Bill Nelson on Shrinking the Fed's Balance Sheet and Reviving Interbank Markets

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 Bill Nelson. Bill is a Chief Research Officer and Chief Economist at the Bank Policy Institute. Bill is also a former Deputy Director of the Division of Monetary Affairs at the Board of Governors at the Federal Reserve System. Bill joins us today to discuss the latest developments in balance sheets and operating systems at central banks around the world, and especially at the Federal Reserve System. Bill, welcome back to the program. Thank you, David. It's always fun to be here. It's always great to have you, Juan, because we share a common interest, maybe a common passion, and central bank balance sheets and operating systems. And I was thinking back to the first time I met you, Bill. I believe it was at an AEI event, American Enterprise Institute event, and we were on a panel discussing the Fed's operating system. We first met then. Yeah. It was George Selgin, yourself, me, and then the host. And since then, it's been a great journey, getting to know you better and learning a lot from you, your newsletter. In fact, I encourage all of the listeners and watchers of the video to subscribe to Bill's newsletter if you're not already. We'll provide a link to that in the transcript so we can do that. Now, Bill, I have a lot to talk to you about today. We know there's a new task force on the balance sheet, so we're going to talk about that later, some of the developments at the Fed. But before we do that, you are a man of international travels, I understand. You've been to Europe, central banks throughout Europe. You've been to the Philippines. Maybe tell us about what's been going on in your life. So I'm in the middle right now of a six-week sort of secondment from the Bank Policy Institute to the Central Bank of the Philippines. They asked me to do an outside review of their new emergency liquidity assistance framework. I went over there, spent a week there meeting them, meeting the various government officials and meeting people in the financial system over there and reading their new draft framework. And then now I'm at home, still meeting by Zoom, but it's complicated with a 12-hour time difference, and working on my report. And then I'll go back for a week in September to present it. So it's fun. It's something a little different, but it also brings back good memories of being more closely involved in public policy work. Now, you also have been to Europe, and I understand you went to several central banks there and did some presentations, discussions with them about these issues. That's right. So I started out in Florence, where the Bank Policy Institute co-sponsored an academic conference with the European University Institute, modeled sort of after our annual Columbia Conference, where we had papers presented that were then discussed by one market participant and one academic. It's a formula that brings people together to talk, and it's worked great. The EUI people are fantastic. But then I went up to talk about the issues that we care about in many ways, the balance sheet, but also the particular link of the discount window and liquidity requirements and how that links into the balance sheet at the ECB and at the BIS and then up at the RICS Bank. Invited by one of your guests and our mutual friend, a pair, Asperger. Fantastic. Yeah. I love the RICS Bank because they have, one, so much history. I mean, if you ever want to go back and get a time series from several hundred years ago, they have it. They have wonderful time series data. But more importantly, they got a really fascinating operating system. We'll talk about that later. Yeah. Probably as we get into our conversations. What were some of the lessons, insights learned from these conversations over there in Europe? Anything you can bring back home to the Fed? Well, the talk I gave, which is available online, and I structured it really as a speech because I had a lot I wanted to say and sort of a lot of things to tie together, walked through basically two things and then brought them together. One was an argument that liquidity requirements should recognize the capacity of banks to borrow from their central bank, that this is a very traditional role of the central bank from their beginning to provide such liquidity, overcoming a lot of sort of misunderstandings about about that issue. And then talking about the desirability of operating with a smaller balance sheet and how one could do that. And then tying those two things together, how in order to get smaller, you have to reduce banks' demand for reserve balances because you can't get any smaller than a size necessary to create those reserve balances. That's banks' demand. Reserve balances are deposits of banks at their central bank, and they're held for liquidity purposes and also for account clearing purposes and these days for investment purposes in some ways. So you have to reduce that demand if you want to get smaller. And there's a number of ways to do that, but one of them is to fix liquidity requirements so that they don't insist on banks holding reserve balances instead. That was actually one of the last things I did at the Fed before leaving was, I might have mentioned this before, but on my monetary policy side, I was involved in the shrinking of the balance sheet and the reduction of reserve balances during the first QT, and wearing my supervisory hat, I was learning that examiners were out there telling banks that they had to hold more reserve balances. So I was trying to get the two sides communicating with each other. That's fantastic. So there's two sides, maybe more than two sides, but at least two sides at the Fed saying different things about what banks should be doing with their reserves or their deposits at the central bank. And I actually went back to the Fed a few weeks ago to give largely the same talk and to maybe serve the same function to talk about these issues. And it was a great turnout. All of these, the turnout, the conversations were fantastic. And there were people from the research side as well as the supervisory side of the talk at the Fed. So continuing. I know some of this is in confidence, so you can't share everything that was shared there, but what was the reception? I mean, do you get a sense, both at the Riggs Bank, the ECB, and at the Fed, do you get a sense that people are like, yeah, this is a good, good point? They're receptive to what you're saying. It was a good, rich conversation. And I think there was a lot of receptivity at all of these institutions for this idea. Good turnout. There's a lot of interest in the idea. There's a recognition that there are also, you know, a number of problems or challenges that would need to be solved when implementing it. And we discussed those as well. And I discussed those as well. But part of my, what I wanted to do was to try to overcome some sort of basic misconceptions that seem to be out there. So for example, there's this idea that liquidity requirements are intended to prevent banks from borrowing from the discount window. Your guests and our mutual friend, Jeff Lacker, said as much in his recent show. I respect Jeff very highly. I don't remember him being at the table when we were designing liquidity requirements. And if you go back and read the Basel standard for the LCR, and if you read the U.S. implementing language for the LCR, you won't find in there anywhere the objective is to keep banks from borrowing from the Fed. In fact, what you will find is the recognition of the capacity to borrow from the Fed woven into the regulation as an option in the form of committed liquidity facilities. So that's one misconception. Another is that central bank lending is this peculiar, odd thing that only happens in crises. In fact, central bank lending is perfectly normal. And it's that normal lending that needs to be recognized. It's the Fed was created back in 1913 specifically to convert loans to reserves or currency to give banks the confidence that they would have that when they need it after the challenges of the 1907 crisis. Yet another is this idea that a bank is only self-insuring if they're holding highly liquid assets rather than establishing robust contingency arrangements or looking to the inner bank market. They're only self-insuring if they're holding those assets. But I keep a lot of money, not a lot of money, but I keep some money in my checking account in case something comes up around my house that I need to fix. I had a sewage problem the other day that caused us to have to move out and I have some major construction done. But I also have gone through the paperwork and taken the time to pledge my home to back up a home equity line of credit so that I have that in addition or in parallel to meet my liquidity needs. I don't see why that money in my checking account is self-insuring, but the line of credit that I established is not self-insuring. But somehow once you sort of allow the people that you're talking to to frame the discussion that way, you've almost lost the argument. It's like, well, who could argue with banks self-insuring? But in reality, that's only a very, you know, that's about a 15-year-old conception of what bank liquidity is. If you go back before the global financial crisis, no bank was expected to meet its liquidity needs with assets. They met them with robust lines of credit and having access to markets and being able to borrow from the discount window. It was considered something that only like a small, relatively unsophisticated, even lazy bank would do to have reserve balances. And I have a quote that I've used in some other writings, but also in the remarks that I gave in Europe of an examiner report back from the late 1970s talking about a small bank that just couldn't be bothered to keep track of its reserve account. And so it just kept a lot of extra money in its reserve balances. And it was pretty significantly and witheringly criticized by the examiner of how, you know, it was just pressing needs of other things, they said, largely, that kept them from actually keeping their reserve balance account down. That is so fascinating. And I'm so glad that you're making that case. Because as you said, the Federal Reserve's initial creation was to provide an elastic currency through what? Through the discount window. And yeah, I got that pushback from Jeff on the podcast. You probably heard we went back and forth. And he's not the first person, though, to do that. I had some other people say, David, you're at a market-leaning institution. How dare you support more discount window lending? And I'm like, well, every other central bank or most other central banks, that's something they do. And it's business as usual. You don't see more moral hazard there. But I thought more about this. There's even a more fundamental question here. Why is it that we assume that liquidity generated through open market operations is more market-friendly than through discount window lending? It doesn't strike me as the case. If the Fed's intervening in treasury markets, that's also a footprint. So to me, it's not a clear-cut case. You can't say we got to go back to the way we used to do things like open market operations, because that also is a market intervention, just as much as a discount window lending. And historically, again, you said that's the way it goes. So I'm glad you're making this case. And I'm glad to see the reception and the energy, which leads me to the next kind of big topic I want to discuss with you. And that is, there seems to be a sea change, a shifting of the Overton window, as I put in a recent substack. In terms of thinking about the Fed's balance sheet, along the lines you just mentioned, maybe adjusting some of the liquidity regulations to incorporate discount window facilities and collateral there. But just more generally, how do we reduce the demand for reserves, make the balance sheet smaller? Even champions of ample reserve systems are getting on board, right? So I want to step back and take stock. How did we get here? How did we get to this moment? Because when we first started talking, I mentioned when we first met at American Enterprise Institute, I think it was Paul Kupiak's event. Very different world, right? No one would even have this conversation. It was almost sacrilegious to bring this up. You, myself, Bill Nelson, we were a small but vocal minority making these observations. But now we're not. Now, I guess I want to ask, how did we get here? What is your sense of why is this conversation happening now? Why are we here? Yeah. Well, I'd like to say it was through the years in my decisive reasoning and good looks. But for those now able to see this on video, they know that that's at least not true. Yeah. But, you know, I guess partly it's worth remembering that the initial journey into a floor system was supposed to be temporary. It wasn't a journey that was taken with intention. In fact, as we've discussed, when the Fed first got the authority to pay interest on reserves, but before it had been accelerated back in April 2008, they looked through their options of conducting policy and they did consider a floor system. And they said, well, we shouldn't do that. That would be a radical departure. And moreover, it would crush the interbank markets. And they pointed to something that I think that you've wisely highlighted lately instead, the voluntary reserve targeting approach as a way to proceed. But so I think that as the crisis receded and balance sheets were beginning to get smaller, a lot of folks in the Fed are very vested in this idea of operating a floor system and have sort of rationalized it after the fact as something to keep doing. But I don't know how much that really resonated abroad, where people had different operating regimes and didn't necessarily blow their balance sheets up by so much. But I also think that the negative public reaction to the losses that many central banks have made abroad that were more visible, smaller, but much more visible than the losses that the Federal Reserve has made, left people with a sort of bad taste in their mouth for operating with a large balance sheet and inspired them to try to move back to a smaller balance sheet. It's interesting. I mean, when I list the cost of having a big balance sheet, I list the same thing that the Fed staff did back in April 2008 at the top, which is that it crushes the interbank market. And it's somewhat ironic that, and this comes back to your point about being pro-capitalist, of which I very much am, that the ECB, the BRICS Bank, the Norges Bank, the Bank of Canada, the Bank of England, and the Reserve Bank of Australia are all shrinking their balance sheet in part because they want to revive the interbank markets. And as your guest Claudio Borio said on the show, an arrangement where a bank looks first to the interbank market and only to the central bank as a backstop seems better than one where all of their liquidity problems are solved by their central bank. And yet, in 2016, when discussing one of their periodic reviews of their framework, one of the FOMC members said, I don't even know that an interbank market is a good idea. And so, it's funny that this bastion, the United States, this bastion of capitalism somehow seems to be falling behind Europe, at least in this regard. But, again, I think that that's gradually over time resulted in a sea change and resulted in all of these institutions looking more to a ceiling system rather than a floor system, which I'm not sure whether that will promote an interbank market either, but we can discuss that. Yeah. So, that is fascinating. And just to go back to this point, you mentioned about the losses being more transparent overseas. So, the Bank of England, they actually had to go to the finance ministry to get a capital injection. I mean, they outright recognized their loss, right? So, they had to be recapitalized. The Rix Bank, you were there recently, they changed a policy where they now – they have a deposit facility where banks earn 0% on a small portion of their deposits at the central bank until the Rix Bank has been recapitalized. And they set this up because they were having losses and they did not want to go back to the federal government to get recapitalized. They wanted to preserve their independence. So, it's striking. The central bank itself was worried about its independence because of the losses which were generating dependence upon the central government to recapitalize them. So, these other central banks who I think you could argue on average, Europe is probably more left of America, right? Their central banks are being more pro-market than the Fed had been. Yeah. That is pretty striking. It is striking. And so, when someone says to me, David, how can you support the more use of the discount window? I'm like, well, look at the other central banks. They're pro-market. They want to revive interbank lending. They want price discovery, right? That, to me, is a powerful story to tell. I agree. And I struggle, I admit, with the fact that some of the ideas I have for promoting the use of the discount window, even though what we're discussing now, it would be used as a backstop. Yeah. And a contingency source of funding, there are also ways that you could reintroduce it, reintroduce the term auction facility, for example, as a way to have it be more regular and ordinary. That could be something that could be done with a floor system or a necessary reserves framework. But there's a tension in my mind about the desire for the Fed to not be involved in markets. I think that the only ultimate way of making people comfortable with the window is that it be an ordinary thing. Yeah. And again, to be clear, just to reiterate the point I made earlier, one way or the other, the Fed's going to be injecting liquidity into the economy. As the economy grows, demand for liquidity or reserves or money grows with it. And you can either do it through open market operations, buy up assets, or you can do it through the discount window. Yeah. And so pick your poison. If you're, you know, one way or the other, the Austrians call it cantillion effects. It's going to happen one way or the other. And to me, the trade-offs are far better for a ceiling facility. You mentioned, if you don't use a ceiling facility, you mentioned we lose interbank markets. We have a bigger footprint in the treasury market, a bigger footprint in repo markets. There's the liquidity concern of Raghu Rajan. There's interbank independent. There's all these other things that I think we overlook. And, again, making discount window regular use, that's a – I'm willing to bite that bullet and go for it. Yeah. I think – I don't know that a ceiling system, which is another system where instead of the central bank being the marginal source of fund at the deposit level, the central bank becomes the marginal source of funding at the lending level, is necessarily the path to reviving interbank markets. You know, I've talked to a lot of central bankers that are doing this about that, and they share that concern as well. And so, you know, it could be that you use that approach to get to where you're going, and then you widen out your corridor a bit. Yes. But there has to be daylight between the bid and the ask in the market being provided by the central bank for there to be an interbank market in which to operate. Yeah, and that's a great point. All of these other central banks are on a journey. Yes. Many of them are slightly different. We don't know where they'll end up. We don't know where the Fed will end up. But we need to start a journey. Let's just get a journey going in that direction. But let me throw this out there, and I know we're spending a lot of time on the ceiling facilities, but we'll get back to the normally programmed schedule on this podcast. But to me, the way I would respond to that concern about a ceiling facility is, a ceiling facility, the long-term growth in reserves would come from the ceiling facility, but like day-to-day liquidity would be met through the interbank market, right? Like one bank needs liquidity today, they go to the other bank and they borrow the reserves. But over time, if there is a structural shortage of reserves, then rates are not going to be in line. They're going to go to the central bank. Is that a fair distinction, more of the structural versus like short-term cyclical? It sort of depends upon the details and how I think about it. So the Bank of England, the ECB, probably other central banks are setting up their system so that they are auctioning off term loans on a weekly basis. And then you're right. If you're not on a day when the auction occurs or if the auction occurs earlier in the day, the auctions aren't going to be the mechanism where a liquidity misallocation or need is going to be worked out. That's going to happen in the afternoon. And that will be either in the interbank market or it will be recourse to the standing daily facility of the central bank or extra funds will end up in the standing deposit facility. But, you know, whereas that auction rate can be at the market rate or close to or be your target, I really think that ultimately the lending rate and the deposit rates have to be a spread apart from those. And many of the arrangements that are now being set up currently are not like that, right? Just like at the Fed, there's very little daylight between the target and what's supposed to be the ceiling facility. So, as I wrote in a piece a little while ago, you know, the Fed actually has a problem to fix because they set up the discount window in its current arrangement to operate with a substantial spread so that people used it as a backstop when it was appropriate to do so but didn't use it all the time. And it's not set up that way. And it's not clear exactly what's going on. But that's my thought. Okay. Well, that's good. But in general, there's a way to wrestle with this, to use a ceiling facility that both recognizes you want to be careful, but you need to use it more regularly as well. And again, to me, it's trade-offs. This is not a perfect world. And you can have all these other issues I just outlined, lack of internet lending, liquidity challenges, the Raghu Rajan story, the ratchet effect, central bank independence concerns, all those things, big footprints in financial markets. Or we can have a little bit more access to the discount window. I'm willing to go with that. But let's circle back to this bigger point about there's now at least a conversation about this, right? I mean, we're not the only ones having this. There's a wide conversation. So you mentioned one big development is the rest of the world has moved in this direction. It looks like the Fed may be moving in that direction. So let's go to a key reason for that. Kevin Warsh, the new Fed chair, and he has a task force on balance sheets, operating system. It was interesting listening to him in this last hearing for Congress. He talked about his words, I believe, where there are, quote, other sustainable equilibriums in terms of an operating system. So he's open to other possibilities. So I think he was a pivotal, like, catalyst. So there was already momentum building, but having him, because once he was nominated as the chair and before he, I think, was officially confirmed, we saw just a spate of papers come out. Laurie Logan and Sam Schulhofer-Wolf's paper came out. Stephen Moran and his co-author's paper come out. Daryl Duffy's paper come out. So there's been a lot of stuff. Once you see the writing on the wall, okay, how do we work within this direction that the new chair is taking us? So, and the panels look great. I mean, I'm pleasantly surprised how the panels are going. One last question on this sea change that we're experiencing. We've listed a number of things. Maybe another issue is the concern about central bank independence. Do you think there's, like, a growing recognition about central bank independence also motivating this move? And I know you've talked to people like Randy Quarles, who's made this. He had a famous quote in a transcript where he said this. But do you think that's also some of the momentum that's pushing this conversation forward? Well, I think that it's certainly one of my more significant concerns. And as you mentioned, it's a concern that Randy has mentioned. And Loretta Mester, former president of the Philadelphia Fed, in the same meeting. So, let me read former vice chair Quarles' quote, because it's wonderful, like many of his quotes. So, this is from the November 2018 meeting, where, actually, Loretta expressed similar concerns. But what Quarles said was, quoting here, Having the FOMC control such a large stock of assets presents what the lawyers in the room will call, from your first-year torts class, an attractive nuisance. And for the non-lawyers in the room, an attractive nuisance is an object that a property owner allows to remain on his land when it is obvious, both that the object will be dangerous if misused, and that misusing of it will be irresistibly appealing to passers-by of impulsive and immature judgment, such as children and congressmen. That's the classic Randy Quarles. Yes, yes. And when I wrote this, I mean, then I followed it up with a list of different ways that the government has sought to use the Fed's balance sheet now. You know, to fund things. And it's not just an abstract concern. Anyway. Well, I guess I invoke that because, like, the Rooks Bank, I gave that example. They developed this new deposit facility because they didn't want to get recapitalized by the central government because they're worried about independence. Yes. Which is, again, tied to large balance sheets. You know, the Bank of England, it had to be recapitalized by the government or the finance ministry. So there is a concern, if you truly recognize the losses from an ample reserve system or from QE, that you're going to be more dependent on central government to help get your finances in order. I mean, you could take the Fed's approach, which you could eventually earn it back over time through a—you call it magic asset, I think is the term you use for it. But so I think independence is somewhere in this conversation and motivating these changes, at least at other central banks. Well, owning assets and funding those assets with interest-bearing liabilities, that in and of itself is sort of a zero—a net zero cost in general, an expectation. You're funding interest-bearing assets with risk-free interest-bearing liabilities, even with longer-term assets. Term premiums can be positive or negative. You're largely making an investment that isn't going to cost you anything. But if you take a lot of risk, sometimes that risk will pay off. In the 2010s, interest rates ended up being much lower than the Fed or anybody expected because of the very slow recovery. And the securities that the Fed invested in under QE1, 2, 3, it ended up making them money. Now, that was swamped by the interest rate risk that they took under QE4. And then, unfortunately, that's when interest rates rose sharply, and they lost a lot of money, a tremendous amount. So, to take the Rixbanks example, well, the Rixbanks, their situation is pretty interesting because they have a very low demand for currency. Central banks are basically, historically, money-making enterprises because they have the monopoly on issuing currency. And currency is a zero-interest loan to the central bank that they invest in assets and earn interest on. So, that's how we were all brought up. And so, it's very natural for us to kind of discount losses as not being very important. But, in fact, their real losses, as I am almost, it's hard to keep explaining to people. No, they're real losses. Even when you consolidate the Fed onto the Treasury's balance sheet, they're real losses. But, nevertheless, the Fed has a very large quantity of currency demand, partly because half of their currency is held abroad. So, they're inherently profitable. The Rixbank, the currency demand is close to zero. So, the way they've set things up is that they were recapitalized. And then, if they need to get recapitalized, they can charge this zero-interest requirement on banks. But then, that goes away once they're recapitalized. So, if you think about their situation, once they're recapitalized, they're basically a trust fund of money from the government that they then can use to invest in interest-bearing securities to pay their bills and keep the lights on. Where the non-interest-bearing required reserves is just a way to top that up. But, they could still need recapitalization if they took risk and they made a loss. So, it's sort of ironic to think about the Central Bank of Sweden and others as they're just trust funds doing monetary policy. But, since there's no currency demand, there is no central bank liability. I mean, the Fed also has the TGA, which is massive and is a big generator of funds. But, that doesn't need to be massive. Other jurisdictions don't handle things that way. They pay interest on their government deposits. So, anyway, I'm not sure if the Rixbank's Central Bank independence is greater with this arrangement. Okay. Because, they're still – if they take risk, they're still subject to the need to get recapitalized. But, maybe they are. Certainly, the ability of the Fed to be funded by its profits has been an important vehicle for its profitability. But, on the other hand, in the United States, there has been very little attention paid to this issue of the losses. That's the point I'm making, is if the Fed had to go to Congress and say, hey, we are having these losses, we need to be recapitalized, so it explicitly recognized what's actually happening right now, there would probably be more attention. There would be more scrutiny, which would undermine the independence. The way that it's doing it now with its – you know, it's the asset that it recognizes. It kind of hides it. It's behind the scenes. Yeah. You don't really see it. I think you're right, David. I think that's a very good point. But, you know, again, the Rix Bank is useful for another reason. I know you're very sympathetic to this point. If you do lose your currency franchise, you're in trouble. You're definitely more vulnerable. And dollar-based stablecoins and the limit could be that very thing. I don't think we'll get there. But this is a point that your former colleague, Jim Klaus, makes, is that if in the limit – dollar-based stablecoins displace all physical currency, and I think that would be probably the scenario, you know, overseas, people start using stablecoins over physical currency, then there goes the Fed's currency franchise, and uh-oh, we're really in a lot of trouble in terms of the, you know, the cheap funding disappearing, because the senior rich is now being captured by Tether and Circle in that scenario. Now, that's a very far-fetched scenario. I don't think it would happen, but that is something that Rick's Bank illustrates well. Right. So, well, let's move on to some other work that you've been doing, and I want to move to a note that you and Lori Bristow wrote about the Fed's early use of a demand-driven system. Again, the bigger context here is a lot of other central banks are moving, or at least claiming to move to a demand-driven system, and lo and behold, I didn't realize, Bill, until I read your note that we've actually tinkered around in that space before. Yeah, I mean, it's so, I have to admit, so it's not exactly a demand-driven system, but it's close enough that the title is kind of catchy and makes it timely. And part of my motivation was to explain to people that borrowing is ordinary, even though, you know, people who are sort of new to the scene see discount window borrowing as this thing that only happens when a bank gets in trouble or being used for bailouts or things like that. That's a very contemporary view. It's a very harmful view of central bank borrowing because it contributes so much to stigma, but it's also just not historically accurate. I mean, as you mentioned, it used to be that the Federal Reserve charged a below-market rate. The story behind how that happened is kind of interesting. Did I ever tell you this story? No, go ahead. Please do. So, it used to be that the money, that the federal funds rate was about equal to the discount rate, and the Fed signaled its policy intentions with a discount rate, or it'd be a bit below the discount rate, as you would normally expect a central lending facility to be. But then in the mid-1960s, the Fed came under a lot of pressure from the Johnson administration to not raise rates. This was the famous Martin being brought out to the ranch and pushed around the ranch episode. And there was a lot of pressure from the Congress. The Congress and the President were all of the same party, and they all pushed against the Fed from raising rates. So, what the Fed did was they said, okay, fine, and they just held the discount rate constant. That was the rate people focused on, but they tightened up the condition in reserve markets so that the federal funds rate, money market rates, moved above the discount rate. And that was in the mid-60s, and it stayed there until 2003, when I had the great pleasure of helping Brian Madigan and Lyle Gramlich and others redesign the window and make it an above market facility. But for, you know, a lot of my career and Jim Klaus's career, he taught me a lot of these things, it was a below market rate. And that meant everybody always kind of wanted to borrow, but there were rules that prevented them from borrowing. And there was also some stigma associated with borrowing. Mostly, he didn't want to get called into the discount officer's room and explained that it's not supposed to be an ongoing source of funding. So, the higher were market rates, the more people borrowed, because the bigger the spread between the discount window and those higher market rates, the more the profitability. So, that would pull out borrowing, and that would create sort of a smooth function between borrowing and the Fed funds rate. So, after Volcker's experiment with what was called non-borrowed reserve targeting, it was reserves minus discount window borrowing, which was used to great effect to break the back of inflation, interest rates rose very sharply, very sharply. And it was, they transitioned into this non-borrowed regime, which had a sort of a shadow Fed funds rate that they were seeking, but they were very much doing it by targeting borrowed reserves. So, they were leaving the system short, like in a borrowed reserve, like in a ceiling system, like in a demand-driven system. They were leaving the system short and filling out the remainder of the reserves with borrowing from central banks. There was sort of always there. It wasn't necessarily the same banks, but there was always an asset, which was borrowing from the banks. And it was the positive relationship between that and the Fed funds rate that created this smooth borrowing system, that function, that allowed them to implement policy. Now, I don't want to completely, I mean, for a long time, I kind of thought, well, honestly, they were really targeting the Fed funds rate, and they knew they were targeting the Fed funds rate. And this was just sort of a backdoor way of not doing it. But, you know, after I read about the 1950s Treasury Accord and other period and read further to research the note you're referencing, I realized that I was being anachronistic. People viewed the world differently. And the money, you know, the money relationships held, and people viewed the world, they really attached a lot of importance between balance sheet items on the Federal Reserve and in people's lives rather than interest rates as the means for transmitting policy. So it's an over, it's a bit naive and a bit, I don't know the right word for thinking too much about your own time and not understanding previous times to just think of this as really a form of Fed funds rate targeting. It was a mixture. There was an element of both. Funds rate targeting had sort of been discredited in a certain branch of economics. So there was an element of sort of trying to not look bad according to that branch of economics, but at the same time recognizing that strictly targeting non-borrowed reserves resulted in too much volatility. That is so interesting. So we've kind of dipped our toes in the pool of demand-driven system, at least temporarily, for a short season. And we survived. It worked well enough. So we could, theory go back in that direction. But again, your caution is the devil's in the details if we were to go that path. Like there's got to be some kind of spread. We still want interbank activity as well as banks to feel comfortable going to the discount window. And there's lots of ways you've outlined already how to do that. So hopefully we can get there. Hopefully this task force will be a part of it. We've got some great people on it. Right. Any other thoughts on the task force? Any of your hopes, aspirations for it? Well, you know, it's certainly a very talented group of individuals, all of whom I know, all of whom I hold in the highest regard. And so it's not completely clear exactly what their mandate will be, whether it will just be reducing the size of the balance sheet, or it will be rethinking how monetary policy is implemented. I'm really not sure whether that implementation issue is going to be on the table because there are ways to reduce the size of the balance sheet that really don't change the underlying structure, right? If you reduce the treasury's general account, or if you reduce bank's demand for treasuries or reserve balances by adjusting liquidity requirements, that will reduce the size, but it won't necessarily revive the inner bank market. It won't put markets first necessarily in the means to, for banks to meet their liquidity needs. So it's great to reduce the size of the balance sheet. That has a lot of good qualities to it, but it's by itself, it won't accomplish at least one of my top objectives, which was to make it the money market arrangements, a more market focused. Yeah. And again, that's what many other banks are doing. They're doing this journey toward a demand-driven system because they want to see a revival of interbank lending. They see an important role for that, price discovery. Also, banks go to each other first before they go to the central bank. You can have it all. You can have a robust ceiling facility and interbank lending according to them. So you're saying that we've got to be careful. We can go part of the way there. We may not go all the way there. In fact, I think you made this point in that note, I believe, from your talks in Europe that just simply reducing the demand for reserves through adjusting liquidity regulations, all these other things we've talked about, doesn't go far enough. I do think that an important part of the path to shrinking the balance sheet is to reduce the balance sheet enough to start to move money market rates a bit above IORB as the Fed was doing last year. And as the year progressed, you could see repo rates moving. I made a little cottage industry of writing a little note saying, two weeks from now, there's going to be a big shortage of reserves because coupon securities are settling and you need to do an open market operation or there's going to be volatility. And then you're going to say, whoops, there's volatility. We had to stop shrinking. Right. And they didn't. And there was. And then they stopped. So if you just do a little management, fill in some of those bigger potholes that they are seen well in advance, they know how to do that. They're not complicated. You can establish a situation where banks have a financial incentive to get smaller. And a lot of the reduction in reserves can just be accomplished by that. If you go back to before September 2019, where there was repo market volatility, another instance of a pothole that I at least, Luke Randall saw well in advance, but the Fed didn't take necessary steps to fill in. Their Fed was saying banks, and this was all long after all these liquidity requirements had come into place. Fed staff was telling the committee that there was only, they only needed a trillion in reserve balances. Now somehow that's three trillion. So I can't believe that the structural demand can at least get back to one trillion if you take some of these steps. Maybe bigger. The economy has grown since 2018, but it hasn't doubled. It hasn't tripled. Right. Yeah. And so, yeah. So I think that that is the engine that needs to be in there to make all of this work. Yeah. Banks have to have at least some incentive when they end up with extra reserves to lend them out. Yeah. So as I mentioned, before, JPMorgan Chase had $400 billion in their account at the Fed at the end of 2023 and at the end of 2025. They had $100 billion. Now, there were a lot of things going on, but one of the things that was going on was that repo rates were moving up relative to the IORB rate. It made sense financially to bring that down. They reduced the size of their reserve balances in 2018, 2019 as well. And at that time, they said the reason was because it makes sense. Money market rates are above IORB. So we haven't quite gotten to money market rates being above IORB, but a little bit more. And then everything kind of pushes in the right direction. Bill, let's talk about a different approach to shrinking the Fed's balance sheet. In the case of Norway, they took a different approach. They took a tiered reserve operating system and they adopted it in 2011. And as you know, because you've looked at this closely, they set quotas. You can only have so many reserves that earn the deposit facility rate. After that, they don't or it's reduced. And so it creates incentives for banks to reduce their stock of reserves. Now, many people have pushed back against that. I think it's a fascinating story. And we've brought a link to my write-up of it in the transcript. But some have said, look, but if we do that, then we have the central bank determining the quotas. That's pretty heavy-handed market intervention as well. But you and others have pointed to a 2008 memo by the Fed, and you've articulated this in your own work. Well, there's a version of this that is more market-friendly, that is driven more by the bank's demand for reserves, and that's a voluntary reserve target. So maybe tell us about that and how that could be a possible option for the Fed as it thinks through its balance sheet task force. Yeah, I'd be happy to. And you're right. I just don't think that the idea of establishing quotas for banks works in a system where you have, I forget how many banks, thrifts, and there's thousands of small credit unions going in there and telling them, this is how much you get to have bank by bank. It's hard to imagine that working. And if you were to make it a percentage of their demand deposits, well, that was the system that caused all the trouble back for the decades before 2008 that led the Fed to want interest on reserves in the first place. That led Milton Friedman to want the Fed to have interest on reserves in the first place. That, first of all, originally, it caused problems because it caused banks to leave the Federal Reserve System. And then it caused problems because it led banks to reduce their demandable deposits through sweep accounts. Back then, sweep accounts meant the opposite of what they mean now. Now they mean sweeping your brokerage account into a demandable deposit. Back then, they meant sweeping your demandable deposit into like a money market mutual fund kind of thing. And reserve balances had gotten down to like $7 to $10 billion. Tiny number that was so low that it was hard to conduct monetary policy. Plus, there was the fundamental problem that forcing banks to hold unremunerated reserves is forcing them to make unremunerated loans to the government. Reserve balances are just loans to the government, like holding T-bills. And why should you and I get paid interest on our T-bills when banks are somehow being forced to hold government securities and not be paid interest on it? It has sort of a Banana Republic feeling to it almost, right? And that was precisely what Milton Friedman objected to, that, you know, this forcing banks to make free loans to the government isn't a desirable thing to do. And that's why the Fed, including Don Cohn, who's been a visitor here, I think will be again, was a very strong advocate of paying interest on required reserves. But then when the crisis came, they actually used the tool to pay interest on excess reserves as a way to conduct monetary policy. Now we're in a situation where to get banks smaller, the thought is, well, in Daryl Duffy's recent paper at Brookings, he spoke about this and others that spoke about having a tiered system. And you're, I think I completely agree. The challenge is, well, how do you figure out how much banks get paid the market rate on, right? And how much do you pay a lower rate on? And under a voluntary system, you let banks choose. It's a great system, but there's a problem with a system that'll circle back to. And I thought your write-up on it was great. Everyone should be reading your sub stacks. They should read that one. And so under a voluntary reserve system, the banks say, we're going to hold about this amount, maybe over the intermediate period, say six-week intermediate period. If they end up with more than that on average, say, then they're paid on that extra less than the market rate. Maybe, who knows, 50 basis points less. If they hold extra, then they're charged a fee which replicates the cost that would come from borrowing from the discount window to fill in that extra amount. Otherwise, banks will just hold a lot extra to avoid ever having to be. So you're kind of, it's counterproductive from your perspective, but it's also counterproductive from our perspective in a particular way. If you're paying banks the market rate on the voluntary reserves and they get to pick those reserves, then banks will continue to see reserve balances as a very cheap way to meet their liquidity needs. There won't be that extra incentive to make the engine go to get the balance sheet smaller because they will be able to set their voluntary reserve targets wherever it will be that they would currently, probably where they have what they're holding now. It would encourage an interbank market in the end of day need, right? But the fundamental driver. So I always, when I describe this, I always envision something where you're paying them 10 basis points less, say, than the market rate on their voluntary reserves. But then you have to get the system just right so that you're delivering a market rate of 10 basis points more than that. And I haven't really thought that one through all the way. So by itself, it's not going to be the fix. It may be one piece of the puzzle. Yeah. You need to do all the other things we mentioned earlier to reduce maybe structural demand for reserves, liquidity regulations, bring back term oxygen facility, do a number of things. But in conjunction with that, maybe bring back a voluntary reserve target. Yeah. And again, the devil's in the details. If you really work these things out, experiment, lots to learn here. But I guess the push is let's at least talk about this, right? Let's consider these alternatives. There are alternatives out there. Lots to learn. We're not stuck in one system. Yeah. Yeah. You know, circling back to the fixing liquidity requirements thing, I want to just emphasize that this has been a passion of mine since about 2012. Maybe you could go all the way back to 2008 when the people for whom the new Basel liquidity requirements were just a twinkle in their eye as they were beginning. And I was sort of consulted as well as for the discount window, how should this come into it? And I fought for years, you know, in the system and in the Basel system and all the way up to get, in that case, committed liquidity facilities recognized. That proved to be a very complicated story today. So the manifestation today is, well, now the Fed is offering 90-day loans, renewable and repayable on request. They still are offering a no questions asked facility. And the point is, you should be recognizing that because as we saw in March, 2023, it was emphasized something I've been saying all along. The reason you should be recognizing it is that a bank that is prepared to borrow is more liquid than one that isn't. And you should be designing your requirements to be accurate reflections of the risk that you're trying to control. And if you don't do that, you're not creating proper incentives. So SVB, for example, about a year before they failed at BPI, they were a BPI member. They called me up and said, do you think if we signed up for the Fed's standing repo facility, that that would be a method by which we could tell the Fed we're going to monetize all these treasuries and agency MBS that we're holding in order to satisfy our internal liquidity stress tests? And I talked to the Fed about this. And as we've seen now from the stuff that's been released, they were failing their internal liquidity stress tests, even though they were awash in liquid assets, precisely because they didn't have a way to convert them into liquidity. And they weren't willing to go all the way to the discount window, but they were interested in the standing repo facility. Well, the answer I got back was no, that the Fed wasn't ready to go there yet. And so they didn't sign up because they didn't use the repo market. And it's expensive to establish the relationships that they would have needed to establish to count. So they just didn't sign up. And then, of course, you know, on Wednesday, they tried to raise money by selling their available for sale securities rather than, say, using the standing repo facility. That brought market attention to their losses. They weren't able to raise capital. Thursday, the run began. Friday, their doors were closed midday, which is why the Fed had to take all these horrible moral hazard-inducing actions of bailing out all those uninsured deposits, of guaranteeing effectively the deposits around the banking system, of creating the awful bank term funding program, which made unsecured loans and did all kinds of things that central bankers aren't supposed to do, all because it wasn't possible to have an orderly failure. And that was because the Fed sent a negative message about being prepared to use the Fed's lending facilities. Now, I'm not saying, I emphasize this, I'm not saying Silicon Valley wouldn't have failed their losses were equal to their capital. And there's a very good chance that even if they had been able to raise liquidity, they would ultimately have failed. But it would have been orderly. And my view has always been, you know, so with going back to my experience where I was more involved in the discount window and I saw it being used to get a small community bank that was in trouble on Wednesday to Friday so that it could be closed in an orderly fashion. That's the dirty little secret about the discount window. It's often used to lend in solvent institutions so that there can be an orderly failure. Almost always those institutions are very small. So my view has always been no institution is too big to fail in a disorderly way, but every institution is effectively too big to fail in an orderly way. And so you need to be able to use those tools to support an orderly failure. And that's the kind of liquidity that simply needs to be reflected in the design of liquidity regulations. Could a SVB today in that situation go to the standing repo facility or is it still. They could. And that's a. Sorry, I'm glad you asked that question because in early 2024, I think it was, or summer 2024, the Fed changed. They issued an FAQ that said banks can now point to the standing repo operations or the discount window or the federal home loan banks as the means by which they would monetize their assets. So they haven't gone all the way to recognizing the capacity to borrow against loans and illiquid assets that make up most of the collateral. That are the ones that are the most valuable growth-inducing things step to take. But they have fixed this and I applaud them for that. It was a very important fix and it was very well received by the banking industry. So just to recap, had this been available or something like this available back in 23, March 23, then the SVP could have like tapped one of these facilities standing repo operations, maybe the discount window. You said they probably wouldn't have gone there. But instead of making that accessible, they instead had to go to the ugly version of this, the bank funding program. Well, no, they were failed by the time the bank funding program was created. The very, you know, the thing that everybody tries to avoid happened. They were closed midday on Friday and the banking agencies and the FDIC. Oh, those are for the other bank. Okay. Yeah. They had to stand up and say, to prevent a run, it really felt like they kind of looked around the banking system and said, there's a lot of unentered impositors out there. So to prevent a run, they had to stand up and guarantee those uninsured deposits. But your point is, if the facilities had been available to SVP in March of 2023, it would have been an orderly resolution of the bank. It may have avoided some of the commotion and concern about other banks. And maybe there wouldn't have been this bank funding facility that had all the moral hazard problems later. That's kind of the story you're telling here. Yeah. And it's not just a hypothetical story. If you look at the end of 2022, Silvergate Bank, which was also very involved in the crypto industry and was also ended up in trouble, it decided to self-liquidate. And so it experienced a very large run on its deposits. I think that they might've been 60%, but they were prepared to borrow from the discount window. So to meet the run, they borrowed against their collateral at the discount window, and then they slowly sold off that collateral and repaid their discount window loan. And then they shut down. And there was no FDIC cost, or there was no need to, there was no alarm, you know, so no potential bank panic. So that's how the system is supposed to work. But for that to happen, then the banks need to be prepared. Now that's a bit different kind of borrowing than what we started this program out with, the ordinary create liquidity as a central bank, but it's still a very important function of central bank lending. Absolutely. Well, we look forward to the balance sheet task force as they process these arguments, think through these issues, look forward to what Kevin Warsh has to say, the FOMC, what eventually they will make of all of these findings. So when that happens, Bill, we'll have to have you back on. We'll further deliberate on where we stand because it's been a great show taking stock of where we are today here in 2026. So our guest today has been Bill Nelson. Bill, thank you for coming back on the program. It's been a pleasure, David. It's always fun to chat with you. Thank you. Macromusings is produced by the Mercatus Center at George Mason University. Dive deeper into our research at mercatus.org forward slash monetary policy. 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 at David Beckworth and follow the show at macro underscore musings. Transcription by CastingWords