more generally about the history.

In the UK fiscal tantrum that happened last fall, because of fiscal reform, interest rates rose. That put a squeeze on pension funds who were forced to sell bonds. When they sold bonds, that drove interest rates up. So, interest rates on bonds went even higher, creating a vicious cycle. The Bank of England then stepped in to try to stabilize this by buying bonds and keeping their price from falling.

But this occurred at a period where the Bank of England was trying to contain inflation — moving away from quantitative easing, when it was buying bonds, to a policy of eventually quantitative tightening, where you're selling bonds. In the middle of focusing on inflation and selling bonds, all of a sudden they're switching to buying bonds in order to prop up their price because of what's happening. That seemed like there was a tension to many outsiders. They're trying to do one thing but because of financial stability, they can't do that.

So, that's what happened in the UK. But, more generally, the argument for why stabilizing financial markets and price stability are at odds is the usual way we think about stabilizing financial markets involves lending to troubled financial institutions. That lending involves the creation of new money that is given to banks that are in trouble. Usually, we think of creating more money as something that's inflationary. That's the inherent trade-off.

This concern about trying to have financial stability by making loans to banks and that creates more money, that tension and the fact that it's inflationary, is something that policymakers and the public worries a lot about. During the financial crisis that started in 2007, I could show you many articles in newspapers and magazines worrying about the inflation to come because of these policies. If you go back to the mid-1980s when the Continental Bank of Illinois failed, there were articles at the time, too, about now the Fed is more worried about bank failures. Will these bailout policies that are under contemplation be inflationary? That's the general tension.

That said, in terms of the history, not everyone thinks that this trade-off is inexorable. One of the things that I learned while visiting the Richmond Fed is that economists here, while understanding the potential for conflict, argue that there are ways to resolve this tension. The policies that people here are focused on are a little bit different than the ones that we considered in the paper. But the theme is kind of similar.

Sablik: Right. And as you mentioned, this is definitely an issue that central bankers have been thinking about for some time.

What specifically motivated you and your co-authors to start exploring this?

Barlevy: I'm glad you asked that question because it gives you an opportunity to give credit where credit is due. This paper is co-authored with three people at Tel Aviv University. This started as a paper that they're working on — this is Daniel [Bird], Daniel [Fershtman] and David [Weiss], the three Ds. When they were working on the paper, they presented an early version to me. I gave them extensive feedback and they decided to invite me to be a co-author on the paper.

In terms of what motivated them to write the paper, they were thinking originally about the Treasury Asset Relief Program known as TARP which, under its original form, the intent was to try to provide financial stability by buying assets from banks during the financial crisis. This is kind of like what we're talking about in the UK of buying assets from troubled institutions. In the UK, we're talking about pension funds; TARP was about banks.

What they thought about is, by buying assets from financial institutions that are in trouble, is there a risk of being inflationary? Is there a way to design these asset purchases — by thinking about at what price you buy these assets — that you might be able to mitigate the tension between financial stability and inflation? What I pointed out to them when they showed me the paper is that, even though this was the way TARP was originally conceived, when it was actually implemented it was not about asset purchases. So, the issues that they were talking about were not really that relevant during the financial crisis.

But the general point that they were making applies to other ways in which central banks try to provide relief and address issues of financial stability — the natural one is, as I said, providing loans to banks that are in trouble during a crisis. What I suggested to them is that the same question that they're thinking about — what price do you pay for the asset — is kind of analogous to what interest rate do you charge a bank when you provide a loan during a financial crisis. That's a question that central banks think a lot about, and that we're providing a new perspective on.

Sablik: Diving into your paper a little bit, you and your co-authors use a Diamond and Dybvig model of bank runs to examine this issue. We don't want to get too deep into the math on this podcast, but I was wondering if you could explain why a bank run is a useful framework for thinking about this question of price stability versus financial stability.

Barlevy: Financial stability can mean different things. For example, if policymakers see the price of an asset rise very rapidly — more than can be justified by fundamentals — should they intervene, for example, by leaning against the wind and reducing asset prices?

Here, we're focusing on a very particular sense of financial stability. When we try to bail out financial institutions by providing loans, we're creating money and that's inflationary. So, you want to focus on that particular question — in bad times, when you're trying to bail out banks that are in trouble, can you do it in a way that doesn't generate inflation — which is quite different from these other questions.

So, I needed a framework in which we talk about what does it mean for banks to get in trouble, thinking about bailouts. This is where models of bank run and the Diamond-Dybvig model, which was the basis of the most recent Nobel Prize in economics, was very useful. It provides a setting in which banks can get into trouble.

A bank run in their model, which is very elegantly crafted, is a situation in which people deposit their savings in banks. Some people need to take money early, some people need to take money late. A bank run corresponds to a situation in which people who are patient and can wait to withdraw are nervous about, well, what if everybody else withdraws. Will there be any resources left for me? So, people withdraw en masse from the bank and the bank is getting stressed because it took everybody's deposits and is holding it loans that are not easily liquidated.

That's a situation in which the bank did nothing wrong, but because people are nervous, the bank is all of the sudden in trouble. That's a situation in which you might want to provide relief to the bank, because you're just trying to reassure the anxiety of these depositors that are nervous about what other depositors are going to do. So, a bank run is a perfect situation, I think, in which to talk about a bad outcome in which there might be a good reason for bailout by a policymaker, like loans from a central bank. We wanted to think about that situation and see whether a bailout policy is inflationary.

Sablik: In a standard Diamond and Dybvig model, when someone withdraws money from the bank, it's assumed that they immediately spend it. You and your co-authors added a feature to your model that allows agents to save that money rather than spend it. How does that help mitigate the conflict between price stability and financial stability?

Barlevy: I think the key problem with bailing out banks, in terms of price stability, is that it requires creating money and that money would tend to be inflationary. So, the question is, how can you make sure that the extra money that you're injecting into the economy doesn't create inflation. The people at the Richmond Fed who have thought about this focused on ways of trying to take out money even as you're injecting new money. The technical term for this is trying to "sterilize" the effect of a bank bailout.

The way we attack the problem is a little bit different. The idea is even when you inject money because you're trying to bail out banks, that doesn't have to be inflationary if the people who withdraw money from the bank because they're nervous don't immediately spend it, even though you're creating new money to lend to banks to give to depositors who are leery about staying in the bank. Because they're patient, they don't necessarily want to spend that money. They're just nervous about saving it within the particular bank that they're nervous about. By providing the money but also giving them a context in which instead of spending it they want to save it, you're preventing the injection of money from actually being inflationary.

The way we model that in the paper, just for simplicity, is you literally store money. The title of the paper is "Under the Mattress." It doesn't have to be that. The key thing for us is we're injecting this money, but we're also giving a reason for people withdrawing money who don't have a reason to spend it to sit on it and not create inflationary pressures.

Sablik: Part of that solution, I think, has ties to the ideas of Walter Bagehot, who was a 19th century English writer. His thoughts on finance and banking are still referenced by central bankers today. What is the Bagehot rule and how is that important to your findings?

Barlevy: The Bagehot rule is a dictum by journalist and intellectual Walter Bagehot that says that, during a crisis, what a central bank should do is lend freely at high rates of interest and against good collateral. We, in our paper, focus on a particular part of Bagehot's rule, which is during a crisis lend freely at a high rate of interest.

We're looking for a policy that induces people who withdraw from a bank to still save money as opposed to spending it. The argument that we have is that by setting a rule that says we're going to charge high interest rates on the banks that we lend to during a crisis, you might help convince people to save their money and store it as opposed to spending it.

By charging a high interest rate, you're effectively committing to draining liquidity in the future when banks have to pay high interest rates back to the central bank. If you're draining that liquidity, then people know that in the future there isn't going to be as much money circulating, and so they don't need to worry about high prices in the future. Now, if they don't need to worry about high prices in the future, they might be more willing to store money today, right? The problem with storing money under the mattress is if you expect prices are going to be high tomorrow, then your money isn't going to be as valuable and can't buy as many goods when you take it out of the mattress to buy in the future.

Sablik: Do these findings suggest any particular policies that central banks should adopt if they're trying to minimize this price stability-financial stability conflict?

Barlevy: What it does is it suggests new instruments for thinking about policy. As I said, people here in Richmond who have thought about similar issues have said you can reconcile the tension by potentially sterilizing an intervention — you'd lend to banks on the one hand, but then you do something else to reduce the money supply. For example, you might sell bonds and take money out of the system at the same time as you're extending loans to banks that are in trouble.

We're focusing on a different instrument, which is the interest rate that you charge on loans. I'm not trying to argue that one instrument is better than the other. I think it's just expanding the set of instruments we have to address this potential tension.

The other thing that I think expanding the set of instruments might help is providing a way to communicate with the public. Let's go back to the example of the fiscal tantrum in the UK that we started out with. Many people commented on the seeming schizophrenia of the Bank of England. On the one hand, it said that it was committed to fighting inflation, and that meant selling bonds. But now all of a sudden it is buying bonds for financial stability. So, what exactly are they trying to do? Our framework suggests an alternative. Instead of buying bonds in order to prop up their prices and prevent this vicious cycle, what in principle the Bank of England could have done is extend loans to pension funds and emphasize that it is charging a high rate of interest when it is offering those loans.

Now, I'm not an expert on the intricacies of the pension market in the UK. So, I don't know which rule is better. But in terms of communication, having a policy of extending loans at a very high interest rate does not sound inflationary. By expanding the set of instruments, you might do something that actually is equivalent in terms of what you're trying to achieve. But by doing it a different way, you can represent it to the public in a different way that creates less confusion.

Sablik: Yeah, that's a great point. I recently had Huberto Ennis on the show talking about the Fed's balance sheet operations. He brought up that point as well that the Fed undertakes these operations both for monetary policy reasons and also financial stability reasons, and maybe it's not always clear when it's doing which one.

Gadi, thanks so much for coming on the show to talk about this research. It was super interesting.

Barlevy: Thank you very much.

Sablik: I'll remind listeners again that you can head over to the show page for a link to the paper we were discussing today. And if you enjoyed this episode, please consider leaving us a rating and review on your podcast app.