Yesha Yadav is a law professor and associate dean at the Vanderbilt University Law School, where she works on banking and financial regulation, securities regulation, and the law of money and payment system. Yesha has written a recent paper titled, *The Failed Regulation of US Treasury Markets*, and she joins Macro Musings to discuss it. Specifically, David and Yesha talk about the implications of the 2020 Treasury market collapse, the fragmented nature of the Treasury market’s regulatory structure, solutions for reform, and more.
Read the full episode transcript:
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: Yesha, welcome to the show.
Yesha Yadav: David, thank you so much for having me, such a pleasure to be here. I'm a huge fan of the show. This is a show that brings us a lot of education every single week, really. So thank you so very much for having me.
Beckworth: Oh, you're too kind. And I want to also just mention Dan Awrey here in the beginning of the show, because he connected the two of us. And your whole group of finance law professors working on these issues that intersect financial markets, macroeconomics, really doing interesting work. I've had a number of your colleagues, Kate Judge, and some others. So it's really neat to bring in the law side of finance and economics here. And I'm delighted to have you on this show because you have an amazing paper you've written on the Treasury market. And before we get into that though, Yesha, why don't you tell us a little bit about yourself? How did you get to this point where you are a Treasury market expert?
Yadav: I would say it's a bit of a long and winding road, David. I grew up in the UK. I went to school in the UK. I did my law school in the UK, and I worked in a gigantic law firm, Clifford Chance where I was working in financial regulation and derivatives markets before the crisis. And that was a really heady time when the markets seemed to work and be productive like the Marvel cinematic universe productive, just everything seemed to click. And I had the sense at that time that the bankers I worked with, the fancy lawyers I worked with, they knew everything. The regulators I worked with, they knew everything.
Yadav: When Bear Stearns started to have its little implosion at that time, I remember distinctly in March 2008, I went into the loo and I remember standing there and remember thinking that in fact folks that I thought knew everything didn't. That the structures in which the markets were working and operating, that these structures were in fact much weaker than we realized. And all of that time I'd wanted to be an academic, but I didn't really know how and what I would have to research. And suddenly, it seemed to crystallize that all of these knowledge structures that we took for granted, that the regulatory structures that we thought made perfect sense in the markets that were so productive and seemingly endlessly generative that in fact, that wasn't the case.
Yadav: And so, that was really the beginning for me of thinking about academia as a serious possibility. Following the crisis, I got to go to the World Bank. I was so lucky to get to work in international financial markets standard setting. Immediately post crisis, that was just a fascinating opportunity. And then finally, I got to Vanderbilt, which has just been a phenomenal home and I have fabulous colleagues like Morgan Ricks in my faculty. And of course, the wonderful folks like Dan, Kate, and others who are just a terrific part of our academy.
Beckworth: Yeah. So you've done some interesting work and it's fascinating, not only do you work on Treasury markets, but you work on financial markets innovation, securities regulations. So maybe at some point in the future we'll have you back on to speak to those issues as well.
Yadav: That would be wonderful. Crypto is blowing up I hear.
Beckworth: Yes. Yes. And you would know, you're an expert in that area, too. Well, let's move into the Treasury market discussion. And before we do that and get into your paper, it might be useful for some of our listeners, now, most of us we'll know these facts already, but for those who don't know the basics of the Treasury market, maybe just a quick overview. How big is it? Just do some basic details about the Treasury market.
Basics of the Treasury Market
Yadav: Yeah. The US Treasury market is just a fascinating market, and it's a weird market. Even for folks that are deeply enmeshed in securities markets regulation and financial markets regulation. It's a funny little market structure. So first and foremost, of course, it's a really important market. Its significance has become amplified since the 2008 crisis. Outstanding debt at that time in August, 2008 was around $5 trillion outstanding in marketable debt. Today, of course, that figure is closer to $22 trillion.
Yadav: So just a quick primer on the market structure, it’s divided into three parks basically. The first part is that primary market that we all know where the US issues its debt, that's the primary market. That's where every single week the US does all auctions and folks buy that debt. The secondary market then is where that debt is traded. And this market is divided into two parts, mainly. The first part is this market, which is called the dealer to client market. It's a market in which folks like hedge funds or China, Japan, France, folks that want to buy treasuries, they come to that space to buy treasuries. This is where the power of treasuries is realized in a most tangible way.
Yadav: That's the dealer to client market. We see around 250 to $300 billion worth of trading in that market a day. And then we have this other incredible space, which is called the inter-dealer market where dealers work with one another to modulate their supply and demand of treasuries. And in that market, again, it's a market which has around 250 to $300 billion worth of daily turnover. Now, one aspect of this market structure, which is fairly interesting and fun is that it's traditionally been dominated by this group of banks and investment banks called primary dealers. They are the 24 currently big banks and investment banks that we all know, and they are the main buyers of US Treasury debt at auction in the primary market. Around 70% on average of debt is usually bought by primary dealers. That figure is, all the research is from early last decade, but it may have changed, but around 70%.
Yadav: So primary dealers have an enormous role to play in intermediating in the secondary market. So they are the key dealers in that dealer to client market. They make that market work. They are the liquidity providers there. And traditionally, they have been major players in the inter-dealer market, traditionally, but not since the last eight, nine years or so. That honor now goes to high frequency traders. So in the inter-dealer market, primary dealers have given up some of their competitive market share. Today, high frequency traders, these securities forms that transact in milliseconds and microseconds, they're the dominant players. 60 to 70% trading volume in the inter-dealer market is dominated by high frequency traders. So that in a nutshell is Treasury market structure across the board.
So primary dealers have an enormous role to play in intermediating in the secondary market. So they are the key dealers in that dealer to client market. They make that market work. They are the liquidity providers there. And traditionally, they have been major players in the inter-dealer market.
Beckworth: And we're going to come back to some of these details later, particularly, the high frequency trading and the role they played and some of the challenges that the Treasury market now faces and has faced. But I want to go back to the number. You mentioned $22 trillion, so that's roughly just under 100% of debt to GDP ratio. But you often hear other people throw out another number. And I think it's about 28, $29 trillion. And that includes this inter-governmental holdings and people who are really worked up about the national debt will invoke the bigger number. But why is it more accurate to invoke the $22 trillion number?
Yadav: I look at the $22 trillion number because I'm obsessed with liquidity of the Treasury markets. Right? The reason why the Treasury market... In fact, the reason why we are able to borrow $22 trillion is because our debt is regarded as being risk free. That means two things. One, that we're not supposed to default in our debt.
Yadav: December 3rd is now past. But we're not supposed to default on our debt. But number two, the very important part is that this debt is super liquid. And what that means is that anyone holding a treasury should be able to buy and sell it easily, cheaply, reliably, and without affecting the prevailing market price. That liquidity is central to making the Treasury market the way it is. These are the securities that the entire world depends on to keep themselves safe. And it's owing to that liquidity that these securities contain. And so for me that $22 trillion figure is the key one, because these are the liquid securities that the globe is consuming as a safe asset to be able to anchor itself into a stable asset.
Beckworth: That's the part of the Treasury market that is essential to the global stability of the financial system. It's the systemic part of what could be a potential crisis of things didn't work out right. That's a good point. I guess the other side of that I was thinking about is that other $6 trillion or so, that's debt that the government owes to itself. Right? It's inter-government debt. So it's not really a true liability, is that fair to interpret it that way?
Yadav: It is an aspect of the government's own balance sheet. And so of course, nominally on paper, it is owed. And one should regard it as being owed. And certainly, when you look at social security holdings, and the treasuries, and other things that are part of the capital structure, then those are regarded as being important aspects of those balance sheets. But for the purposes of understanding the market and its fragility and the impact of the treasury on a global scale, it's really that $22 trillion figure and particularly the marketable treasuries and how they're traded. That's really the salient topic that is important for my research, and I think what the world really cares about.
The very important part is that this debt is super liquid. And what that means is that anyone holding a treasury should be able to buy and sell it easily, cheaply, reliably, and without affecting the prevailing market price. That liquidity is central to making the Treasury market the way it is.
Beckworth: Okay. Well, let me read the introduction to your paper to motivate your paper and the fascinating discussion it generates. So I'm going to begin reading from the beginning, just take a few excerpts out. I'm going to ask you a question. And you start off by saying, "In March, 2020, as the COVID-19 pandemic ripped through the economy, the then $17 trillion market for US government bonds or treasuries was brought to the brink of failure. Because investors rely on treasuries to keep them safe during crisis, the potential collapse of treasuries presented an unthinkable doomsday scenario for global markets and the US economy.
Beckworth: The Treasury market was supposed to be the safe haven for investors that needed to sell treasuries, to raise cash or buy them as protection. Instead, as panic took hold and investors tried to cash out, the market faltered to a crawl.” You then go on to say, “the suddenness, notwithstanding the ill time collapse and risk-free Treasury markets is unsurprising and overdue." So for you, this was unsurprising and overdue. What do you mean by that?
The Collapse of the Treasury Market and its Implications
Yadav: The warning signs were there, David. Right? This was not, or should not have been a surprise. The market structure of this market is creaking and fragile. And that is owing to two aspects. And as a lawyer, the aspect that I look to most fundamentally is the regulation. Right? First and foremost, the regulation of this market is structured in a way that I regard as being just fundamentally ill-equipped to dealing with modern risks of a very modernizing, technologically advancing Treasury market structure. Our regulatory system for the US Treasury market simply does not work. The way in which the US Treasury market is regulated today is through a very fragmented system of oversight that looks to at least five separate federal agencies, none of which has any primary authority to act on its own to regulate the market.
Yadav: So you have the Fed, the New York Fed, you have the Treasury, you have the CFTC, you have the SEC, you have FINRA, OCC, too, is going to be playing a part because of the bank dealers. None of these agencies has a lead role in pushing a regulatory agenda. So even though we have a lot of very expert, super smart, experienced agencies, there's a lot of cost embedded in taking even small steps towards common sensical regulation. So what has happened in US Treasury market structure is that the market structure has evolved at a pace that should be expected towards electronification. We now have HFTs that are big in this market, as they are in many, many other markets. That's normal. It's a part of the market's furniture today.
Yadav: The dealer to client space even is electronified, even though it's OTC, but it's increasingly electronic. And of course, this market is super interconnected with every other asset class out there. It has to be. It's the Treasury market. But all of these deep interconnections were missed as they have to have been missed because the regulatory system that we have is fragmented. Now, to give you one example, in 2014, the US Treasury market suffered what was then a flash rally, as it was called when Treasury market prices on October 15th, 2014 went haywire for about half an hour in the morning. No one knows why, no one could have known why. The government produced a report in 2015, where the agencies confessed that the regulators did not themselves know that high frequency trading had become a critical part of the inter-dealer market. I find that shocking.
The regulation of this market is structured in a way that I regard as being just fundamentally ill-equipped to dealing with modern risks of a very modernizing, technologically advancing Treasury market structure. Our regulatory system for the US Treasury market simply does not work.
Yadav: Right? It's on paper. It's codified. It's part of that report that even something as fundamental, and basic, and obvious as that was not captured, and we should not be surprised. In a fragmented market structure, we're going to have gaps, cost to communicate institutional mandates that restrict the flows of information, very difficult job in producing and understanding of what the interconnections of this market look like. So I am not surprised that even common sensical rule making has not happened here. Therefore, we should expect this market to suffer liquidity problems, disruptions, glitches, flash rallies, whatever, like every other market does. It has to happen.
Beckworth: It's just striking though, as you outline, and you just said that the most important market in the world, no one regulator or no one entity really knows with complete certainty what's happening in all parts of the market. And you think maybe that would not be the case, because this is such an important market. Now, we'll come back to the market structure and the fragmented oversight, those agencies that you mentioned. And you have some proposals we're going to get to at the end as well that hopefully will improve that and make it better.
Beckworth: But let's say for the sake of argument that the Treasury market had been perfectly regulated, we had better oversight. People knew what was going on. Would it be safe to say that with all that in play, the pandemic still would've strained the Treasury market, all the pressures it created? So maybe we shouldn't hope for too much in extreme tail events like that?
Yadav: It's an amazing question, David. I think it's a brilliant question because obviously it gets to the limits of good regulation, too. Right? And the necessity of this market is one that we cannot overstate. So obviously, when the Treasury market went into March, 2020 it was facing enormous selling pressure, as we know. So we were having selling pressure from governments. We were having hedge funds. We were having mutual funds that were all exerting enormous selling pressure. As the recent inter agency working groups report also highlights, there was also buying pressure from prime money market mutual funds that were trying to get around $300 million worth of Treasury bills by them. Right?
Yadav: So there's a lot of liquidity pressure being exerted into this market. But that's exactly what this market's supposed to absorb. Right? The US Treasury market is designed, it's supposed to, it's entire purpose it's to provide the add final anchor to stability when every other market is collapsing. That's its job. Its job is to be able to absorb enormous liquidity, one-sided liquidity pressure if it has to, because that fundamentally is the function of the US Treasury market, to keep the country and to keep the world safe.
Yadav: Now, if we had had perfect regulation, then certainly, that's not going to avoid the pressure that the market is facing, but the market may have had better tools to absorb that pressure. And particularly, if primary dealers and others may have been able to equip their balance sheets better, to be able to sustain extensive market making at a time when their market making services were badly needed. So there is no perfect solution. But I think we could have equipped the market better. So, today there is enormous opacity in the US Treasury market. Even the secondary market is opaque on a real time basis.
Yadav: Even regulators don't have perfect data. In 2017, a new reporting regime was brought in for trades as late as 2017, but at least it was there. But that reporting regime still has gaps. For example, hedge funds don't have to report. Many HFTs that are not subject to FINRA oversight, they don't have to report directly. And so there are these big, deep data gaps that are embedded within the regulatory structure and within the market, which means that regulators are not getting the full picture on a real-time basis as to how that trading is working, where the risks are, where the selling pressure might originate. How that selling pressure might originate, what the interconnections are between the futures and the Treasury markets, and how that might impact the pressure in the secondary market.
Yadav: These are all aspects of informational understanding that would help regulators to plan better, help primary dealers and HFTs to get their balance sheets in a position where they can absorb the impact. And maybe to also ensure that maybe primary dealers have some liquidity support, if they need it to the extent that we may need to handle such enormous demand on the Treasury market at that time.
Now, if we had had perfect regulation, then certainly, that's not going to avoid the pressure that the market is facing, but the market may have had better tools to absorb that pressure. And particularly, if primary dealers and others may have been able to equip their balance sheets better, to be able to sustain extensive market making at a time when their market making services were badly needed...there is no perfect solution. But I think we could have equipped the market better.
Beckworth: I want to hold off on your big solutions to the end, but I do want to throw one question out there to your related to maybe fixing this data gap problem. And that is the Office of Financial Research under FSOC, do they have the legal authority, the statutory authority to go fill those gaps if they really wanted to?
Yadav: Right now, I think that is difficult even for the OFR, even for the Treasury. There are institutional mandates that prevent information sharing on a large and system wide scale from the administrative standpoint. So for example, when 2014 the flash rally happened, the different agencies had to negotiate with one another for data. So for example, it took almost a month, I think, I could be wrong but almost a month for the CFTC to negotiate to give up its data so that the problems could be analyzed. Obviously, banking data is super sensitive as well. And so the banking regulators had a hard time deciding the conditions in which bank reporting and treasuries trading would happen under these new regulatory regimes that have been put in place.
Yadav: So there is enormous sensitivity around treasuries data. There are issues regarding the pooling of data. And so even going out by itself, an expert like the OFR would still have difficulty because it would have to negotiate with the several agencies involved here to get basic data. Now, one other issue I just want to flag for you, David, is that the US Treasury secondary market is also connected to the repo market. I have another paper with my co-author, Pradeep Yadav, and there we talk about the interconnection between the repo, treasuries backed repo, and the secondary trading market. As you know, David, as conversations with Morgan you've had, the repo market is completely opaque. Right? It's informally sensitive by design. And so there is that massive gap that exists in this repo space where primary dealers are hugely involved. And so even with the OFR gathering data on the secondary market, it would still need to get a picture of the repo market. And that's hard.
Beckworth: Very fascinating. And again, very sobering that this most important financial market in the world has all these challenges. One other question about the Treasury market, then we'll move back into the fragmented oversight of it. And that is the debate that I often have with other people about what actually happened in March, 2020 to the Treasury market. I mean, you've outlined much of it already. But I like to call it a plumbing problem. The Treasury market itself was, in a broader sense, solvent. The federal government wasn't going to run out of money per se. It wasn't because people thought the treasuries were in solvent and they had a dash for it, it was literally the pluming. That's the view I've taken.
There is that massive gap that exists in this repo space where primary dealers are hugely involved. And so even with the OFR gathering data on the secondary market, it would still need to get a picture of the repo market. And that's hard.
Beckworth: I've had others like, "No, no, no. There was just too much debt out there. Too much big deficits running and the system couldn't handle it." And I'm guessing you would bring in a third perspective, the regulatory structure and the lack of development in the Treasury market contributed to it as well. But how would you step into this conversation or what would define March, 2020?
Defining the March 2020 Treasury Crisis
Yadav: Yeah. So my perspective here is that the regulatory system, does not, cannot currently and anticipate the problems that will afflict this market because it is not structured to do that. So the regulators are ill-equipped to get a handle in all the data, develop a picture of the interconnections between the markets, and therefore have some sense of what kind of constraints need to be put in place. And assistance on primary dealers and others that are key liquidity suppliers in this market.
Yadav: What do I mean by constraints? Well, liquidity providers in this market are free to leave anytime they want. Primary dealers, even though they're so important to the health of this market have no obligation to actually be there. Neither do high frequency traders, or indeed anyone that is a key liquidity supplier in this market. There's no mandatory need for them to remain on this market and trade, even in a crisis.
Yadav: So that means that when the going gets tough, when it's hard, when it's costly to continue providing liquidity, the incentive on liquidity providers, generally, is to think about leaving. Particularly when there's a large amount of uncertainty and the kind of exposures they might face, or the one-sided demand for cash or treasuries is to leave. That's rational. They should be doing that, arguably if they're compliance systems are flashing red than saying, "Get the hell out of here."
My perspective here is that the regulatory system, does not, cannot currently and anticipate the problems that will afflict this market because it is not structured to do that.
Yadav: But the question for the public good and for regulators is how does this impact the Treasury market? And it feels like it introduces a scenario in which liquidity might be wonderful on a great and sunny day. But when things get heavy and difficult and hairy, then the liquidity is liable to disappear. And that's exactly not what you want to happen in the Treasury market, because this is the market that's supposed to undergird the entire financial system, particularly where that financial system is in crisis. So for me, the couple of causes here are to get back to your question, the regulatory structure is ill-equipped to put constraints in place. I think we need to understand what constraints are needed because liquidity provision in this market is looking increasingly fragile.
Beckworth: Okay. Very nice. Well, let's go back to the point you made earlier about the fragmented framework of regulatory oversight. And you list five main agencies that are responsible. Then you have five plus, because you also bring in these other regulators like the OCC. But let's walk through them and tell us what role they play. Again, this is a fragmented structure here. But let's start with the Treasury. What role does the Treasury department that itself play in the Treasury market?
Who is to Blame for the Fragmented Regulatory Structure of the Treasury Market?
Yadav: The US Treasury is obviously critical to it. It regulates the auctions, mainly it is the key agency that is important for ensuring that the auctions are flowing smoothly. So the US Treasury has a key role in that, and obviously, is a central player given the skin in the game for ensuring that the treasury is properly funded.
Beckworth: Let me ask this question, who determines the debt maturity structure of the public debt? That's the Treasury. Right? And do they have a committee that advises them? Is there some method to their determination of the debt structure?
Yadav: This does change. For example, we have a new 20 year bond that's been introduced. And so, these are conversations that are happening within the Treasury to think about what the most optimal funding model for the Treasury would look like. There have been various iterations as to what would work to make the auctions go smoothly. So for example, there was a time when we never used to have auctions on a regular basis. And so the market sometimes would freak out when the government would suddenly go to market to ask for debt. And so now we have this principle of regular and predictable auction. So, this is a central part of regulating the credit provision for the treasury, which is regular and predictable auction.
Yadav: So the market knows what's happening. Primary dealers know what's happening. So these procedural issues are very important for the Treasury to think about, to make sure that the funding that it gets is at lowest possible cost. And obviously, the Treasury is also responsible for making sure everyone gets paid on time. And that can sometimes be an issue. I think it was 1979 when the checks went out a bit late. And so it was a technical default in the US Treasury, technical. And that actually had some repercussions, too. So, it's very important for the Treasury to make sure the process for the auction goes and is set up smoothly.
Beckworth: So I guess it's important then to have a sense of the demand ahead of time of those auctions? I mean, they must talk to the market participants, "Hey, we're going to issue tens, twenties, what are you going to buy?" I guess there's some coordination, some feedback they're getting from the markets. Right?
These procedural issues are very important for the Treasury to think about, to make sure that the funding that it gets is at lowest possible cost. And obviously, the Treasury is also responsible for making sure everyone gets paid on time. And that can sometimes be an issue.
Yadav: Yeah. I mean, this is why it's great. I mean, this is why, it's not just the US, but other economies, continental Europe, the UK, and others all rely on the system of dealers. Right? On a system of these core banks and investment banks that are the key buyers of Treasury debt. So the reason why this is great is because these institutions can then go out and get an understanding of how much treasury would be needed. So they are then able to inform the US Treasury and other treasuries that this is what the likely demand is going to be.
Yadav: Now, this system is not perfect. There has been a concern lately that some treasury auctions don't get the kind of demand and uptake that one would like. So in February, 2021, for example, one auction towards the end of the month went particularly badly. There wasn't that much demand for it. The Treasury was caught by surprise. It affected the US Treasury market. The liquidity in that market dried up that day, had a bit of a problem. The system is not perfect. But the idea here is certainly to ground it in a framework that is as metronomic as we can possibly enable it to be. That ensures the market's not surprised. We get information from the primary dealers. We work on a system that ensures that our debt is being issued at lowest cost.
Beckworth: Okay. Let's go to the Federal Reserve Bank of New York. What role do they play?
Yadav: Well, the Fed is the Fed. It's a key banking regulator alongside the OCC. It's our lender of last resort. It's the Fed. As we know from March 2020, we'll step in if there's a problem. But why it's here is because it's our central bank. It is also, of course, a major banking, the major banking regulator alongside the OCC.
Beckworth: Yeah. And as we saw in March 2020, it's the final backstop, too, to the Treasury market if it push comes to shove. You mentioned in your paper over a trillion dollars and repos right after the market's froze up, and then more after that. It's the final backstop. I wonder if you have any thoughts on the fact that the Fed's balance sheet has gotten so big so that it's effectively doing a sizeable portion of public debt management. So, I think if you add overnight repos, plus reserves, you're looking at liabilities close to over $5 trillion. And as you mentioned earlier, we got $22 trillion in our public debt. So the Fed's becoming a non-trivial player in the management of public debt. Is that a problem or not a big deal?
The system is not perfect. But the idea here is certainly to ground it in a framework that is as metronomic as we can possibly enable it to be. That ensures the market's not surprised. We get information from the primary dealers. We work on a system that ensures that our debt is being issued at lowest cost.
Yadav: I think I wouldn't call it a problem as such, I would think it's a state of affairs. It is a part of our ecosystem and we need to figure out what we do with it. In other words, when we do have a regulatory reform in this area, and there's talk of it now, how should the Fed plug itself into this dialogue when it is in fact a key player in this market? If we decide to go for a solution that includes some kind of central clearing, how should the Fed be involved in that clearing house? Right? What kind of impact will the Fed's regulatory oversight of primary dealers have when its own balance sheet is essentially very prominent in Treasury management? What kind of oversight will it need to give to those who are essential in maintaining the liquid liquidity of these treasuries? So, these are all important questions. I think we need to think about them because the Fed is not going away here in this space, particularly in the absence of solid ex ante information constraints, whatever. We need the Fed now as a backstop the Treasury market, there's no avoiding that.
Beckworth: All right. The SEC and FINRA?
Yadav: Great. SEC and FINRA are the key regulators for the securities firms. That obviously includes the HFT players that are now critical liquidity providers in this market. They also oversee major platforms that host treasuries trading, for example, BrokerTec, which sees almost something like, almost 80% of inter-dealer treasuries trading on its platform. So the SEC and FINRA regulate securities forms, as well as the SEC thinking about BrokerTec in its space.
Beckworth: All right. And the fifth one is the CFTC.
Yadav: Yeah. The CFTC is our primary regulator for derivatives, including of course, treasury related, treasury link derivatives. So that's where it comes into the its own. It's the derivatives regulator. And so it plugs in from that perspective.
Beckworth: And then you mentioned a few others sprinkled on, the FDIC, the OCC, the Federal Reserve itself, the banking regulation side, because banks are engaged with Treasury securities as well. So it's just a smorgasbord of different regulatory bodies. They each have their hand in a certain part of the cookie jar, trying to get their piece out. And as you said, it creates some challenges. There can be turf battles. This might be institutional morass, not able to quickly get information out. So it's created challenges. And on top of the regulatory challenges, you also outline some structural shifts in the Treasury market itself. So maybe walk us through them. You've touched on some of them already, but go ahead and walk us through them again and what has happened.
Yadav: Yeah. So the primary dealers have historically been completely dominant in all parts of this market. So the primary market, the dealer to client market, the inter-dealer market, it's been their domain. Now, of course, they're facing a lot of pressure from HFTs in the inter-dealer space. And indeed there's talk of some HFTs muscling into the dealer to client space. So that means that in some ways, the "franchise value" of being a primary dealer is being eroded by a lot of competition coming in from the new traders, the new HFTs, which are smaller firms, securities firms. They are traditionally less regulated than banks are. Their balance sheets tend to be smaller. Some of them are not even regulated as broker dealers, so they don't report their data to FINRA. So they have even fewer regulatory obligations on them.
Yadav: So it's a much more eclectic market. And it's one where the primary dealers are under pressure from the competition side. And of course, as we know from the banking side, they also have balance sheet pressures to ensure that they have sufficient cash and securities and compliance reserves on hand, as they need to, to maintain their safety and soundness. So they're facing pressures on that side as well as on the competitive side and the inter-dealer market. So that's been one of the arguments that has been put forward to say that primary dealers may in fact be losing an incentive to play a deeply engaged role in maintaining the health of the market as they might once have done.
Beckworth: Well, let's talk a little bit more about the high frequency trading that's now so important to the Treasury market. And I think most of us, including myself, understand high frequency trading has applies to stocks. You've read books, and lots of stories told. But how does it work with treasuries? You mentioned they're typically smaller firms, do they typically have similar algorithms so they would all react together to a certain time? I mean, how does that system work?
The High Frequency Trading System
Yadav: Yeah, I mean, HFTs are commonplace in the market, in the equities market. We've been very experienced with HFTs. It's a function of a market that's speeding up. Computers, electronics, communications, data, everything, it's going to happen in treasuries. It has happened in treasuries. It was inevitable. And of course, HFTs have important advantages. I think there's studies that talk about the efficiencies, the lower trading costs, the lowered spreads in the US Treasury markets. So, these can all arguably be seen as positive externalities in some respects. But equally like every other market, there are also fragilities that are being introduced.
Yadav: For example, one can have HFTs because they have smaller balance sheets, leave the market very quickly because they have to. They may use algorithms that are potentially similarly programmed. So, just to give you some examples, this is not always the case, but to give you some examples. When the flash rally happened in 2014, the 2015 report stated that there was a large amount of wash trading that was happening in that episode.
Yadav: And what that means is that folks were just trading with themselves all the time, all the time. It was almost like the algos were just programmed in a way that was just on this wash cycle in the washing machine, just repeatedly just trading in the same way. I mean, that's a function of the fact that you're pre-programming these things to respond to almost all category of evolving events in the market. And sometimes they may just not be able to cope.
Yadav: And if they can't cope, you have to press the kill switch. Now, interestingly, in the flash rally episode, HFT players actually remained on the market. They just reduced how many orders they were posting. The report stated that the primary dealers actually left quicker, but the HFTs did remain. Now, in the March, 2020 episode, the HFTs, the level of liquidity became much, much shallower. And so we've seen episodes here where liquidity providers of all shapes, in fact, HFT’s primary dealers are becoming increasingly temperamental in their willingness to provide ongoing liquidity to the market.
Beckworth: Do the high frequency trading firms make the job of the regulator even more challenging?
Yadav: Yeah, I think they do because it's a different category of firm. So, they are a securities firm that tends to have smaller balance sheets. They are new to the space in treasuries relatively. In addition, they're a little bit harder to regulate because they're playing a important securities market function, but in a very systemic space. So they are trading an instrument that is a lot more homogenous than, say, trading a whole bunch of different equities would be. Treasuries represent the credit risk of a single issuer, which is the US government in various different maturities. But fundamentally, that's it is. But they're trading a fairly homogenous asset.
Yadav: It's a systemic asset. They are smaller securities firms, but they bring advantages like liquidity, and speed, and efficiencies, and so on and so forth. So how do you calibrate regulation to take advantage of some of their positive effects, but also make sure they have the resources to maintain market integrity and safety within a highly systemically important market where it is essentially the credit risk of a single issuer, so the impact of a single event can potentially be market-wide? [The HFT market is] a really hard category of firm to regulate because regulators like the fact they bring liquidity, they lower the cost of trading. But it's a systemic market. So how do you calibrate that regulatory cost to introduce competition without putting the market at risk? It's such a hard problem. It's a really difficult puzzle.
Beckworth: Well, that's a great segue into your proposals. So you have a proposal for improving the public oversight of the Treasury market. So why don't you spell that out for us?
[The HFT market is] a really hard category of firm to regulate because regulators like the fact they bring liquidity, they lower the cost of trading. But it's a systemic market. So how do you calibrate that regulatory cost to introduce competition without putting the market at risk? It's such a hard problem. It's a really difficult puzzle.
Improving Public Oversight of the Treasury Market
Yadav: Wonderful. So, a main proposal here is to get the FSOC involved in coordinating. I think it's a fairly mild mannered proposal, to be honest, I don't think there’s anything very radical about it. I think Treasury Secretary Yellen had introduced her five priorities. I think it should be added as one to bring oversight of the Treasury's space to coordinate oversight of the Treasury's space and get all the regulations in same room in a much more fundamental way, create a really workable, and reliable, and codified memorandum of understanding, much deeper and regular than it is today.
Yadav: Reduce the institutional barriers to sharing information and just make that sharing happen. Come up with a plan. We can do better than this. So I think our first, my first proposal here is to coordinate under the FSOC. In a related paper, *Fragile Financial Regulation* with Pradeep Yadav, I also talk about having a formative market making obligations on primary dealers and HFTs, larger HFTs to require them to the best possible degree that they can manage to stay on the market in times of trouble.
Yadav: This harks back to a market structure that used to be operational in the '80s in the New York Stock Exchange, for example, where market makers had to remain on the market to provide liquidity, even in a crisis under the so-called specialist system. I think, and we think in this paper that having a formative market making obligations in Treasury markets is germane to the nature of the market. It's a market that needs to remain liquid, even when there's a crisis. We're not saying that primary dealers in HFTs stay there 'till they become insolvent, but at least they put some resources ready to deploy in the event that there is some crisis in the Treasury market that constrains its liquidity.
Beckworth: And FSOC has the statutory authority to do this type of work?
Yadav: Well, it would have work with all the other agencies. So it by itself doesn't, but it would have to engage the SEC, the Fed, the OCC and others to promulgate the rules or come up with a new rule, or maybe Congress needs to do something. I mean, come up with a new law that would ensure that the Treasury market is equipped to deal with the emergent risk that it's facing, and the enormous demand for its services, particularly given the kind of crises that we've been facing lately.
Beckworth: Well, I had Kate Judge on the show before, we talked about her earlier. And one of her concerns is that FSOC hasn't been used as well as it could be, particularly in the previous administration. But going forward, do you think it'll have the teeth, have the interest of leaders to really use it in the direction you want it to go?
Yadav: Yeah. And I think there should be bipartisan support for this. I mean, I think the FSOC has traditionally been vilified in certain quarters. And I think the FSOC has a lot of gains that can be offered at least in bringing regulators together in the same table. It has the authority to do that. It should be doing that, particularly in this space. And everyone of all stripes, of all colors, political colors should be on the table here to really strengthen the Treasury market. That I think is, should be appealing to every political stripe.
Beckworth: Yeah. It seems like a very non-partisan suggestion.
I think the FSOC has a lot of gains that can be offered at least in bringing regulators together in the same table. It has the authority to do that. It should be doing that, particularly in this space. And everyone of all stripes, of all colors, political colors should be on the table here to really strengthen the Treasury market. That I think is, should be appealing to every political stripe.
Yadav: I would imagine.
Beckworth: Yeah, for sure. All right, let's go to your next proposal. And that relates to increased use of central clearing.
Increased Use of Central Clearing
Yadav: So the central clearing solution and Darrell Duffie at Stanford, he's been very actively researching and think about this solution as well. This solution, I think it's a very complicated one. I think we need to give very serious thought to central clearing. Obviously, central clearing has brought multiple advantages to different markets. It's a key part of our market that maintains a lot of market integrity. And maybe you can do the same for treasuries. But what we have to be careful about is ensuring that we don't create a clearinghouse that is the most systemically risky institution anywhere in the galaxy.
Yadav: I feel that this could be the intergalactic death star if things go wrong. And this is the last possible outcome that any of us would want. So this solution has to be dealt with, with kid gloves. There's a lot of interest in thinking about this. The regulators are intrigued by this solution. I think there's a lot to be said for that and thinking about it. But we need to be really careful.
Yadav: And Dan's work here on clearinghouses has to be lauded just to highlight the immense riskiness that these institutions can pose. And so with the Treasury market potentially becoming a market that's centrally cleared, including maybe the repo market attached to it, this would become the most systemically risky institution anywhere, anywhere, anywhere. And so we need to be so careful about that.
Beckworth: Okay. Just a few more details on central clearing. So this is the idea that there'd be a clearinghouse where everyone would come together, and they would trade through this intermediary, this central clearing intermediary. And doesn't the Treasury already do some of that? So the question is just increasing the degree of its use. Is that right?
Yadav: Yeah. So there is clearing in the US Treasury market today. The Fixed Income Clearinghouse Corporation, the FICC is a clearinghouse that clears trades for FICC members that are the primary dealers. So when primary dealers trade with one another, that gets to be centrally cleared. But only about 13% of all secondary trades in treasury I think are centrally clear. I think that's the right number. And it's very ad hoc.
With the Treasury market potentially becoming a market that's centrally cleared, including maybe the repo market attached to it, this would become the most systemically risky institution anywhere...And so we need to be so careful about that.
Yadav: Now, as anyone who studies central clearing will tell you, that kind of solution is a nightmare. It's a nightmare because you want clearinghouses if they're clearing to clear as much of the market as possible. That helps netting. That helps the clearinghouse make sure that it's has enough transactions. It can do all sorts of set off, and risk management, and monitoring, and so on, and so forth. So the FICC here are clearing only a small fraction of the market is hugely problematic just from a systemic perspective because it just does not get to see the market as a whole. It's just doing a tiny portion. It's gets all the risks of clearing without many of the benefits that a larger clearing mandate would provide, which is set off risk monitoring as well as fees, obviously, for those who are using the clearinghouse.
Beckworth: Now, this is all very fascinating. This discussion of central clearing is at it applies to treasuries. Because I remember going back to the great financial crisis. And Dodd-Frank, one of the things they aimed to do was to bring derivatives in out of the shadows so we could know what was happening. So they made use of central clearing. So we have better data on derivatives. But then all that risk gets concentrated in those nodes of central clearing. And this is the same concern you're raising for Treasury. If we did increase central clearing that the central clearers would, I think, would be potentially a very systemically important financial entity. Now, is the hope that if it did come under pressure that the Federal Reserve would step in and save the central clearing entity?
Yadav: I think that's basically the only solution that we are going to have. There is no doubt here that the Fed would have to come in and contain the damage. This is without question the doomsday scenario of all doomsday scenarios. To be honest. And so our central bank would have no choice but to protect, not just the Treasury market, but the entire economy, global economy, everything, our existence, our planetary life to have to basically step in and do it.
Beckworth: So I like the death star analogy, it’s very, very good here. I mean, it would be a very-
Yadav: I can't help myself.
Beckworth: ... catastrophic.
Yadav: I had to get it in somehow.
Beckworth: So you're a big Star Wars fan. Huh?
Yadav: I'm a big sci-fi fan. Everything, yeah.
Beckworth: Okay. Very nice. Well, sometime I'll have you on to talk about the economics of sci-fi. But back to Treasury market, let me ask another question then. The Fed recently introduced its standing repo facility and it also has introduced during the pandemic and it's going to keep it permanent now this Foreign International Monetary Authority repo as well. So do these things help in any way some of the concerns you have with the Treasury market?
The Impact of the Fed’s New Facilities
Yadav: Yeah. I mean, I think they provide this sense of security. I mean, there was a lot of foreign selling pressure in March, 2020. I think approximately $275 billion decreased in the NY Fed. I think data suggested that over the month of March. So the extent to which we can mitigate some of the liquidity pressure in the Treasury market, that's a good thing, obviously. Furthermore, having a standing repo facility provides that liquidity backstop that can be super helpful for primary dealers and others. But again, it's not an excuse or it's not a way out of doing some real research and reform of the underlying market structure. That is not solved by simply having these ex post facilities that simply mitigate some of the harsher effects of a very fragile market structure. So, I think they go some ways towards at least providing some degree of peace. But they don't solve the need for us to do some real work. And we're not there yet, not even close.
Beckworth: So it sounds like you might be concerned that the standing repo facility might induce complacency in addressing some of these structural needs?
Yadav: Yeah. My worry is that it reduces the urgency. I mean, regulators have stepped up, to their credit, they have been thinking about this more recently. And I think the inter-agency working group and other regulators have highlighted the need for Treasury reform. But it's so hard to make it happen, given this very fragmented structure that we have in place, the complexity of the US Treasury market itself, it's interconnection with the repo market. And furthermore, of course, its basic centrality to the global economy. So it is a very tough nut to crack, but I think that if we do nothing, we are in a position where we might, in the long term, harm the reputation of the US Treasury market as the safest market in the entire world.
Beckworth: That's interesting the way you frame that, that it's hard to get anything done and we don't want to rely just on the Fed. But the Fed has been the only game in town it seems like on increasing scale. Right? I mean, think of climate change, we're asking the Fed to do more and more because Congress isn't doing enough. Or we're asking the Fed to step in and save the global shadow banking system because we don't have approach there as well.
I think the inter-agency working group and other regulators have highlighted the need for Treasury reform. But it's so hard to make it happen, given this very fragmented structure that we have in place, the complexity of the US Treasury market itself, it's interconnection with the repo market. And furthermore, of course, its basic centrality to the global economy. So it is a very tough nut to crack, but I think that if we do nothing, we are in a position where we might, in the long term, harm the reputation of the US Treasury market as the safest market in the entire world.
Beckworth: And so here in this particular area, we would love these regulators, we would love Congress to get board and get a better sense of, and better regulation on the Treasury market. But in the absence of that, we're good because the Fed has a standing repo facility and we'll backstop the Treasury system. And that's not the ideal outcome.
Yadav: I would imagine not. I think everything that we have learned in regulation in Dan's work, and Morgan's work, and Kate's work, and others within our ecosystem highlight the need for are avoiding that kind of moral hazard. We should be doing more to strengthen our system ex ante than always having to rely on exposed solutions that are open ended.
Beckworth: All right. One last question before our time is up. So one of the other proposals that's been put out there is for regulators to loosen or relax the supplemental leverage ratio on banks. So, have more, I guess, excess capacity on their balance sheets to absorb some of these times when they're stressed in the Treasury market. What are your thoughts on that proposal?
Relaxing the Supplemental Leverage Ratio
Yadav: So the SLR is currently being studied. So the SLR ratio is currently being studied to see whether or not it needs to be lightened in times of trouble. Now, obviously when we had COVID, we provided SLR relief to ensure that banks would be able to take on more treasuries, as well as take on more deposits. It has been very unpopular, as well as popular with the banks, unpopular, too. So there's a lot of debate surrounding it, but I think there's a need for more data here. And particularly for thinking about ways in which maybe the SLR might be nuanced in ways that could potentially allow it to be relaxed and particularly troubled and struggling times whilst otherwise being maintained to strengthen and secure the banking system.
Yadav: But we know that primary dealers and others have huge importance in providing liquidity for treasuries that can exert enormous pressure on their balance sheets, to the extent to which they might need some temporary safety valves, that is currently having to be studied. And I think it needs to rightly be studied to figure out what kind of potential nuanced solution we can bring into place. So it's not like we have to say, "No SLR versus SLR." Maybe there's some kind of way to nuance the application to make it more calibrated in ways that ensure systemic stability, but also ensure that the Treasury market remains viable, and healthy, and properly intermediated.
I think everything that we have learned in regulation...within our ecosystem highlight the need for are avoiding that kind of moral hazard. We should be doing more to strengthen our system ex ante than always having to rely on exposed solutions that are open ended.
Beckworth: If I recall correctly, the bank of England changed its SLR, but it adjusted it such that the capital requirements didn't go down overall. So there's a way to tweak it and maintain a middle ground there. The other thing that's striking about the SLR to me is I've had Bill Nelson on the show a few times, and he used to work at the Board of Governors as a Fed staff member there.
Beckworth: And he noted how, when the SLR was being written, I guess, to around 2014, 2015 when it first was being put out there, the expectation was that Fed's balance sheet would get much smaller. And instead, it's gone the other direction. And larger the Fed's balance sheet is, the more reserves banks have to hold, and it becomes a binding constraint. So they were not at all imagining the SLR becoming the binding constraint it can be in crisis. So I think it is important that we update our thinking on this issue, take a middle ground. But that we understand that how a law was written back four or five years ago may be outdated in some regard.
Yadav: And I think one of the aspects that we need to also think about is that many financial institutions have very automated compliance systems. So the SLR exists as a compliance constraint. So that can hinder the ability of banks and others to be very loose with their ability to quickly provide the resources they may need to the relevant desk, and so on, and so forth to be able to marshal the resources they need in a hurry.
Yadav: And so understanding the institutional picture of the reality that primary dealers are facing from the compliance side, how quickly they can deploy resources, how fully they can do so. And understanding the public role that they face vis-a-vis intermediation could perhaps motivate us to think about ways in which we are calibrating the solution without undermining and jeopardizing systemic safety, which is obviously paramount. And we've achieved so much from the Dodd-Frank Act to get it to the place that we're at today.
Beckworth: Well, on that note, our time is up. Our guest today has been Yesha Yadav. Yesha, thank you so much for coming on the show.
Yadav: David, thank you so much for having me. It's been such a pleasure. I have learned so much from your questions and I would be so excited to do it again.
Beckworth: Alright, thank you.
Photo by Chip Somodevilla via Getty Images