Manmohan Singh on the Meaning of Money after the GENIUS Act

Manmohan Singh is the editor-in-chief of the Journal of Financial Markets Infrastructure. Manmohan returns to the show to discuss whether money still matters, the impacts of the GENIUS ACT, the lobbying show down over stablecoins in the US, stablecoins impact on the Eurodollar market, and much more.

Subscribe to David's new Substack: Macroeconomic Policy Nexus

Read the full episode transcript:

This episode was recorded on September 10th, 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 Manmohan Singh. Manmohan is a former IMF official and is currently the editor-in-chief of the Journal of Financial Markets Infrastructure. Manmohan joins us today to discuss the meaning of money in the digital age, its implications for the financial plumbing system, and for central banks. Manmohan, welcome back to the show.

Manmohan Singh: Thank you for the opportunity. Again, last time we’ve met over COVID, so this is one-on-one now.

Beckworth: It’s great to meet you in person here in the capital. We just work a few miles apart from each other. You were at the IMF, and I’m here in Arlington at the Mercatus Center. You recently stepped down from IMF. Are you excited for this new transition in your life, this new stage of your life?

Singh: Yes. It’s been a good 26-plus years. I saw most of the world, how money works and central banks work, and all sorts of it. It’s a very big rubric, the IMF. You could have a job which is so different from the previous job. Yes, I would like to move on. I would like to write a book on some of the themes I’ve been doing. I think if time permits, I should have one out by Christmas.

Beckworth: Fantastic. Now, I got to know you and your work with Peter Stella. The two of you work together a lot and did work on collateral, on money. Peter is also a returning guest of the podcast. I love the fact that you have actually traveled the world. You have visited multiple central banks. You have looked at balance sheets. You’ve looked at collateral. You have dealt with the infrastructure of the financial system in many places around the world. You’re someone who cares deeply about the pipes of the system. Last time we talked, I remember we talked about how some of the pipes can be clogged, not operational.

Singh: Or rusted.

Does Money Still Matter?

Beckworth: Or rusted. Collateral is a key part of the story here. We’re going to get into that. A lot of this is motivated by my interest in stablecoins, what is moneyness. Let me begin there. Does money still matter in this digital age? What is moneyness? Does money still matter? 

Singh: Yes. Given where the world is heading and some of these topics which you might trip over, I’ll take money a step further because I’ve written on moneyness. There was a very nice piece in FT with one of my colleagues, Phil Prince, which basically talks when you look at moneyness—and moneyness is not a bipolar issue, it’s a spectrum—when you break down moneyness, an integral part is collateral services is offered, the way collateral is handled.

In QT, when you take money into the system and release collateral, the demand for the collateral, the moneyness in the collateral you release may be more than the money you bring into the central bank balance sheet. I have always looked at moneyness, and I think the way the world is heading backing stablecoins, as you mentioned, with collateral, you are getting into the moneyness world.

Beckworth: I believe last time we chatted, we brought up this very point that you’re alluding to, that QE was supposed to stimulate the economy by taking out long-term bonds and lowering yields, but you brought this other point. There’s a counterforce that it does. It also removes important collateral so that there’s less transaction assets, less velocity. On one hand, we look at the surface effect that we observe, but we don’t see this backdoor plumbing issue that’s being adversely affected by QE.

Singh: Exactly. There are times, when you remember the eurozone crisis, acute shortage of bonds, there are times when moneyness is not in favor of collateral. That’s why I said, ISLM or whatever framework you pick up, LM curve may pivot, and the pivoting could be in a way where you actually extract moneyness by such operation rather than put in money. I’ve never said QT was reversed.

I always was in favor that if you went for QE with a big number, and yes, you did drop the long end, the QT steps were very puny. If you wanted the reverse, then you need to go with that. I think the Fed started with $45 billion, they made it $90 billion, even ECB, because I think you didn’t get much for it. I think it was a slow trickle, and I don’t know how the markets felt about it. There were too many things going on with QT. Then the COVID comes and this and that. It’s very hard to do analytical and good econometric work because there were too many changes. QE happens, QE2 happens, QE3 happens, then you start QT, and then COVID happens. It’s very difficult, but intuitively, I think these were small steps. Again, all this is experiments. We were never taught QEs and QTs.

Beckworth: Exactly. I was just thinking even the new Federal Reserve framework that was just announced at Jackson Hole a few weeks ago, doesn’t even mention balance sheet policies. It mentions the inflation target. It mentions what it’s going to do. Balance sheet policies have become so regular now, and we’re learning these experiments. They didn’t include that.

Going back to the Great Recession, this is really where it all starts, and I believe your work became really important is it wasn’t just QE. It was that the Great Recession, we lost a bunch of collateral. A bunch of mortgage-backed securities disappeared. A bunch of assets that were used as collateral, they disappeared. Then we pulled out other collateral through QE, large-scale asset purchases, and we put some back into QT. We’re just figuring this all out.

Let me ask this question here. One of the big concerns in your work is there wasn’t sufficient collateral or the right type of collateral. Are we in a better place now? There’s been a lot of debt issued. Do you feel like the world’s in a better place in terms of the amount of collateral, or are we still facing challenges?

Singh: The world is a big place, and oftentimes the collateral world may be regionally very different. The Fed did not have as much acute problems as the eurozone had when they came after the eurozone crisis. I know they changed the SEC lending policy. They made sure ECB and the national banks released the bonds. If I understand correctly, there was very good margins to be made, and many of the national banks, Belgium, they did quite well on releasing the bonds.

Monetary policy, collateral policy differs. For example, in Japan, there are days where there’s no JGB activity, zero. What I call, if you really want to understand where collateral velocity is zero, look at the JGB market because everything gets stuck in the prefectures. They have tried to improve that too. Even in the Asian time zone, my understanding is US Treasury is more liquid than the JGB market. Things are changing.

Before COVID, they have tried to see what else can replace the US Treasury market. They want to make it more Asiatic thing. I have seen memos where JGBs could be swapped with the Hong Kong dollar. JGBs were acceptable in Hong Kong accounts, and because of the peg, Hong Kong dollar is more amenable with the Tokyo financial market. 

Beckworth: It’s complicated.

Singh: It’s complicated, small steps. When you get used to it, the inertia with Treasuries and dollar is there. Geopolitics are changing. I don’t know where the world is heading. 

Beckworth: Most of our listeners will know what JGBs are, but for those who don’t, that’s Japanese government bonds, which are an important asset in the global financial community. Your point was the Bank of Japan had bought a lot of them up, and there wasn’t much activity. The market’s relatively flat, inactive, and that’s a problem.

Singh: That actually hurts JGBs. On the other hand, when you see Chinese Govies trying to make it into LCH one way or the other, it’s not a trivial topic, but I wrote a paper where in global plumbing, can we see new entrants? 

Beckworth: That’s the London Clearinghouse.

Singh: Exactly. Just like you have the SDR basket, which picks and chooses the most blue-chip currencies—dollar, euro, Swiss, even the Japanese yen—similarly, in my view, and I don’t think I am exaggerating, if LCH accepts collateral, I think it’s one of those pristine things because you have to go through filters. LCH, unlike, say, the eurozone crisis matrix, had very different haircuts than the eurozone. Very, very different because if something goes wrong, that is one of those too-big-to-fails for the UK. It’s like three-big-to-fails. It’s probably the biggest pocket out there, even probably bigger than Barclays if something went wrong.

Beckworth: A lot of systematic risk wrapped into the London Clearinghouse, and you’re saying that they’re now looking at Chinese collateral, government bonds.

Singh: Initially, Euroclear may be amenable with the Chinese Govies in dollar or in euros, but I could see that if some of the geopolitics world, mBridge and all, I don’t want to get into this stuff, but there’s a lot happening on that front, if Chinese renminbi makes it to LCH as legitimate collateral. Other emerging markets have tried it. I don’t think South African rand made it, but if Chinese Govies, you often start reading Chinese Govies, or at least the renminbi may need more appreciation. There have been recent blogs on that. 

You see, the number two position after US Treasuries, Chinese, Japanese, and Italian bond markets are very, very big. They’re all close to each other, so I’ll call them number two. Can you imagine if that makes it to the bucket? Italian bond, BTP, made it thanks to the ECB matrix. Not on its own. Not on its own, but the structure of the payment system out there. Here, if you make it, that’s a market decision. I don’t know when that’ll happen, but it could happen, and that will change. If you think about stablecoin topics, and if the Chinese wants to issue the stablecoins, a stamp from LCH will go much more than anything else I can think of.

Stablecoins

Beckworth: Wow. Not only could we fall behind in AI and electric cars, we might fall behind in collateral issuance to global markets. There should be a shot across the bow for US policymakers, if you are listening. Look closely at the developments with the London Clearinghouse, what they might accept, the Chinese stablecoin. Let’s talk about stablecoins. Let’s jump in. This is why we were talking, I think we were talking actually online, and you had maybe seen some of the other podcasts we did on stablecoins.

Really, what’s driving a lot of the conversation is the GENIUS Act. It’s been passed in the US. I’ve had several guests on, I’ve talked about how that is really going to supercharge an already first mover advantage to dollar-based stablecoins versus other currencies. It’s a growing market. It’s a large market already. Tether, US Coinbase, also the two largest ones, but Tether’s the largest one. Of course, that’s outside the US, but they’re dollar-based. The GENIUS Act of 2025 is passed, so it brings all these stablecoins within the regulatory perimeter. That gives them certainty. They can participate. They don’t have to worry about regulators changing on them, rules changing on them.

I want to bring this up because it relates to collateral, it relates to moneyness. Let’s begin with a list of assets that can be used to back up the dollar stablecoins. There’s got to be a one-to-one backing. I’m going to bring these up because then I want to talk to you about the collateral implications of these things as we go forward. According to the GENIUS Act, you can have your stablecoin—and just to be clear, a stablecoin, it’s very cost-effective. It’s peer-to-peer, low transaction costs. It’s effectively a transaction asset that someone views as a dollar, just like using a dollar bill, but it’s digital.

While they can be backed up by our US coins and currency, demand deposits, which was interesting to me, and I was really surprised to see that, but I also read in the details there are limits to how much you can back up with a bank account. The FDIC can set limits on how much of a stablecoin is backed up because you can imagine a stablecoin that exceeded the FDIC insurance deposit limit. You don’t want a bunch of uninsured deposits funding or backing the stablecoins.

US Treasury bills, notes, and bonds, I was a little surprised to see bonds in that list, but they said they have to have 93 days of maturity or less. Basically, Treasury bills, that’s what we hear mostly about. Repurchase agreements backed by Treasuries can also be a part of the assets backing that. Reverse repurchase agreements, similarly. Money market funds. Then central bank reserve deposits directly. That’s an interesting conversation.

You’ve written about this, but can you imagine a stablecoin issuer having a Fed master account? I know the Fed’s been very reluctant to do that. We had the narrow bank, it’s not quite the same, but it wanted to literally have an account and arbitrage the interest on reserves. Basically, you have a deposit, goes straight into the master account. Do you ever foresee that happening, the Fed ever allowing a fintech, like a stablecoin, to have a master account? Because that’s been so contested. There’s a lot there, I know.

Singh: There’s a lot there, so let me take it one by one. I’ll take your first question first. There are a lot of litigations. You mentioned TNB. There is the Wyoming Custodia there. I think if you go to one of those websites, there’s a lot based out of Puerto Rico. There’s just not one, two of them. Getting a master account, I think the state of Texas has a master account. A lot of people said, how come that state has it and some others don’t have it?

Getting a master account doesn’t mean LOLR, lender of last resort. Let’s make that clear. You may get a master account. The politics here are changing. It’s a very dynamic world. Who’s getting what out here? In fact, I remember a couple of years, two, three years ago, somebody did get a master account and it was U-turned, partly because of some of these issues about, I won’t say legit or illegit. How do you make decisions? Discretion is the right word.

I want to step back because this is politics. This is what’s happening out there, but it won’t be LOLR. This question actually gets into stablecoins and their model. Will this model work at zero interest rates? Because we have had zero interest rates. At zero interest rates, unless your business is very well diversified and you have something else like, okay, I don’t make money on the seigniorage from the HQLA, I may have custody. I may offer custody and make a few pips on the custody.

In fact, recently, one of the large stablecoins wrote a letter to FASB that stablecoins and T money market funds, tokenized money market funds, I don’t want to get into that, but basically saying that, could that be deemed as cash equivalent from an accounting purpose? Cash equivalent, that’s the word. This is another thing. What I’m trying to get at is there is demand to get some interest, knowing very well that this interest rate cycle does touch zero once in a while, who knows when.

Then these stablecoins, right now, they are in a very juicy 4% or 4% plus environment. The fact that one of them tried to go through BlackRock and dip into the RRP. Remember, RRP, IOER, these were done for MonOps, monetary operations reason, strictly speaking. RRP was done because of the leakage from IOER, the floor as well so they made sure Fed funds rate remains contained. The Fed funds rate is the right one with diminishing volume, but that is the modus operandi of the Fed right now.

There is a lot of traction master account. If you get master account, what do you want after that? Do you want IOER? Do you want RRP? If I remember correctly, Bank of England had a draft paper and they said, no, for fintechs, you may have a central bank account, but there’s no interest on excess reserves. That gets into where MonOps comes in and where the use or abuse of the MonOps come in. Remember, RRP was never to be there forever.

Beckworth: Yes, it was supposed to be temporary.

Singh: It was supposed to be temporary. Now you cannot use RRP through money market funds. There’s a lot going on in this space but my take is because rates can touch zero, I do not know whether stablecoins world will do. They are trying to preempt that scenario where they may be hitting zero. There are attempts, there are definitions, and I want to come back to something more simplistic. You and I may have read a long time back what was money. It had some basic characteristics, noninterest-bearing, privacy, and the third one was uncapped.

These were basic things. Okay, the world has moved on. I remember the paper, one of my grad school papers was Kocherlakota’s 1996 “Money Is Memory.” It’s a beautiful paper. You don’t see these papers. It survives. 30 years later also, I sometimes say, what does this mean? If your memory, if you had infinite memory, now we have an AI world. There was no AI world in 1996. Basically, if you remembered everything, would you need money? That was the punchline.

Then there was another very fine paper, Kahn and Roberds, about privacy: “Money Is Privacy.” Now you look at the basic tenants of money, privacy. Does money have interest? You and I always carried dollars. We never got interest for the money you’re carrying. I’m coming into this M1 world. M1 wants to dip into the M0 to garnish some interest, some RRP, some way. Because there is a day where interest rates could be zero.

There’s a lot in this space. I don’t think anybody has any idea. There are lobbyists. There are models. I will tell you one thing. Moneyness, for me, is not singleness. It’s the moneyness part of it. Where is the money moving? Now, stablecoins. You mentioned stablecoins. It’s backed by these five, six types of assets from coins, demand deposits, money market fund, repo, reverse repo, central bank reserves. I had a piece with Charlie Kahn, the same Kahn-Roberds paper. Some of these people, their minds are very clear.

We titled it, “If Stablecoins are Money”, five years ago, they should be backed by reserves, central bank reserves. Reserves did not, for us, include money market funds and this and that. Even MiCA, the one in Europe, it has a higher threshold for central bank reserves or cash deposits. Because remember what happened with SVB, how things unwound in two, three hours. New York was closed. California was not closed. The Treasury market is still the most liquid market. Geopolitics are changing, but will the short end change so drastically that in two, three hours—?

From a liquidity standpoint, both are safe. HQLA is safe and money is safe. From a pure instant payment, T0, reserves beat HQLA. Reserves is on the central bank balance sheet. You cannot have a better counterparty. HQLA does move in price. Of course, you will say this is a 93-day duration. It’s very short term. I totally agree. They made sure it’s not long end because with the duration, prices can move a lot.

Here, there’s another interesting aspect to complete the full story. I received one of the papers as part of my journal job a year and a half ago. This gentleman, Dennis McLaughlin, had worked for LCH. Very nice paper. I had two fantastic reviewers look at it, and they said this makes sense. The punchline was banks, the global banks, the HSBCs, the J.P. Morgans, the Citibank, the Stancharts, will they be amenable to ring-fencing stablecoins? As for the GENIUS Act, will they ring-fence it or a subsidiary this way?

Right now, for example, J.P. Morgan’s instant payment business, there’s a new name, but much of the platform is such that their instant payment or their conventional T1 business, the T+1 business, is fungible. The overall balance sheet has a lot of intraday float through netting. These guys, the global banks, are very astute in netting.The punchline was the banks, these global banks, need about 30 minutes to get most of the netting. In other words, as Asia closes, Middle East opens, and then London opens, 30 minutes. He has backed it up. It’s good. He’s got real data. By one hour, you get most of your netting.

In this world, and I don’t know whenever GENIUS Act, are the banks going to be players? Clearly, if the business remains fungible, the T0 and T1 world, in other words, there is a float subsidy which accrues all across the business, I think there’s less of an opportunity cost than ring-fencing maybe hundreds of billions of instant payments which can grow. If this market genuinely is in demand, it can grow. Ring-fencing it, it becomes an exogenous choice, not an endogenous choice. In a T0, T1 world, you’re not locked into HQLA. In today’s world, 4%, maybe tomorrow, 3%. That’s a set rate. It’s exogenous and it’s tied to the HQLA rates with the short end. That is your rate.

Today, you have opportunity bigger than 4%, you don’t need to get into stablecoins. There’s always a float angle. I’m very curious if tomorrow’s or day after tomorrow’s world, you will just have fintechs, these new fintechs dominating with their footprint. Right now, stablecoin market with the two giants, they make up more than 95% of it, or will these banks come in? I don’t want to get into interoperability and all that, but most of the banks are walled gardens.

Their stablecoins will remain with their clients. Stablecoins does have the underlying economics to be interoperable. Now, this definition is relatively new. If you’re on various platforms, various DLTs, is that interoperable? I don’t see the world where JPM Coin will go to Citi, will go here and there. That may be a pure definition of interoperability, but if in tomorrow’s world, they remain the business fungible, I think they will play in this game.

Beckworth: They will still be around.

Singh: They will be around, and there will be stablecoins from the bigger banks. If you ring-fence it, then I think you’re tying their arms.

Beckworth: Haven’t we ring-fenced it already with the GENIUS Act? Haven’t we?

Singh: Yes, I said the GENIUS Act does say that, but as we speak now, their T0 and T1 business is fungible. They do issue stablecoins. Will the fungibility remain? When does the GENIUS Act strictly start binding on them that their stablecoins will have to be ring-fenced?

Beckworth: I know there’s been talk in the news that these big banks, they were really concerned, so they lobbied hard to make sure the GENIUS Act was tailored, at least somewhat, to their liking. One of the restrictions they put in place is you can’t earn interest on stablecoins under the GENIUS Act. That way, they figured Tether couldn’t still deposit from the big banks, but what they soon discovered is that you can put your stablecoin on an exchange that does pay something.

There’s always this learning, there’s things going on. I want to go back to this point you raised earlier because stablecoins aren’t paying interest, at least under the GENIUS Act. You’re making the case that, I think I heard you say, stablecoins don’t have a very good business model in certain worlds of interest rates if they’re really low, but they’re not really earning any interest now, are they? Oh, I see, on their assets.

Singh: Their HQLA. HQLA is assigning rates.

Beckworth: If they go to zero, then they’re in trouble in their low-bound world?

Singh: Unless they are diversified. Some of them, I don’t want to name them, but some of them may be in the custody business, so they may get zero on interest but they can survive. Because remember, most of the fixed costs of their stablecoin has been incurred. I forgot to mention this Brazilian footnote. There’s been papers around, and I was privy to one of the papers where the larger Brazilian banks deal with the PICs, the instant payment. The more volume they are handling, the more liquidity needs and upfront financing needs because it’s all gross.

That means the bigger you are, in simplistic terms, the less of the netting because you are dealing with a gross. They’re asking, I was speaking to Reuters, they’re asking for regulatory relief, that if we get into the gross business—and this goes back to this paper I had mentioned where T0 does mean lesser netting, and the banks, a lot of banks, have that advantage. I will be surprised if banks remain a big player with the ring-fencing because then, at least there’s no asymmetry in the rule, right? I mean, the fintechs have to ring-fence, you guys have to ring-fence, but from this logic, the Brazilian example, if you give them 30 minutes to net, they’ll be happy. Asymptotically, you are ring-fencing them. It’s a question of 30 minutes. Do you see where I’m going?

Thirty minutes is not separately asking them to ring-fence, but 30 minutes gives them enough that they can compete. I mean, they can still compete, but I think, as I said, if you ring-fence them, just like fintechs, it becomes an exogenous thing because you then are constrained by the HQLA returns. If you allow the fungibility, because there is some economics in the fungibility, there’s some genuine economics, and they may be asking for this.

Beckworth: The challenge for stablecoins is that they’re generating the revenue through the assets they hold that back up the stablecoin. Of course, if they went to the world where you would want them to be holding reserves, master accounts, and then the question becomes, well, should they earn interest on those reserves or not? Then banks might have a different view than the stablecoins do. You’re saying the 30 minutes that’s required in the limit for these big banks to compete is sufficient to compete with a stablecoin world where stablecoins, they go around the world, they’re peer-to-peer, they’re quick.

Singh: The banks are very smart. They may still be able to compete, but as I said, you tie them because you then define the HQLA return as that subsidiary is returned versus today, they get float. Just to give you an example, if $100 billion are removed from one of these big banks, pick a bank, pick your favorite, Standard Charter [Bank]  or whatever, $100 billion moves into a subsidiary, there are some benefits because for $100 billion of less deposit, let’s say all the customers want this, you may need less equity.

Remember, some of the leverage ratios get softened, so there’ll be some balance sheet relief, but then they have not seen a paper where that balance sheet relief compared to that exogenous HQLA return compared to the loss of netting on that $100 billion. I haven’t seen such work. In my mind, there is economics to argue that the 30 minutes may alleviate us. I’m not saying I still think it’ll be equal. No equal symmetry if they also ring-fence, but they have a case where I think they could also argue, look, we’re asymptotically reaching this T0, we need 30 minutes, then this is our model.

I could see some of this unfold. Otherwise, if it’s completely symmetric, they may find stablecoin not to be as attractive as it may be for the fintechs. They may find lending, the real sector, the credit economy, those things may be much more attractive.

Beckworth: You might be separated out, payments from credit creation. That may be the reality in the future is that banks have to accept that maybe they’re no longer in the payment business as much as they are in the credit creation business. I think we need to probably spell out a few terms that we’ve been throwing around here. I know many listeners will know, but T1, T0. T1 means one day to clear, one day to process. T0 means real time. You think that a stablecoin usage, if it’s widely adopted, gets us to a real-time payment world, T0. It’s putting pressure for us all to converge to that type of world.

Singh: Let’s take a step back, how this has evolved in the last five, seven years. If you look at after Basel III, the bank balance sheets have been restricted. This is where I mentioned about, do we need new pipes or rewiring of the plumbing? I wrote three, four years ago that I was looking at these banks’ numbers and I come back to collateral. I see the world from collateral and moneyness and go to money rather than the other way around.

These numbers, the collateral, the way they pledge the markets, there are various avenues. I see a lot of articles on repo. Prime brokerage is big. Collateral comes in from prime brokerage, from reverse repo market, from SEC lending, and also from the derivative margin. There’s a whole pool that comes in. Given the balance sheets and the leverage ratios and all the other ratios, these numbers were not. The Basel III has really restricted these balance sheets.

On the other hand, the pipeline of debt, especially in the US, is large. You’re in the $35 trillion range. Who knows how far this will go? I’ve seen blogs about the stablecoin markets being north of $2 trillion. My point is, yes, there may be some subsidy to the intercept of the US Treasury because of the demand, but my point is the dealer banks are not going to be able to make markets in such large ones. Just imagine there was a 16-wheeler truck that was pulling maybe $16 trillion of debt. The 16-wheeler truck will have to pull $64 trillion.

Beckworth: It cannot do it.

Singh: It cannot do it. That’s why, coming back to some central banking facilities, I’m not surprised, trying to fine-tune the FICC with CME, all these types of CCP activities and netting. You remember September 2019 and February 2020, the spikes? The banks may see some money on the table, but the balance sheet doesn’t allow you to pick up that change on the table. Somebody has to come in. That’s why central bank balance sheets are still large. FEMA, the foreign investors, can go bilaterally to offload rather than go to the market.

Then the standby repo facility, both are at least $1 trillion. I’m saying the reason these facilities and these avenues exist is if there is some turmoil, geopolitical or whatever, you have bilateral offloading, onloading of the Treasury market. It may help dampen the Treasury market spikes, but again, the plumbing is being bifurcated. That’s a better word. I’m not saying rusted, bifurcated. It makes sense during Lehman, you have very strong central bank—

Beckworth: What do you mean by bifurcated? What’s the two parts?

Singh: You have the dealer banks making markets. As I said, collateral comes in. 

Beckworth: They have to, by law, to be a primary dealer?

Singh: Yes, but if your balance sheet doesn’t allow, then you can’t. If you saw central banks, basically, at least out here, were issuing on the runs and buying off the runs. This is the September 2019 to February, March of 2020. The dealer banks can do this, but they didn’t have the space. You would rather pick up, I don’t know, 100 pips rather than 10 pips. If you have prime brokerage and you have hedge fund clients, your balance sheet is for them than to clean up on the run.

Beckworth: The one is the primary dealers, and the other, the bifurcation, what’s the other?

Singh: Central banks have these. Think of pipes. Before Lehman, central bank footprint was very small. You know the central bank balance sheet. It wasn’t very big. It was less than a $1 trillion, at least in the Fed. Post Lehman, things happened, and things had to happen. Somebody had to come in. Then the eurozone crisis happened. As I said, this goes back to my comment about QT being very small.

Beckworth: I see what you’re saying.

Singh: The pipes remained, and the pipes now, we have some extra pipes. I think the standby repo facility came up. 

Beckworth: Reverse repo is another extra pipe, right? Now money market funds can tap the balance sheet.

Singh: Exactly. These are pipes where you have direct collateral money transactions with central banks. Pre-Lehman world, it was a market, the market pipes. Everything came to the broker dealer. The plumbing is bifurcated in a nice, pure world. If there were no crisis, you would remain in the pre-Lehman world. We never went back to the pre-Lehman world. There are these pipes which will remain exactly because if something happens, you have a $1 trillion in the FEMA, you have $1 trillion in the standby. The RRPs was north of $2 trillion once upon a time. It is coming down. These pipes remain. Is there a way? I know your favorite question. Can we come back to the pre-Lehman? In my mind, getting a pure market pipe without these bilateral central bank pipe would be back to the corridor, not the floor. I have written about it.

Beckworth: Unlikely that I’m hearing you say it’s unlikely.

Singh: It is unlikely. They are comfortable. I think these pipes, because if something goes wrong, these pipes will come out again. You got used to these pipes.

Beckworth: I think George Selgin has made this point, though. In an ideal world, you could have these pipes ready to use during a crisis. Then after the crisis, you clean up and you shrink the balance sheet. You don’t use the pipes, but we still use the pipes because I think a lot of it has to do with what you mentioned earlier. The regulations restrict what these primary dealers can put on their balance sheet, and just the regulations. And Bill Nelson and others have argued there’s also some cultural phenomena. The bank inspector, supervisors, they expect you to hold a certain amount of deposits at the Fed. It gets more complicated. It’s not so easy just to dial back. Also, the US financial system is unique. In the ECB, they’re pushing hard to shrink the balance sheet and go to demand-driven operating system.

Singh: Even in the UK.

Beckworth: In the UK, even more. Those, particularly Europe, the financial system is much more bank-driven. The US, we’ve got a lot more securitization, and therefore, a lot more money market accounts. We rely on those to intermediate finance. That’s why the reverse repo facility is important. I understand the difference in the structure of financial systems can also have a bearing on what’s a practical operating system in the US. 

That’s why, in my dream world, which I know you said is unlikely, and you’re probably right, but I would love to see us move from a floor system to demand-driven and something like a ceiling system where not just banks would go to the discount window, but the standing repo facility to central clearing was open to money market account, was open to all kinds of counterparties. That would be an important pipe. It’s not a very important pipe now. That would be a great world, in my view, because it’d minimize the footprint of the Fed.

Singh: No, you’re right. The more they can do in a market-based solution, like the CME, FICC, when I speak to them, they say, it’s hard. One is under Chicago Fed, one is under New York Fed. The more you can get netting from that, the less likely you will use these FEMA pipe and the standby repo pipe. If you don’t use the pipes, maybe the pipes will never be used. The deal of balance sheet is constrained. I think you might have seen many papers. Basel did what it had to do. The way the debt’s been issued all across, especially in the US, they cannot complete markets everywhere. Some markets will lead to spikes.

Beckworth: Yes, and we rely more on these principal trading firms, these high-frequency traders, the hedge funds that they step in. They pick up the slack, but the problem is they also get out quickly, whereas these primary dealers, they’re more stable, relatively stable, but they just simply can’t be as big of a player.

Stablecoins and the Eurodollar Market

We’ve been talking about stablecoins and the implications for the plumbing and stuff. Let me step back and get away from some of the details really in the weeds here. Do you see the rise of the stablecoins more akin to the emergence of the Eurodollar market or the money market funds in the 1970s? What is a better historical analogy to look to when we think about the emergence of stablecoins?

Singh: We wrote about this and we drew parallels with the Eurodollar market, especially if it’s not in the reg perimeter. Out of the two stablecoin issuers, one is under the GENIUS Act. It’s in Europe. The other one is not in the reg perimeter, at least in the US. In the Eurodollar market, it survived. It stayed. The monetary authorities, there was always incompleteness, but there was a unique thing about the Eurodollar market. These were all large banks.

Now, I’ll make a subtle point about stablecoins. These were large banks, so if you really needed to understand offshore activities between Tokyo and London, you could. These are banks, and central banks can get information on banks. When you come to this world, with them, no central bank account, and the regulations are very different than the ones on banks, a few things come to my mind. Remember how the old-school open market operations are? It’s reserves. The central bank takes reserves, takes out, again, back to money and collateral. It’s as simple as that now.

You’re backed by HQLA. This whole stablecoin world is HQLA. It’s not backed by reserves. That’s why my bias towards reserves. In this HQLA world, you have another player in the securities market. The way you target money supply, which was you were monopolist, you can tweak it and absorb collateral or release collateral, and you can target some. Now, it gets very difficult because there’s another player who, on the back of HQLA, is also calling that money. It’s very new. It’s technological change. It has some benefits, but I’m not saying I’m fond of central bank monopoly on money.

Moneyness matters. Moneyness has always been there, even more so in the last 20 years. It got too much before Lehman. It’s come down after Lehman. There was a piece we wrote, central banks have to be careful about stablecoins, especially if it’s not backed by reserves. My take is that the way central banks’ monetary operating system is in this HQLA regime, it is now slightly biased towards the fiscal side because you see the US Treasury. If there’s demand, they issue. You have something which is not under the Fed reserves-backed fractional system. It’s a narrow bank backed by something which comes from the fiscal side.

This will be very interesting. Bottom line, we have seen in the case of crisis, uninsured deposits have been taken care of many times. I can see if these players are big, let’s use the phrase, too big to fail. If the HQLA goes below par, for whatever reason, I don’t know, geopolitical reason, whatever, I cannot see any of them also going under.

Beckworth: The Fed will be there to backstop them, whether they’re under GENIUS Act or not. Tether outside in some other country, they’ll still be backstopped because they are part of the global dollar system.

Singh: Let’s go back to Dodd-Frank. You have a few conditions on who picks up the tab. I think one of the conditions was Fed has to get an okay from Treasury. If you’re backing it by Treasury, obviously, there’s an implied put over there. This world is just opening up. These numbers are supposed to go from $250 billion to $2 trillion. Can you imagine that type of market and people getting used to money? Now, coming back to the very old school, there will be many who will still want to live in that old-school private fiat currency world who may not be comfortable with stablecoins. I don’t have anything. I’m, in that sense, technologically a lot behind.

Beckworth: You don’t have your digital wallet with stablecoins on it.

Singh: I do get my points on my Visa card. For some of us in the advanced economy, you can give me stablecoins, but I get enough miles every month and here and there that I don’t usually pay for local flights. Now, there are other issues. In fact, if you look at some of the decomposition of fee from the MasterCard, Visa account, it’s very interesting how they decompose the fee. Very interesting. That’s going to change. I think people who are heavy users and accrue points and use a lot on the Visa card, they’ll still—

Beckworth: They won’t be as interested in stablecoins.

Singh: They will not. Number two, I can see this other world, the Venezuelas, Argentinas, where the Tethers are very dominant. Then there’s another angle in eurozone. Visa, MasterCard may be erased. There’s no reason. They have enough technology there. Why are we paying two American giants some revenue? I’ve heard that from very reliable sources that they may do this capped CBDC or whatever you want to call it that erases Visa card. Every jurisdiction is having its own internal debate. This is a very new world.

For me, it is always through the lens of moneyness. Maybe from the reserve backing by the Fed, this will be backed by the Treasury. We’ll go back to Dodd-Frank. A lot of parallels are being made very quickly, like this letter, tokenized money market fund. Are they identical? This and that. I just don’t know where this is going. I can see these are shades of moneyness. I have always seen the interplay between money and collateral.

When you asked me, is this money? I said, moneyness is a continuum. The dealer balance sheets often found more money in a Treasury bond at certain times and less money in it. This is where I mentioned that in QT, we may have messed up on the moneyness because of what may be more money for the market.

Beckworth: Let me repeat a point that was made by Rashad Ahmed on a previous podcast. It deals with the possibility of a financial crisis or financial stress or a run on these stablecoins, which is a possibility. That’s what a lot of people are concerned about. This will happen. I agree with you. I think implicitly, we already know the Fed will backstop them if push comes to shove. I think on top of that, this point, what Rashad Ahmed make was, at least in emerging markets, one of the big challenges has been that balance sheets, both households and corporations, they have dollar liabilities on one side, but they have local currency on the asset side. Whenever the dollar fluctuates in value, you’ve got this currency mismatch.

If you start having dollar-based stablecoins on the asset side, you have assets that are in dollar terms, you’ve got liabilities in dollar terms, you begin to reduce the currency mismatch. Maybe it won’t be a one-for-one wash, but you would actually reduce the prospects of this global financial cycle tied to the dollar. There might be other things that cause problems. On one hand, you worry about the runs. On the other hand, everything is more in sync. If the dollar were to go up or to go down, if everyone’s got dollar assets on their balance sheets, the world could actually be a better place. There might be less global financial volatility, even if there’s an occasional run on a stablecoin.

Singh: Now, let me break your question into two parts. You mentioned household balance sheets, and then there is the sovereign central bank balance sheet. If the household, assuming there’s no friction, they can use, through the Tethers or whatever, they themselves can match and reduce the dollar mismatch. I can see it over there. From the central bank viewpoint, central bank balance sheets, I wrote a piece on what’s an optimal central bank balance sheet. There’s no clear-cut answer.

In a good world, again, a world where plumbing is not bifurcated, and you’ll have some very fine old-school people, Ulrich Bindseil comes to mind, your balance sheet should basically be CIC, currency in circulation, and required reserves, and some equity. We looked at it. Actually, at IMF, we had the benefit of the statistical department. If you use this rule of thumb, the balance sheet should not have these very large items, which happened because of the Lehman issue, excess reserves bigger than required reserves, so much so that they take no more interest on required reserves.

A normal balance sheet is 90% plus a clean balance sheet, relatively speaking, where you just had M0, CIC, plus required reserves. If you abstract some of these advanced economy aberrations of time, and you look at the real world, since you mentioned the use of dollars in other parts of the world, I have looked at some of them. Some of these balance sheets are generally very high M0 to central bank balance sheet. I remember Morocco, Albania, Uzbekistan, very high. In other words, there’s not much other items.

Some of them actually have to work because you need international reserves. Who knows, this may be changed if the households are more hedged. China and Brazil would increase required reserves because if you needed $100 billion in assets to cover yourself from a balance or payment issue, but your M0 was $50 billion, you have to do something else to come up to $100 billion because that’s what you needed. Assets would drive that type of thinking in the central bank. There are other line items like public sector deposits at the central bank which don’t get reimbursed. I’ve seen all sorts of balance sheets across the world.

How do you fill your liability side if your asset side, your demand for dollars, say, India, $600 billion, you need that much reserves relative to what you may just get from M0. There are times where other line items came. On the other hand, I’ve seen the same logic being used very cleanly recently, in the last 12 to 18 months. Argentina took a lot of haircuts on the asset side, Argentine central bank on the asset side. What happened there? M0 to central bank balance sheet went up, which is a positive. The bigger M0 to the central bank balance sheet is a positive rather than M0 being like 5% or 10%. That means you’ve got 95% of other stuff which should not be there.

When these guys took a haircut, significant haircut, asset side shrunk. M0 is not changing. If asset side shrunk, what happens out here? Then the equity had to shrink. Then now the M0, which was like M0 to total balance sheet size, was roughly 10%, is now 25% because of cleaning up. There are rules of thumb you can follow, but the idiosyncrasies of certain countries, public sector deposits, and sterilization. Sometimes balance sheet size may not change, but line items get swapped. Certain central banks can issue notes. The interest rate structure can change. Notes is a different interest rate versus long-term bonds.

It’s very interesting. That’s why there is no paper, a good paper you can say that’s a very good paper on optimal central bank balance sheet. It is very, very heterogeneous across the world. How do you link it to your whole stablecoin issue? If households start hedging themselves and you think there may be less mismatches at corporate, household, then goes to corporate, goes to the banking system, then I would imagine and I can see two points. I can imagine lesser need for dollars because your overall economy structure is better hedged, assuming that happens. I don’t know how it will happen.

Number two, with the geopolitics in play, as much as there is demand for dollar hedging because of what happened in the recent years with Russia and some other countries, some people may unwind from dollar. There are two things which could happen. The demand for dollar at the sovereign central bank level could go down, A, because you are better hedged and you allow stablecoins, and B, they are diversified initially to other G7 countries. Coming back to some pieces which may sound esoteric, if Chinese Govies get into LCH type of logic or the Enbridge logic, you may start holding.

As I said, Chinese renminbi is part of SDR. That’s a big step. Under Ms. Lagarde, they were considered integral part of SDR. For me, the next bucket is very similar to SDR would be Chinese Govies acceptable collateral at LCH, which would be in sync with their logic of mBridge, the logic of internationalization of renminbi. Who knows, if this happens, their stablecoins banked by renminbi may also be in demand around the world. 

If you look at UAE, you look at Singapore, Hong Kong, they are not sitting idle. This demand for dollar at the micro level hedging is a plus in some sense and a demand for dollar, but the flip side is it gets very complicated, but it’s very new and it’s very intriguing. In fact, I would like to spend more time on this topic. Now that I am retired, I can write a little bit more fully. These things come once in 30 years. There was a time when money demand function changed with this ATM machines.

If you look at Townsend’s work of 1980, you look at some of the paper, “Money Is Memory,” I don’t know. I don’t know in the next two, three, five years what’s happening, whether there’s a crisis or not, but GENIUS Act is here. It’s not going to be erased. There are some very deep pockets in this game. Money may not be the same anymore. Moneyness will remain, and which aspect of moneyness do you want to pick on, but money is changing.

Beckworth: In other words, another shock to the velocity of money due to financial innovation via stablecoins. We don’t know where this is all going to end up in terms of what is the money aggregate, what is money, what is the final form of money people widely use. We need to stay tuned to Manmohan Singh’s work. He has been our guest today. Thank you so much, Manmohan, for coming on and explaining all of this to us.

Singh: Thank you again for the opportunity. It is unfolding very quickly, but it is very interesting. These paths don’t come every day.

Beckworth: We’re living through history.

Singh: I think so.

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.