Laurence Bristow on What the Fed can Learn from the Reserve Bank of Australia

Laurence Bristow is a former staffer at the Reserve Bank of Australia and currently is a Vice President and Research Associate at the Bank Policy Institute. In Laurence’s first appearance on the show, he discusses the differences between the Reserve Bank of Australia and the Fed, The RBA’s change in operating systems, what a demand driven system actually looks like, the motivation for the RBA to make this change, calls for changes to the operating system within the Fed, and much more. 

Subscribe to David's new Substack: Macroeconomic Policy Nexus

Read the full episode transcript:

This episode was recorded on November 20th, 2025

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 Laurence Bristow. Laurie is a former staffer from the Reserve Bank of Australia, or RBA for short, where he played a key role in the redesign of the RBA’s monetary policy implementation system. Laurie joins us today to discuss the RBA, its new operating system, the broader push among many central banks toward a demand-driven operating system, and finally, its implications for the Fed’s operating system here in the United States. Laurie, welcome to the show.

Laurence Bristow: Good day, David. It’s a pleasure to be here.

Beckworth: It’s great to have someone here with an Australian accent. I think this is a first for us, but also someone who really knows central bank operating systems. You were in the inner sanctum. You were in the RBA helping redesign its implementation or operating system, so I’m really excited to talk about that in a bit. Before we do, though, tell the listeners about yourself, your career, how you got to this place.

Laurie’s Career

Bristow: Yes, thank you. Before we get started, I just want to say that the views I express today are my own and are not representative of my employer, the Bank Policy Institute or the Reserve Bank of Australia. I got my undergraduate degree from the University of Queensland, majoring in economics and finance. I think coming out of that degree, I was interested mainly in research and in monetary policy, so I went on to get a one-year master’s degree again from the University of Queensland, after which I joined the Reserve Bank of Australia in February 2021. The first team that I worked in at the Reserve Bank of Australia was in the risk management department. That team is split into two, so in the side that I was in was the international side, and in that team, we managed the risk on our foreign reserves portfolio.

The Reserve Bank of Australia has about a $60 billion Australian dollar portfolio held across seven currencies, which it stands ready to use if there’s any dislocation in currency markets or if it feels that the Australian dollar is trading a long way away from fundamentals. The other side of that risk management team is the domestic side. It was a very exciting time to join the RBA in February 2021 because the RBA’s balance sheet was expanding due to QE. We were giving out long-term loans to banks and taking collateral against that, and so the risk of the domestic side of the portfolio was a big topic of conversation.

After my time there, I was rotated into the monetary policy implementation team, which was a fantastic time in Australia to be in that team, because soon after I joined the monetary policy implementation team, we began a full-scale review of our monetary policy implementation framework. Since working at the RBA, I have recently come across to Washington, DC. I’m working for the Bank Policy Institute as a vice president and research associate. I found out about the role through my boss, Bill Nelson, who I know is a friend of the show.

Beckworth: Absolutely.

Bristow: I was writing a working paper on reserve demand in Australia for which he provided a peer review. We got to talking, and he offered me the job to come across to Washington, DC, and I’m very excited to be here.

Beckworth: It’s always great to have someone on the show who likes to nerd out about central bank operating systems. There aren’t many of us, but those that are around, I love to have them on the show. You bring, again, a unique perspective. You can tell us about the changes that are happening there, what it’s like to actually do a review of an operating system. I’m eagerly anticipating the Fed will one day take on the process of doing their own review. So far, it has not happened. We want to talk about this because I think it’s very timely for the Federal Reserve. There’s been a lot of conversations, a number of speeches.

Governor Michelle Bowman had a talk about the Fed’s balance sheet. She’s the first Fed official in recent memory who has said, “I would like to go back to a scarce reserve system,” which is pretty shocking. Now, she’s probably a minority. However, Stephen Miran had a recent speech as well. He’s a governor, and he talked a lot about the regulatory constraints on the Fed’s balance sheet. In fact, he talked a lot about the points that your boss, Bill Nelson, has often echoed on this podcast. There are two champions there, at least, wanting to reconsider. Lorie Logan has also had several speeches. Of course, she supports the ample reserve system. A lot of, I think, robust discussion going on about the Fed’s operating system. I think the time is ripe to maybe have a review. We can learn from the RBA, from the Reserve Bank of Australia, so I’m excited to talk about it. 

Reserve Bank of Australia

Let’s move to the RBA. Maybe before we get to the operating system, or what you call the implementation system, maybe just tell listeners a little bit about the RBA, if they’re not familiar. What do they do? How’s it governed? What does it target? Those basics.

Bristow: Yes, fantastic. The RBA’s inflation target is set out in a document called the Statement on the Conduct of Monetary Policy. That document records a common understanding between the government and the Monetary Policy Board on key aspects of monetary policy. The RBA operates with a flexible inflation target, aiming to keep consumer price inflation between 2% to 3%. The extra detail on that is that in the setting of monetary policy, the board aims for inflation to return to the midpoint of that target range. I think in terms of how the RBA is governed, it’s changed quite recently because of the RBA review.

When I was there, we just had the Monetary Policy Board and the Payment System Board. Coming out of the RBA review, one of the recommendations was to add in a Governance Board. That Governance Board is responsible for RBA operations like banknote issuance and the RBA’s provision of banking services to the federal government. The Governance Board was still something of a hypothetical entity when I was there, so I can’t comment more on the inner workings of that. I think one thing to note that’s interesting about the RBA’s governance framework is that the RBA sits separately from the banking regulator in Australia.

There’s some communication at the staff level between the RBA and the prudential regulator. The collaboration at the senior level happens through this coordinating body called the Council of Financial Regulators, on which the RBA governor, as well as the chair of the Australian Prudential Regulation Authority, sits, and they communicate on matters relating to financial stability.

Beckworth: That is so interesting. In Australia, the setup is such that monetary policy, a central bank, is distinct from the main bank regulator, which is different than we have here in the US. The Fed has both responsibilities and, I dare say, probably the most important of the federal bank regulators. There’s many, but one reason given for that is, how can the Fed implement the discount window if it doesn’t have all the supervisory data, if it doesn’t know what’s going on in the banks? Here’s an example where a central bank is able to function. You have your own version of the discount window, the lending facility at the RBA, and they are able to do their job. I guess they talk to the bank regulator. It is possible. I know there’s other countries like that as well.

Bristow: Yes, that’s correct. I think the discount window in particular is interesting. There are two versions in Australia. There’s the overnight standing facility, and there’s what we call exceptional liquidity assistance. When banks access both of those facilities, in the past, there has potentially been some stigma around accessing those facilities. Part of that stigma is born from this question about whether the RBA should notify the prudential authority upon a bank drawing down on one of those loans.

Beckworth: Oh, interesting.

Bristow: That’s the case for the overnight standing facility. Then we collaborate quite closely with the prudential regulator on any loans given through the exceptional liquidity assistance facility.

Beckworth: That is really intriguing. Again, just another food for thought that you can have a central bank that is distinct from bank supervisory duties. I’ll just say one more thing about this before we move on. This has also become a political question in the US with the new administration. There’s this push to maybe move all the bank regulators into one place or to consolidate them. Some critics would say, it’s one thing to give the Fed budgetary independence. As you know, the Fed funds itself through seigniorage. It doesn’t need to go to Congress for appropriations. That’s fine because we want them to be independent.

Is it really reasonable to have bank regulatory duties also funded from that? Because every other bank regulator in the US government has to go to Congress, is accountable to at least the budgetary appropriation process. There has been this question, this gray area, but it’s just fascinating to see we have a case, the RBA, where they are distinct. I know there’s other countries that have done this as well. 

All right, so they have an inflation target, a range. I recently chatted with Raphael Bostic, the Atlanta Fed President, and he actually is someone who’s championed for the Fed to have a small range.

I’ve looked at a number of countries. A lot of countries have 1% to 3%. I do think a range would be useful. That’s where I would land on this. I think it allows central banks to be flexible to supply shocks. In terms of trade shock in Australia, when you’re a small open economy, I think it’s important to have some flexibility like that. Great example of central banking proactiveness. RBA is one of the first countries to adopt inflation targeting, the avant-garde, if I can call it that, of central banking. Now they’re taking the initiative in terms of thinking through their operating system.

I know the ECB’s had a review. The Bank of England had a review. Always exciting to see forward thinking, not just doing what you’re doing because it works well enough. Maybe well enough is not optimal. Maybe we could have a better world. It’s great. There are people at the Fed and Capitol Hill who are listening to the podcast. I listen closely to what Laurie has to tell us about insights from Australia, in addition to his cool accent and stuff. One quick question about Australia. Do you ever go to the outback and explore the animals and amazing things out there?

Bristow: It’s one state that I haven’t been to in Australia. It’s the Northern Territory, so I don’t have any stories from the outback.

Beckworth: Maybe that’s a good thing. We don’t want central bankers too adventuresome. We want them to be very calm and focused. Maybe I’ll travel there one day for you, and I’ll give a report back. 

RBA’s New Monetary Policy Implementation System

Back to central banking, though. You’ve described the RBA. Now let’s talk about the new implementation system. Walk us through the backstory. Maybe the beginning. Why did they feel there was a need to have a review?

Bristow: This is of the central bank monetary policy implementation system?

Beckworth: Yes.

Bristow: One thing to note about the RBA’s implementation system, it’s different from the Fed in that Australia didn’t experience really a recession during the GFC [Great Financial Crisis]. As a result, they didn’t conduct any quantitative easing during that period or at any point during the 2010s. As we were coming into the COVID pandemic, Australia was still operating in a scarce reserves regime. In response to the COVID pandemic, we had a gamut of unconventional monetary policies that we implemented. Those included bond purchases for market functioning. It included bond purchases to support the three-year yield curve target.

We also did traditional QE, where we just purchased a large number of bonds. I think importantly for the review of the framework, we also extended three-year fixed-rate loans to banks. Those are collateralized. Because we expanded the supply of reserves so much and the period of time for those loans was three years, as those loans were rolling off, we knew that the supply of reserves was going to decline and decline quite rapidly, even from a high level. It naturally prompted this question of: Prior to the pandemic, we had the scarce reserves regime. We’ve increased reserves by a lot, and we know that they’re going to drain in three years’ time.

Do we want to go back to the old system of scarce reserves, or do we want to keep reserves at a higher level on a more permanent basis? I think that was a natural point at which we asked the question of what should our framework be going forward.

Beckworth: Tell us about the framework review, the operating systems framework review. How long did it last? Who was involved with it? What role did you play in it?

Bristow: The framework review really started around the time that I joined the team in May of 2022. There was a lot of staff involved in it at the RBA. All the way from the analyst level to the governor, we had a series of presentations to staff, to inform them about the monetary policy implementation system and the challenge that we were facing. I think one thing that maybe is true of other central banks as well is that the level of education about the nuts and bolts of central banking tends to be quite low. Part of that process was educating staff on what the key issues were and having them weigh in.

We also briefed the board. I briefed the board on the state of demand for reserves at a certain point. Then, really, the review came in two stages. The first stage of the review was deciding what operating system do we want. Do we want a scarce reserve system? Do we want a demand-driven system? Do we want to stay in a floor system? Once we had decided to switch to a demand-driven system, there was then an extensive process of exactly what the details of that system were going to be. What is the price of our ceiling facility going to be? What is the tenor of operations that we want to offer? How frequently do we want to offer our open market operations? All of those nitty-gritty questions were part of the second stage of that review. We presented to the board for decision at both stages.

Beckworth: How long in total did the review take?

Bristow: We started in 2022. The first stage of deciding what system, it was announced by our assistant governor in April of 2024. Then the second stage of the review took a year, from April 2024 until April 2025.

Beckworth: Yes. I remember reading this in the news this year, April 2025. It was really interesting because I forget which official it was that made this speech and announced it. He said, among other things, that they would no longer be reporting the settlement balance rate. It’s still there, but where it used to be the anchor rate when they had the QE, the ample reserve system, it no longer was going to be important moving forward, even though it’s there in terms of anchoring that cash rate. The ceiling facility would be much more of the anchor moving forward, which I was like, “Wow, that’s remarkable. It’s amazing.” It took several years. It’s a deliberate, drawn-out process. Again, Fed officials, if you’re listening, let’s do this here. 

Maybe help us understand better this notion of a demand-driven operating system, because I think the RBA is probably the one farthest along on this journey. It’s a journey, to be clear. It’s not like overnight, you can suddenly turn on a dime and stop and do it. You actually have to get there. You have to shrink the balance sheet, implement these new tools, refine the existing ones, and such. 

The ECB has also talked about demand-driven system. Now, theirs is a little harder to understand because they talked about a demand-driven floor system, which I think that’s a little bit of contradiction in terms, although I know why they say that. Bank of England is also a demand-driven approach, but they’re all getting there on a journey. Help us understand what is a demand-driven system.

What Is a Demand-Driven System?

Bristow: I think officially, the RBA is calling its system a demand-driven system with full allotment, open market operations. I think the way that I like to think about it is it’s a ceiling system with a roof on top. In the past, in the corridor system, the RBA controlled the quantity of reserves and let the market set the price. That price that the RBA targets is the cash rate. What they’re doing in the current system is they’re controlling the price and letting banks determine the quantity of reserves that they want to source from the ceiling system. Another way to think about it is that the RBA is willing to lend theoretically an unlimited amount at a fixed price to its counterparties to put a ceiling on money market rates.

Now, the trick here is that the RBA’s open market operations are only offered once a week on a Wednesday. There are four days in the business week where this facility isn’t available to put a ceiling on rates. You could imagine that collectively, banks decide to borrow a certain amount from open market operations on a Wednesday. Then let’s say on that Friday, there is a big tax take and the supply of reserves drops in a way that banks didn’t expect when they made their decision to borrow from open market operations. That could cause some volatility in money market rates. It could even cause money market rates to go above the rate on the RBA’s ceiling facility.

That’s where this roof-on-top concept comes in. That’s where we have the overnight standing facility. The overnight standing facility is priced above the rate on open market operations, and it offers overnight funding for banks who need it in this interim period between Wednesdays. That’s the ultimate ceiling on the system and stops money market rates from becoming too volatile.

Beckworth: You have to have some robust ceiling facilities, and you have your regular one, then you get one above that if there’s stress in the markets or heightened demand. Maybe that’s a better term. Stress may be too strong, which is interesting.

Bristow: I think one of the things that is fantastic about the RBA’s system is the lack of stigma at both of these facilities. One of the things that the RBA, maybe it’s luck but maybe intelligent design, is when we are in the floor system, after all of the unconventional monetary policies, we never stopped doing open market operations. We never had to reintroduce a new facility when we were transitioning back to a demand-driven system. The RBA kept offering its open market operations even when the supply of reserves was extremely high. I think one of the reasons they did that was they wanted banks to continue to regularly borrow from that facility and to maintain a lack of stigma in that facility.

Now, it’s true that the overnight standing facility hasn’t been used in the same way. Historically, there’s been some stigma attached to borrowing there. One of the things that the RBA has done as part of its announcement of its new system is it’s released a joint statement with the Australian Prudential Regulatory Authority that says, “We want banks to use the overnight standing facility in the regular course of business. We won’t think badly of banks for doing that because it’s important for the RBA’s control over monetary policy.”

Beckworth: So interesting. A few observations here. One, I like how you frame this. Basically, this is a demand-driven ceiling system. That, to me, is the nice language it encapsulates. Demand-driven ceiling system, but with a roof. It’s an interesting contrast to the floor system, but guess what, it’s a leaky floor. There’s at least in the US, a leaky floor system, so we had to have the overnight reverse repo. It’s like the same idea, but on the very top. On the bottom, you had a floor. You had the administered deposit rate, the central bank in the US, interest on reserves. It’d be the settlement rate, I guess, and RBA.

In the US, it proved leaky. Rates would slip through that floor, and so you had to put something to catch it. Now we have something on the top. We want to have robust—it’s not a corridor system—but a robust interest rate corridor at the top, at the bottom. That’s cool to have the flip of a leaky floor, just in case you want to have a robust roof on top of it. Demand-driven ceiling system. I’ve become enamored with this. I’ve had severalposts on mySubstack. Listeners, go check them out if you haven’t already. Here’s a couple ways I’ve thought about this. I want to get your response.

Advocates of the floor system say, “Look, one of the most beautiful things about the floor system or the ample reserve system is that there’s ample liquidity.” Governor Waller says, “Why would you want a system where banks have to go look in their cushions for spare cash or coins?” The idea being you’re trying hard to find some liquidity and you don’t have enough, so you’re looking in all places. That’s how he describes scarce reserves as opposed to ample liquidity in a floor system or ample reserve system.

To me, I think you can answer that question with a ceiling system. I would say a ceiling system has latently ample liquidity. If a bank needs it, they have no fear of going to the facility, whether it’s lending or repo facility, and say, “Look, I want it.” As you said, there’s not the stigma there. It’s a normal part of business. Whether you have in the floor system explicitly ample liquidity or in a ceiling system, you’ve got latently ample liquidity; you get it both ways. I think the costs with a ceiling system are far less than some of the costs with a floor or ample reserve system. Any thoughts?

Bristow: I think it’s a great point. I think one of the advantages of the ceiling system is that there’s this tradeoff that exists between banks having a lot of cash and them having the incentive to lend that in private markets. I think the thing that’s nice about the ceiling system is that banks know they can borrow from the Reserve Bank at their ceiling system if they need liquidity. That, if it’s priced appropriately, means that banks generally have enough cash to operate. One of the things that’s unique about the design of a lot of these ceiling systems is that the facility they offer is only once a week.

In that intervening period, even if system liquidity is ample, there can be some volatility. I think that volatility is important for inducing the private market activity that is valuable.

Beckworth: Fascinating. You want some price discovery. You want markets to do their thing, but you want to be there as the backstop should it get too volatile. You want some volatility, not too much volatility. So fascinating. Again, I like this term. I coined it when I was thinking about this initially, when Bill Nelson was here. We were talking about the changes at the ECB and the Bank of England at the time, and then RBA. To me, it’s latently ample liquidity in a ceiling system.

Even some of the proposals here in the US were beefing up the discount window or better use of standing repo facility, the idea is, for example, to use your collateral at the discount window to count toward, say, the liquidity coverage ratio or some other regulatory thing. You don’t have to have electronic money on your books, but you know you have it implicitly, latently at the Fed, if you need it. Of course, the problem in the US is there’s stigma. Maybe these facilities aren’t designed as efficiently as they are at the RBA.

Again, I challenge our listeners, don’t get hung up on, oh, but we have explicitly ample liquidity. We don’t need to go dig in the couch for change as a bank. You know what? We don’t need to dig in the couch because we have a facility that gives us latent ample liquidity. Anyways, I belabor that point. 

Interbank Market

Any other thoughts on the RBA system? You said you created a roof to create just enough volatility so there’s some price discovery. Let me ask this question. Does that suggest that there’ll be some interbank overnight unsecured lending? That’s basically dead in the US. The Fed funds market is a shell of its former self, GSEs and foreign banks. One of the critiques is, why do we use it? In fact, Lorie Logan has said, “Let’s move on to a repo rate.” I know some people aren’t as excited as she is about that. I understand her point because what is the Fed funds rate? It’s not really an informative price. Is the RBA’s push to have a little bit of volatility to make that interbank signal have some value?

Bristow: I think this is a great point to distinguish between the central banks that have announced that they’ll be operating a demand-driven system. If you look at, for example, the Bank of England, they have a demand-driven system, but the price of their ceiling facility is equal to the rate that they pay on their reserve balances. What that really means is that they’re in this demand-driven floor system because the price or the opportunity cost of holding reserves is close to zero. There’s no incentive for banks to lend out that cash or redistribute those reserves amongst themselves at the end of the day in the interbank market. I think the RBA is on the opposite end of the spectrum in that initially, our price under the ceiling system was 15 basis points above the rate on reserve balances. In April 2025, we increased that rate from 15 basis points above to 20.

The opportunity cost of holding reserves in Australia is much higher than it is at the Bank of England. I think that there are really two reasons that we did this. I think number one consideration for us was just the size of the balance sheet. One thing that’s interesting to note about Australia is that the size of the repo market is much smaller than potentially in other places. Even when we were operating under the scarce reserve system and only injecting $2 billion to $3 billion of reserves, the RBA was 30% of the repo market, which is quite a large footprint.

You can imagine that under a system where you’re going to be operating a larger balance sheet, the RBA could potentially be 50% or 60% of the repo market. I think that was our sense when we were looking at demand. We were thinking our balance sheet, we might have to inject up to $200 billion of reserves, and we want to be able to decrease the size of our balance sheet. The other option was to keep the size of our balance sheet large and use other asset purchases to keep reserves high. That was just going to entail a lot of risk on the RBA’s balance sheet, which was something we weren’t comfortable with.

I think you raise a great point about the interbank market. In Australia, in particular, it’s very important because our target for monetary policy is the cash rate. The cash rate prior to the pandemic only had about $4 or $5 billion worth of volume each day, and that volume dropped to zero on a lot of days when we were operating the floor system. Australia doesn’t currently have an alternative to switch to, so we don’t have a benchmark rate that is SOFR. There is a beta rate that’s being produced by the Australian Securities Exchange, which we’re calling SOFIA, but it’s not live, it’s not operational.

The RBA, I think it needs to engender some volume in the cash market just to keep its benchmark rate for monetary policy robust. I think that was one of the reasons why we chose to increase the price of our ceiling facility. It was to induce more activity in that market and to allow banks to redistribute reserves amongst themselves at the end of the day, rather than supplying more liquidity.

Beckworth: Big changes are happening at the RBA. Now, I mentioned earlier, this is a journey. Have they finished the journey, or are they still on the journey?

Bristow: It’s hard to say whether we’re finished with the journey or still on the way. One thing to note is that in the corridor system prior to the pandemic, the cash rate target was set 25 basis points above the interest on reserves. Our ceiling system is currently at 20 basis points above the interest rate on reserve balances. We’re very close to where we were prior to the pandemic in terms of the cost of reserves in the system. It’s just that we’re picking the price now and not the quantity. I think we’re probably done for now until we see the supply of reserves decline further as the bank’s bond purchases roll off.

I think the RBA will continue to reassess whether it’s hit its steady state in terms of the size of its balance sheet, the amount of private market activity it wants to see, as well as its own footprint in financial markets.

Motivations for a Demand-Driven System

Beckworth: Laurie, you just alluded to another central bank who’s doing a demand-driven system, although their spread is much smaller. It is interesting to see this general movement among many central banks. I think the ECB counts, the Bank of England, Bank of Canada, to some extent, the RBA. Here’s my question. How did this all start? It’s almost like this movement is happening, right? Obvious answer would be, QE, how do we deal with QE? How did they all land on a similar journey? Again, differences, but what do you think? Why did they all want to go to a demand-driven ceiling?

Bristow: I think the biggest motivation for switching to a demand-driven ceiling system is that during the country’s experience with the floor system, they found out that the demand for reserves is highly uncertain. It’s difficult to estimate, and you can run up against the steep part of the demand curve without knowing that it’s going to happen. I think when I was at the RBA, the story of the Fed in September 2019 was repeated often as a cautionary tale for other central banks. I think one of the important things to note in that discussion is that people were quite worried about the skewed distribution of reserves amongst banks.

I think that point is basically that even if you have a large buffer above demand, it still might be the case, because the distribution of reserves is skewed, that some banks need to borrow into bank markets to replenish their reserve levels. When they try and do that, they’re accessing a market that has largely withered as a result of a lack of activity. It takes banks and other market participants a while to reboot those systems to increase their credit limits with their counterparties, for example.

Even when the supply of reserves is above what you estimate and maybe even at the top end of your range for where you think the demand for reserves is, you can still get volatility in money market rates or spikes in money market rates, which means that you lose control of your monetary policy. I think that was the story that we told ourselves was that we don’t know where demand is, we think it’s increased, and we don’t want to experience this volatility that the Fed experienced in September of 2019. The demand-driven system solves this problem to some extent because you allow banks to expand the supply of liquidity automatically when they need it.

I think that’s the sense in which central banks have all decided to operate a demand-driven system, because I think this problem of estimating demand and demand shifting around over time is common to all of the jurisdictions.

Beckworth: That is a profound point. The critique of the scarce reserve system given by proponents of ample reserve is, you’ve got to forecast reserve demand every day. That’s tough. In ample reserve, you’ve just got to push the reserves out there and you’re one and done. Your point is actually, no, it’s a lot more complicated because you don’t know where that curve goes steep. You don’t know where you flip back from ample reserve into scarce, and every day you’re like, “Are we there yet?” There is a huge amount of uncertainty even in an ample reserve regime.

Moreover, as Bill and others have noted, you can’t just say, inject a certain amount. Say you inject $500 billion through QE into the US, and, oh, we’re done because what happens is the reserve demand, the structural part of it, grows over time. That $500 billion may not be enough at some point. That may only be enough to get you back into the scarce reserve. Your whole point is it’s not so simple. Ample reserve system is not so simple as advocates make it out to be. Whereas ceiling demand, you let the banks figure it out for you.

Bristow: I think that’s right. In the corridor system prior to COVID for Australia, there’s this argument that it was hard to operate. You had to forecast the supply of reserves every day. I think that is true. Sometimes we got it wrong and we would have to go in and inject more reserves in the afternoon. You didn’t really have to forecast demand because it was very stable. It didn’t move around a lot. You really would just aim for a constant level of reserves. Whereas in the floor system, as the level of reserves is declining, you have to forecast where demand is going to be, which is very hard. It’s highly uncertain. You also have to keep all of that infrastructure for forecasting the supply of reserves because you want to know if supply is going to dip close to your estimate for demand. You’re actually doing twice the work. You’re forecasting where demand is.

Beckworth: That’s a great point.

Bristow: You’re forecasting where supply is. There’s more uncertainty over demand in that system.

Beckworth: That is a profound point again, is, in some ways, it’s even harder, the ample reserve. The knowledge problem is sometimes even more pronounced. This one last point, what you said about sometimes even in an ample reserve system, reserves get concentrated in some banks while other banks need it. You can have volatility. You can have problems. Whereas in a ceiling system, demand-driven, each bank knows what it needs. It goes to the facilities and it’s satisfied. On so many levels, it seems to me like, I don’t want to say a slam dunk case, but man, a no-brainer.

Demand-driven, it provides latent ample liquidity. Banks know when they need to go. Now, of course, the critics will say, “The devil’s in the details, David, because if you don’t have good ceiling facilities, which is fair, then this is just a pipe dream. Maybe it is a slam dunk case on paper, David, but in reality, you’re dreaming, man. That’s not going to happen.”

Bristow: I think there’s one thing to note about the slam dunk case, and that’s that you’re relying on each individual bank to borrow from your ceiling system, but you’re also relying on banks collectively to borrow enough reserves. Because there’s not so much experience with demand-driven systems in the current environment, there’s this critique that maybe there’s a collective action problem. Maybe banks expect to borrow from other banks instead of from the central bank. Because of that, maybe there’s not enough reserves in the system or the banks collectively don’t borrow enough for money market rates to trade around where you want.

That’s one of the things that I think the RBA will be monitoring for as it transitions into its demand-driven system, is just how well banks are going to forecast their needs, how well banks are going to forecast the supply of reserves in between open market operations dates, and how well they solve that collective action problem of who’s going to borrow to increase the aggregate supply of reserves.

Beckworth: If there were a shortage of reserves, aggregate level, wouldn’t, by definition, banks go to the facility to satisfy that? If rates are starting to go up, if I’m a bank and there’s a shortage overall, yields are going to start to go up. Guess what? I’m going to go to the facility where they’re cheaper. To me, it’s like an automatic fix. Am I missing something?

Bristow: It is in a longer-term sense, but you have to remember that the open market operations that Australia and other central banks are doing are only once a week. I think that’s where there could be some sticking point, is if they forecast reserves within the week to be higher than they end up being, or the realized number is that there’s some intraweek volatility. That’s the thing that potentially is worrying, and that’s why RBA has this roof to control— 

Beckworth: That’s why this is still a journey.

Bristow: That’s why it’s still a journey and a learning process.

Beckworth: We’re still going to have to figure out how often should we engage in these operations? How big should the spread be? All these details need to be fleshed out because we really don’t have any data points. We have a great theory on paper that I’m terribly excited about, and I’m so delighted to see the RBA do this. 

Bank Policy Institute Money Market Symposium

Let’s take these ideas, as awesome as they sound to me, and apply them to the Federal Reserve. I want to do that first by talking about a symposium that BPI had, a money market symposium or conference. You guys were all a part of it. I understand these issues came up; these very issues came up. Tell us about that event.

Bristow: We had an event. There was a bunch of market participants, some Fed staffers, myself, and Bill involved in the discussion. I think the discussion was mainly around developments in money markets recently and the Fed’s steady state implementation system and balance sheet. The context for the discussion is that repo rates have recently risen above the interest rate on reserve balances. In some cases, there’s been trading above the rate on the standing repo facility. The question out there is, will the Fed increase the supply of reserves? Will it respond to this action? Will it begin reserve management purchases going forward?

There was a lot of discussion around the standing repo facility and how effective that tool has been in capping repo market rates in the current environment.

Beckworth: People acknowledge the problem, though. This is a great example. Right now, repo rates are above. Guess what? They aren’t using the ceiling facilities like we would want them to. We don’t have all of our ducks in a row, so to speak, to even maybe begin thinking about the demand-driven ceiling system. We would want to see something down in the discount window. Bill’s had a lot of great suggestions. Again, I’ve championed Bill’s views. He has suggested bringing back something like the term auction facility, like TAF, so you regularly auction—that’s what the RBA does—this full allotment, you set the price, and you can come get as much as you want.

Maybe we do something like that so it becomes normal operating procedures. In fact, during the Great Financial Crisis, when the Fed did it, banks took it up and it had a different name. It wasn’t the discount window. Maybe that’s part of it. Also, Bill and others have argued having collateralized lines of credit at the discount window so you can tap into those, use those for liquidity requirements, and just make it more normal to be using that. That would be the discount window standing repo facility. There’s been proposals to have it open more often. I know they’ve made some changes recently.

Also, there’s been talk about bringing in central clearing to the standing repo facility. That way, the standing repo facility could have more counterparties, at least on the other side of that central clearing entity in the middle. Were these issues also discussed?

Bristow: Yes, I think mostly related to the standing repo facility and less so around the discount window. I think the interesting things that we heard from market participants around the standing repo facility is that there is still some stigma surrounding borrowing from that facility. I think that stigma has two sources. One is that primary dealers don’t want to explain to their CEO that they’re borrowing from the Fed. The other one that was mentioned in the symposium was that occasionally, participants get questions from their credit rating agency representatives about why they’re borrowing from the Fed and why they aren’t using other private market sources.

I think that’s the first time I’ve heard of that specific argument. I think my takeaway from that discussion is that there is still stigma surrounding borrowing in the standing repo facility despite record volumes in that facility in recent times. I think there were a number of suggestions offered by participants in the symposium to reduce stigma of the standing repo facility. One option that was mentioned is just to reduce the price to below the top of the target band to encourage usage by market participants. One solution that was offered also was just like you say, recast the language around the facility to say that it’s for monetary policy implementation purposes, and it’s not a backstop on rates.

Another technical solution that was offered would be to move the auction that’s in the morning to be even earlier, to be closer to where a lot of the volume happens early in the day in the repo market, so that participants can use it in their regular course of business rather than after most of their business has taken place. 

One other solution to stigma at the standing repo facility that we heard would be to separate participation between banks and primary dealers. I think it would be easier to explain taking funding from the standing repo facility for primary dealers if it was only primary dealers that were a part of the facility, partly because it feels a lot more like open market operations did back in the day and it’s borrowing in the regular course of business.

Beckworth: One of the big issues that I see for making the standing repo facility more useful is the fact that the US financial system depends less on banks; the European system is bank-driven. Here, there’s a lot of banks, but there’s also a lot of nonbank financial firms, credit creation going on. You need other entities besides the banks to help intermediate what the Fed is doing when the Fed wants to inject liquidity. To me, it seems like it’d be especially important to bring in that central clearing because then it would allow, I daresay, hedge funds or other private credit creation entities out there that are playing a role.

Maybe there’s some concerns about them getting access to the Fed’s balance sheet, but the central clearing is one way to address that. Isn’t that a big issue too, that the very structure of the US financial system is not just banks coming to the Fed, it’s you need these other financial intermediaries to also play a role?

Bristow: I think that’s right, and that was one of the solutions that was discussed in the money market symposium. I think it’s one of the things that was mentioned in the latest FOMC minutes as well was potentially opening up the standing repo facility to be centrally cleared. I think there was some discussion in the minutes about potentially increasing systemic risks surrounding the central counterparties as a result, and having them more intimately connected to the financial wiring and also increasing Fed’s footprint in financial markets through that. I think that solution is less of an immediate-term solution and maybe more of a medium-term solution for the Fed because it would take some time to implement.

Beckworth: Fair point. Right here and now, we got repo rates above the target, and so we need to do something about that. Long-term, structurally, it might be a fix to do it. All of this is important and also interesting to people like you and me, as I mentioned. We like to nerd out on these issues, but also, there’s politics, there’s personalities at play. To me, I was really surprised to see, as I mentioned earlier, on September 26, Vice Chair Michelle Bowman gave a speech where she literally said she would like to see a return to scarce reserve system. Let me just read an excerpt from her talk, why she would want to return to a scarce reserve system. Again, I think she’s maybe the only person explicitly saying this, other than Steve Miran kind of hinted at this in his talk that he gave just recently.

She goes, “First, a smaller balance sheet would minimize the Fed’s footprint in money markets and in Treasury markets. Second, holding less than ample reserves would return us to a place where we are actively managing our balance sheet. Identifying instead of making signals of market stress, in my view, actively managing our balance sheet would give a more timely indication of stress and market functioning issues, as allowing a modest amount of volatility in money markets can enhance our understanding of market-clearing points.” 

Which is what the RBA is striving for. They want a little bit of volatility, so there’s some price discovery. If needed, maybe the red alert sign goes off. Again, I think the ceiling system would also be better. I guess what I’m saying is let’s not stop at scarce reserve. Let’s go all the way to something like a demand-driven ceiling system.

Bristow: I think it’s an interesting point. Maybe what you’re touching on is, is there a difference between a corridor system and a demand-driven system at the same level of reserve? What’s better? Do we want scarce reserves with a corridor system, or do we want a small number of reserves equally, but control money market rates with the ceiling system?

I think if you’re going to decrease the supply of reserves to similar to what they were in a scarce reserve system, but you were going to operate a ceiling facility, I think the one change that you might have to make from something like what the RBA and the Bank of England are doing is that you might have to operate your ceiling facility on a greater frequency, rather than once a week, and allowing reserves to fluctuate through that week and create that volatility. If you were going to operate with a very small number of reserves, there would be some volatility inherent in having the supply of reserves quite small.

That would be enough, I think, to induce some activity between banks, and then you could offer that facility every day. Then you would also see that activity in that facility would give you some indication of where the market-clearing price is as well.

Beckworth: Those are great points and food for thought in terms of if we ever do get to the ceiling system. Michelle Bowman, Vice Chair Bowman, gave this speech in September. A little bit earlier in August, President Lorie Logan of the Dallas Fed also gave a talk where she said, “Look, if you want to go smaller, you’re going to have to get those ceiling facilities working.” She came up with a bunch of proposals for making the discount window standing repo facility more robust. I’m very sympathetic. Now, to be clear, President Logan later had a speech in October, so after Governor Bowman’s, but President Logan had a speech titled “Ample Liquidity for a Safe and Efficient Banking System” where she went full-court press making the case to keep the ample reserve system. 

Then just recently, Governor Stephen Miran, on November 19, had a speech titled “Regulatory Dominance of the Federal Reserve’s Balance Sheet,” and he made, again, a lot of the points that your colleague Bill Nelson has made about the regulatory issues on constraining the flexibility of the Fed’s balance sheet to adapt. Any thoughts on Governor Miran’s speech?

Bristow: I think the focus of Governor Miran’s speech yesterday was at the intersection of banking supervision and monetary policy implementation. Maybe I can offer an anecdote from Australia as well. I think a similar dynamic happened in Australia as happened here, except in different contexts. When the supply of reserves was abundant in Australia, it was around the 2023 crisis in the US, around SVB. One of the things that came to light was just how quick bank runs can be and how much potential liquidity banks need to facilitate that.

I think one of the things that the Australian Prudential Regulatory Authority asked Australian banks to do was to reconsider their monetization risk over a short period of their high-quality liquid assets. I think what that did is it made banks really reconsider how many reserves they had to hold. Really, what that is, is supervisors telling banks to hold reserves independent of decisions around the monetary policy implementation framework. I think he’s right in his characterization of the issue where supervisors are telling banks to hold reserves. The central bank, at least in Australia, was late to that conversation.

Beckworth: Maybe that’s the argument for keeping supervision at the Fed so they can have some control or say so. It’ll be interesting to see what Vice Chair Bowman’s new push to downsize and reconfigure bank regulation. I wonder if she’s going to do anything about this issue, too, in order to help monetary policy implementation. 

With that, our time is up. Our guest today has been Laurie Bristow. Laurie, thank you so much for a fun conversation on central bank operating systems.

Bristow: It’s a pleasure, David.

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.