Banner for 2025 Financial Markets Conference

May 19, 2025

Policy Session 2: The Increasing Role of Nonbank Institutions in the Treasury and Money Markets

Dallas Fed president and CEO Lorie Logan moderated the conference's second policy session, which focused on the increasing role of nonbank institutions in Treasury and money markets. Panelists included Lou Crandall, chief economist at Wrightson ICAP; Deirdre K. Dunn, head of global rates at Citigroup Global Capital Markets Inc.; and Nate Wuerffel, head of market structure at BNY.

Transcript

Lorie K. Logan: All right. Good afternoon. Welcome back from lunch. It's a pleasure to be back at this excellent conference, and I just want to start by thanking Raphael for inviting me to be here; and thank you to all the organizers from the Atlanta Fed for putting together such a great panel. It's great for me to be able to introduce the panel, and I'll start with Deirdre Dunn, head of global rates at Citigroup Global Markets, and chair of the Treasury Borrowing Advisory Committee.

We also have Lou Crandall, chief economist of Wrightson ICAP. And my former longtime colleague at the New York Fed, Nate Wuerffel, who's now head of product and market structure at Bank of New York. It's great to have all of you here, and I'm really looking forward to the conversation.

Our topic this afternoon is a timely one: the role of nonbank financial institutions in Treasury and money markets. But before we jump into our dialogue, I thought I'd offer just a couple of remarks to frame our conversation this afternoon. The Treasury market and money markets sit at the very core of the financial system; I know this group knows that quite well.

The Treasury market finances the US government, provides a safe and liquid asset relied on by investors worldwide, and creates a benchmark for broader long-term interest rates. Money markets establish overnight, risk-free interest rates that are building blocks for all other asset prices. They finance a wide range of assets, and they are where the Fed implements the stance of monetary policy. And these markets, of course, are tightly linked, because one of the main money markets is the repo market where Treasury securities are financed.

These markets are extraordinarily deep and liquid, as befits their systemic role, but they're not invulnerable. Observers have cited potential fault lines, including limited intermediation capacity, buildups of leverage, and uneven risk management. When economic shocks occur, investors often seek to transfer large amounts of risk and raise large amounts of liquidity in both the Treasury and money markets, and such a surge in demand for intermediation requires an elastic supply of intermediation.

Yet we've seen, since the global financial crisis, that the balance sheet capacity of bank-affiliated dealers has not kept pace with growth in the amount of Treasury securities outstanding. So, there is the potential, as we saw at the onset of the pandemic, for the demand for intermediation to overwhelm the supply of intermediation and create market dysfunction.

So, in principle, approaches to enhancing the resilience of intermediation capacity could address both banks and nonbanks, and I'm sure our panelists and I could fill an entire panel with ideas on bank intermediation. But our focus today is going to be the rest of the market—the nonbanks, and their increasing role in the Treasury market and in money markets—and I'm looking forward to a robust discussion of the state of Treasury intermediation by nonbank financial institutions, and what we can be doing to strengthen it.

Excessive leverage can amplify challenges from insufficiently elastic intermediation in stress episodes—perhaps precisely because repo markets ordinarily function so well and Treasury securities are so safe and liquid, it's possible for NBFIs to obtain large amounts of leverage against them. Yet levered positions can be prone to unwind rapidly, adding to intermediation demands. Buildups of nonbank financial institution leverage in trades, such as the cash-futures basis, can therefore be cause for concern if the risks are not managed appropriately.

Some observers have voiced worries about such a buildup earlier in the year, so it was reassuring to me to see that the basis trade did not materially amplify the market volatility that we experienced earlier in April. Strong risk management mitigates vulnerabilities from leverage, and the SEC mandate for broader central clearing marks an important step to strengthen risk management in the Treasury cash and repo markets. Data collected by the OFR shows that a large fraction of non-centrally cleared bilateral repos are conducted with no haircuts, and in some circumstances other components of a cash borrower's portfolio may effectively provide the margin on these trades—but it's not always the case, and repos that are margined on a portfolio basis could be more difficult to move to different counterparties in times of stress.

Broader central clearing, I believe, will ensure that risk management is stronger, more uniform, and therefore more resilient. So, in our conversation today I'll be interested to hear how each of our panelists think the transition to broader central clearing is proceeding, and how it's influencing the overall structure and resilience of the market.

And lastly—as my Fed colleague Roberto Perli, manager of the System Open Market Account, noted in a recent speech—resilient funding liquidity makes the market as a whole more resilient. A strong Federal Reserve monetary policy implementation framework forms an important part of that resilience by ensuring ample liquidity will remain available across money markets at rates within or near the target range for the fed funds rate.

It's my view that rate control is not just about keeping the fed funds rate in the target range. The fed funds market is small, and the FOMC's desired stance of monetary policy must transmit smoothly into larger and broader markets, especially the repo market. So, the Fed's rate control framework has two main components: the ample supply of reserves generally keeps money market rates close to the interest rate on reserve balances, and our standing facilities—including the discount window, the standing repo facility, and the FEMA repo facility—help prevent rate spikes in the event of large, unexpected shocks to reserve supply, demand, or the distribution.

The ceiling tools have proven their effectiveness over time; however, I do see some opportunities to strengthen them going forward. Depository institutions have greatly improved their operational readiness to borrow from the discount window, and I think we should continue to reinforce the value of that readiness, and readiness also is a partnership. At the Federal Reserve Banks, we're working to enhance our capacity to serve customers efficiently when they come to borrow, drawing on feedback we received from a recent request for information by the Board of Governors. And as part of that, we're working with the FHLBs to improve interoperability between our lending facilities and theirs.

I think we should also continue to emphasize that borrowing from the window is an appropriate way for healthy banks to meet short-term funding needs, not something investors or rating agencies or supervisors should criticize or question. I think we can also enhance the standing repo facility; as my colleague Roberto described in his recent speech, experiments and market outreach by the New York Fed's Open Market Trading Desk have found that conducting and settling the standing repo facility operation in the morning, in addition to current afternoon timing, makes the facility more effective by addressing intraday funding needs.

And I'm pleased that Roberto announced earlier this month that the Desk plans to soon introduce regular early settlement standing repo facility operations. Central clearing of the standing repo facility operations would also make the facility more attractive and enhance rate control, in my view—and that's because central clearing would allow bank-affiliated dealers to net down their balance sheets when they borrow from the standing repo facility and lend onward to other nonbank financial institutions.

Desk outreach has found that the lack of netting is one of several reasons why counterparties currently report high hurdle rates for borrowing from the standing repo facility, along with reporting requirements and supervisory considerations. I'd also note that although the SEC's clearing mandate does not apply to the Fed, centrally clearing Fed operations on a voluntary basis would align with and support that transition.

So, I'm looking forward to your thoughts on each of those, and your own perspectives on how to make the Fed's tools as effective as possible. So, hopefully that lays out the framing of the issues I thought we could cover, and I thought we'd just start the conversation with the big picture: How did markets—and especially nonbank participants, whether it's end users, hedge funds, PTFs (principal trading firms)—how did they handle the high volatility that we saw in early April, and what do you see as the outlook for these different types of NBFIs in the years ahead?

And Deirdre, I thought I'd start with you on the Treasury cash markets, and then go to Lou on the money markets and get a sense of what we're seeing in those NBFIs, how they handled April, and what you see going forward.

So, Deirdre, over to you.

Deirdre K. Dunn: Yes, thank you, Laurie; and thank you all for having me. I'm very excited to be here with my esteemed panelists. So, early April really was exceptional. It was a pretty short duration of exceptional volatility—eight or nine trading days—but the scale of the moves that we saw, and the scale of the breakdown in traditional correlations that we saw, made risk management and risk intermediation very challenging for, really, a wide variety of market participants. And I think we saw volume spike, we did see significant risk unwinds, and we saw an inability to sit with risk that was really driven by some of the breakdowns in how people traditionally hedge things—and also exacerbated by, I think, the style of news flow that was coming out in that period of time.

We can talk about the Treasury markets, but I think it's important to talk really about the broader markets at the time as well. Where we saw the most significant dysfunction—thankfully, in some ways—was not the Treasury markets, and I would make a distinction between market pricing and market functioning. We saw a significant change in market pricing in the US Treasury markets, but where we saw more of the dysfunction, I would say, was actually away from the Treasury markets.

In the rates market specifically, probably what people focused on most was the unwind of the swap spread trade, which is mostly held by opportunistic and levered players. The breakdown between swaps and Treasuries that we saw was significant; swaps and futures also was significant. And you saw, with the VaR shock, many participants take steps to reduce their risk exposure there.

The basis trade was frequently reported and looked at, but really was very stable, all things considered, relative to everything else in that market; and Treasuries performed very well. I think there were pricing concerns in Treasuries, as investors began to talk about really the breakdown in the traditional countermoves that you see between equities and rates; seeing US Treasuries sell off as equity sold off was concerning, and as the dollar was weaker, a surge in golds, in contrast of those two things, had some investors questioning how strongly US Treasuries—where is the safe haven status? Is that a risk? Are we at risk of investors looking to potentially allocate less cash to US Treasuries, or US financial assets overall?

Those questions really did cause some consternation in the market, and filtered through to the US Treasury market in terms of pricing and price performance. But in terms of market functioning, we were able to see very significant risk go through in off-the-run Treasuries, and with on-the-runs we were able to see the market continue to function with reasonable degrees of price transparency and risk clearance. And I think a lot of that is down to funding markets, and the stability that we saw in funding markets—which really is a credit to some of the facilities that exist now, what we've learned and put into practice over the past 10 years as we continue to look at the US Treasury markets in the official sector and in some of the private and public partnerships.

But I'll let Lou comment more in detail on the funding markets.

Lou Crandall: Thank you; and I thank the organizers for inviting me. So, Lorie referenced a speech by Roberto Perli from 10 days ago, and he went through a lot of metrics about how the Treasury cash market was stressed but functioning. The inevitable volatility and prices that you get from a shock like that led to an erosion in some liquidity metrics like depth of market, but it was not disproportionate to the increase in price volatility.

He did not go through the same sort of exercise with the funding market, because frankly, there weren't any interesting data to talk about. The rates were modestly elevated for a couple of days relative to the seasonal trend at the peak of the volatility; you didn't see dysfunction in the cash market, but you saw severe strains. You really didn't even see that in the funding market, and as Deirdre was saying, that was critical to maintaining the stability of the cash market.

In talking to people about why the funding market performed so well, the thing that comes to mind first and foremost is that thank goodness this happened in the first few days of a calendar quarter, rather than the days leading up to a quarter end. Because we've done a lot over the last few years to mitigate the really irrational distortions that we allow to emerge on each quarter end; but they're still there, and they're still an accident waiting to happen. If that price shock had occurred just when we were at the seasonal low point in funding market liquidity, the outcome probably would have been a lot messier.

So again, the window dressing contractions are still an accident waiting to happen, but we dodged it by a few days this time around. The other thing I heard from some customers—it's interesting; as you talk, I'm sure this has been your experience as you talk to different people about how they came through this process: everybody's experience seems to have been somewhat different. One thing that people noted, though, is that as they were dealing with individual balance sheet constraints, the fact that we had moved so much repo activity into voluntary clearing through the fixed sponsored repo program, created elasticity in individual conversations about balance sheet access.

Logan: Bringing Nate into the conversation: Nate, you've been looking at these issues of nonbank financial institutions in the Treasury market, I think going all the way back to 2014, during the flash rally. And maybe putting together the conversations between Deirdre and Lou, maybe provide a little bit of perspective of who are the large, nonbank financial institutions in the market that we're really thinking about, and how do you see that evolving? And both of them mentioned something that didn't happen during that April period; we didn't see big unwinds of leverage, perhaps because the funding market was so stable, particularly in the cash-futures basis trade. Why do you think that is?

Nate Wuerffel: Yes, sure. Thanks, Lorie, and thanks to President Bostic for this excellent conference. I think, when you look at April, the uncertainty really started around trade and tariff policy, but then that spilled over into a number of what I would say are fundamental things that affect interest rates—outlook for macroeconomic growth, inflation, budget and fiscal policy—and so, it's really no surprise that you saw the amount of price volatility.

But as I think Deirdre and Lou were saying, when markets really get dysfunctional, I think of three things you tend to see: one being high levels of price volatility, the second being a deterioration in market functioning indicators, and then I put in its own third category, disruptions in funding markets (when funding market pressures start to grow). And that's what we saw in most of the large dysfunctional periods in the past in the Treasury markets.

And it's that last piece that I think is also key to understanding the nonbank financial participation, and how they respond to stress, too; because most of the leveraged transactions in the Treasury market involve some funding component, and when that funding component gets very expensive you have to unwind those transactions. That's what we saw in the cash-futures basis trade in the pandemic; it's likely perhaps what happened on some of the swap spread trade transactions, but the dog that didn't bark was the cash-futures basis trade, in this case, among probably a number of other dogs.

But I think the fact that funding markets were as stable as they were helped with market functioning. Maybe one other thing I would add to Lou's list is, the unsung hero might be also the ample reserves framework itself, that there was a lot of liquidity in the system.

When I think about the Treasury market and how it's evolving, I was looking at some of these statistics in the last decade; from 2015 to present, foreign holdings of US Treasuries have dropped from 50 percent of the market to 30 percent of the market, and the remainder has been filled really with a lot of nonbank financial institutions—so mutual funds, money market funds, and hedge funds. Hedge funds now own $2 trillion, pensions are $1 trillion, money market funds are $7 trillion—so, combined you're talking about 30 percent of the market is now owned by these nonbank financial market participants.

I think those changes in composition aren't necessarily a sign of a problem, per se, in the Treasury market; in fact, I think it means that the Treasury market—the safest, most liquid market in the world—has a very deep and broad participation around the world, and so that's a really positive sign. But it does mean that the behavior of holders of Treasury securities is going to change as the composition changes, so if you have a lot of nonbank financial participants, or you have a lot of levered holders of Treasury securities, then their response to those three variables, whether it's volatility or liquidity or funding pressures, is going to change how the Treasury market itself responds.

And so, there's been a lot of work that's been done, as you mentioned—whether that's the CFTC's MRAC committee, the Treasury Market Practices Group; the TBAC (Treasury Borrowing Advisory Committee) has looked at these issues. And thinking about what are effective—what does "good" look like, in terms of managing the risks of, for example, nonbank participation in the Treasury market, or leverage and hedge fund holdings in the Treasury market?

Logan: One of the groups you didn't mention in that category were the PTFs. Any observations or lessons learned, either from Deirdre or Nate, on the way PTFs behaved during the April volatility?

Wuerffel: I'm happy to start. I'm sure both of these two will have views on it, but in 2014 when the flash rally happened in the Treasury market, that was the first time I think that eyes were really open to the fact that electronic trading had entered in force in the US Treasury market, and interdealer space in particular. And there we saw this very rapid movement—sort of like equity market flash crashes, but in the Treasury market.

And in the interdealer space—so, there's two segments of the Treasury market: between dealers and their clients (that's dealer-to-client trading; it's about half of the market), and then there's dealers offsetting risk with other dealers in the interdealer market. That interdealer market—half the Treasury market, half the trillion dollars of activity on a daily basis—is pretty much dominated by electronic trading in the first instance, but then also, 60 percent of it is actually high-frequency trading firms, or principal trading firms, as they prefer to call themselves. And that activity is risk-sensitive, so the way that those entities play in the Treasury market is that they're very sensitive to price, and they adjust their risk by offering less depth in the market.

Depth is basically how much you can trade before the price moves, so a high frequency trading firm can move its price very, very rapidly, and they'll just shrink the amount of depth that they offer. Dealers tend to widen bid-ask spreads, but what we see in these periods of volatility is that high-frequency trading firms tend to move their depth very quickly, and we saw depth really decline in April.

Dunn: I think depth declined at the same time as volumes increased dramatically. The sheer volume going through the system in Treasuries far eclipsed anything we saw before; it didn't even come close . . . it blew by COVID, without looking back. I would also highlight that the electronic trading and the PTFs that you highlighted are mostly concentrated in on-the-run, or some of the next few off-the-run, securities—where, obviously, as the scale of just sheer US Treasury debt outstanding rose, the percentage that also is considered off-the-run (that trade tends to trade slightly more bespokely) continues to grow as well.

And perhaps you could make the argument that that segment is almost more dependent on, or contingent on, funding availability; and I think what we saw in COVID actually really was much more pressure and stress on the unwind in the off-the-run market as people really looked to raise cash. That was a very significant distinction from this.

The other thing that, as Nate was speaking, it occurred to me that I wanted to point out is that the scale of economic uncertainty that was introduced in early April I think was a significant challenge, not just for the Treasury market—and in some ways, less for the Treasury market than other markets—but for what is the broader economic outlook. And so, I think if you felt on March 30th that you were standing on solid ground with what you would say your economic outlook was, or your outlook for inflation, or your outlook for growth—by a week or so later, the entire marketplace was really questioning a lot of that.

And so, as a financial market participant who's used to expressing views on the economic outlook through financial markets, when you're much less certain the first thing you do is probably reduce risk to that; you otherwise unwind expressions of that risk. And to be honest, I think other markets are probably more levered and more direct expressions of that risk, whether it was inflation markets—the TIPS market was much more challenged than the nominal Treasury market—or whether it's in very short-dated swaps markets and the moves there, and so that might have been a contributing factor as well.

Logan: So, I wanted to talk a little bit more about the concerns around leveraged Treasury trading, and that goes back long before the period in April. We've seen multiple reports on these risks. So, the TBAC did a study on the cash-future basis trades; the TMPG has released best practices for Treasury repo; the CFTC advisory committee report on effective risk management practices for the basis trade. So, there's been a lot of work, a lot of focus, on these types of risks in various public and private-public partnership organizations.

Nate, how do you see these risks evolving? What impact do you expect these different best practices to play? Do you think we didn't see that type of issue because of the work and attention that's been done coming through these various efforts, or do you think the issues were different and it's an area we still need to be closely focused on?

Wuerffel: Yes, I think risk management practices are still an area that needs focus. You mentioned the collection of margin; so, margining is a very, very common risk management practice to guard against counterparty credit risk in times of market volatility. So, if tomorrow your counterparty is not there and you were expecting cash or a Treasury security, you're now facing counterparty credit risk. And so, margining those transactions can guard against that, and that's typically why it's used in so many products. But in bilateral repo transactions, it's often not; 70 percent, I think, was the OFR statistic for bilateral repo. That has a zero margin or haircut.

Now, there could be a number of reasons for that. It could be that there's a series of transactions between, let's say, a dealer and a hedge fund that are managed on a portfolio basis. And so, the Treasury piece is zero margin but there's other mitigating factors, and I think that there are different ways and techniques to manage risk between counterparties. But I think the question in my mind is: How consistent are those practices across the industry? And to the degree that you experience a default—and there are a lot of different types of practices you may have—you may not be able to address the fallout from that kind of default in a consistent way.

The TMPG best practices that have been proposed basically call on market participants to use margin in addition to other risk management techniques, and that's for all Treasury repo. When you look at all Treasury repo, there's about $5.5 trillion a day; $2 trillion of that, roughly, is centrally cleared already—so that has really good risk management practices in place, because CCPs collect margin on those transactions.

But for the stuff that's not centrally cleared, and even for trades that are centrally cleared, but it's really the dealer, the member, that has to post margin; it's not their clients that have to post margin. So, even in a centrally cleared world, I think you're going to need best practices for that around the collection of margin; and the TMPG has a good set that are out there. The MRAC committee goes beyond just margining practices and talks about counterparty onboarding processes, and doing stress testing, and market and liquidity risk, so a whole range of things you can do to manage the risk in the space.

Logan: Dierdre, I know you worked on the Treasury Borrowing Advisory Committee's report on the cash-futures basis. Any other key points you'd want to raise to the audience in terms of risk management attention, or issues that you all hoped to get more focus?

Dunn: Yes; I would say the TBAC really welcomed the opportunity to dig in on the Treasury futures basis trade, because at the time we felt that the industry focus was squarely on hedge funds, ownership of US Treasuries, and leverage and the growth in that segment. And as Nate pointed out, it's not necessarily a bad thing, and many people would actually argue that the hedge funds are providing a significant transmission mechanism, in terms of what Treasury actually needs to issue to fund themselves (which is cash securities), and what many other market participants—heavily asset managers—are looking to buy (which is in future space, for many operational and expense and other considerations).

And it really is the asset manager side of the equation that drives the basis to exist, to begin with; and I think that was significantly underreported, especially as we moved into mandatory clearing. But I think that the broader marketplace has a much stronger understanding of that at this point, and I think the continued work on mandatory clearing—and the best way to roll it out smoothly, and best practices in repo—are just all going to continue to enhance market resilience.

Logan: So, you brought up central clearing as a key piece of this risk management. I think, Nate, your focus was on uniform margining practices. What are the other key risk management features of central clearing that you think are important in driving this initiative forward?

Wuerffel: I think central clearing is one of five key planks that the official sector has underway in the Treasury market. They look at things like data and transparency, trading venue oversight, intermediation capacity, central clearing—and I'm missing one, but it'll come to me eventually. But central clearing is probably the most significant of the five work streams that the official sector has, and that's because it's . . . really, I like to talk about how it will reassemble the way that the Treasury market operates. Inserting a CCP in between two counterparties to a trade reduces the counterparty credit risk between those counterparties; but because you're operating in a Treasury market, which is the world's most important sovereign bond market, it also reduces financial stability risk by removing that counterparty credit risk from the system.

And what it does is, some people have looked at that and worried about the centralization of risk—and it is true that central clearing concentrates risk in the CCP, but that's by design. It's not a flaw; the design of the system is to bring the risk there so you can see it, you can measure it, and then you can manage it effectively. And that's done through things like margining practices, default management practices, having a liquidity facility at the CCP. I think CCPs have generally weathered pretty well over the last decade, and through the pandemic; I think they've weathered pretty well, and so really what it's doing is trying to institute that risk management regime into the Treasury market.

Our view is that that's a net positive. There are some aspects of it that will mean significant changes, not only because the Treasury market is so diverse, you have so many different types of players, they all have to find somebody who will clear their trade for them—and so they have to go sign legal agreements, they have to revamp their operations and technology. But then somebody's got to pay for the margin; that's one of the costs. It's like capital and insurance policy for times of stress—so that, too, has to change.

So, the economics of trading in the Treasury market will be quite different. I think that in times of stress, it will really pay off because what we've seen in the past in repo markets, in very, very high stress periods, you can start to back away from providing funding to your counterparty because you're worried that they may not be there tomorrow. But if you both trade and now you face the CCP, you're much more likely to continue to provide funding and to provide liquidity in the Treasury market in times of stress (of course, in normal times it means you have to pay a little bit more, to pay for that insurance).

Logan: Deirdre or Lou, how do you think that implementation is going? Parts of it have been delayed, and I think the goal is smooth transition; and a smooth transition is really important, so you want to make sure you get it right. How's it going? And if you wanted to raise, what are the two key things that need to get done in order to have that smooth transition? What are the things you'd want to highlight?

Dunn: I think it's going well. I think there's buy-in and support for it, largely, across the board. I do think it is a huge lift, and the original timeline that was laid out was very aggressive, and I think there were a lot of concerns with meeting that. Obviously, we've had a one-year delay, which I think hopefully the industry can meet and achieve.

I think what would give me more confidence about achieving that well would be more clarity on exactly, providers: Who is going to be offering, and what is the model they're going to be offering, and is there going to be a done-away option, and what is the accounting policy for that look like? I think there's a big difference between having one CCP that everybody uses, and multiple different offerings that need to intersect with one another, for example. And we can see that in many other markets where we have similar kinds of activity, and so I think understanding that better and being able to price that appropriately, and for people to be able to make decisions—in some ways, uncertainty is harder than risk, and so we need some of the certainty around those issues, I think, to move it forward.

I would also highlight that, alongside the scale of growth in the Treasury market, we've seen an even more significant growth in just the number of tickets, or the technology needs in order to support that growth in the Treasury market. Because as we've gotten more electronic trading, the number of tickets that you see in any given day, the number as a specific example—every month end where you see a lot of active rebalancing, especially after a month like we saw with the price moves in April, the amount of tickets that you see go through in the three minutes right around four o'clock is just remarkable.

And so, we need to make sure that as we move to a more centralized model, a lot of the service providers across the board can scale up with that. And I think that's going to continue to be increasingly important as the Treasury market continues to grow. So, as we design this in central clearing, we have the opportunity to ensure that we can offer that scalability.

Logan: Lou, anything you'd want to add on implementation, and how you're seeing it in funding markets?

Crandall: Just one minor point. I still think the current implementation schedule is extremely aggressive. If we had started out with this implementation schedule, the industry would have been able to plan a phased transition more easily; getting an extension midway through is less efficient, and still seems like a heavy lift (but I'll obviously defer to the practitioners on this one).

Logan: Do you all feel like the conversations that need to happen are happening, or is there a place you'd like to see public-private sector come together to help meet those timelines and concerns?

Wuerffel: I think there's a lot of progress that is being made. In some of the past large-scale transformations in financial markets, we've seen public-private partnerships and committees. I think that there's good progress in central clearing. I would say that I think there's going to need to be real, laser-like focus. Lou, you point out, there's a lot of work that needs to be done under these extended time frames.

So, I think we'll need leadership on the public sector side, in addition to the private sector side. I just wanted to mention two things that I think are really important. One is, to centrally clear, you need good clearing models that are safe and manage risk well, but also have capital and funding cost efficiency. The biggest cost of centrally clearing is that there's a lot of capital that is required to guarantee your client transaction in the central clearing, and then you have to pay the funding of the margin.

So, there's a lot of work that's being done by FICC and some of the new possible CCPs to lower the cost, the capital and the funding costs, of these transactions. We're supporting a number of those on our triparty repo platform, and I think that those types of solutions will be really important to make sure it's not such an expensive build.

And the other thing, you mentioned done-away; for those of you who don't know the lingo in the Treasury market—and even Treasury market participants don't know it—but a done-with transaction just means you execute a trade with somebody (with a dealer), and then that dealer, in addition to trading with them, they agree to clear your trade. So, the clearing is "done with" the transaction.

On a done-away trade, we agree to trade. But then there's this third-party entity that says, "Hey, I'll clear the client side of the trade." So, they're "done away," they're away from the trade. I think that that could open up competition in clearing in the Treasury market, and could lower the cost of clearing; so, we're very supportive of that. And it exists in swaps and derivatives, but it doesn't exist in the Treasury market.

Dunn: I think we need more accounting clarity in order for that to be.

Logan: So, you both agree that the market does need to move toward having these done-away mechanisms ready, but the question is: What more work needs to be done to set that up and move it at a faster pace?

Dunn: I think there needs to be accounting work done. I think it needs to be a viable business on its own, which right now I think it isn't; so, people wouldn't voluntarily open it up, I don't think. So, as a standalone, that's going to be very, very challenging.

Wuerffel: Yes, I agree. I think in concept, it's compelling. You have firms that are very well equipped to trade, and you have firms that are well equipped to clear; and it's different costs, different talent, different infrastructure. You don't often ask your best race car driver in the world to also be the mechanic, right? And so, there's a reason why you could divide up the labor, but the question is, can you build a model that, on a standalone basis, is actually economically viable?

Logan: We've talked about central clearing being an important part of this increased resilience for the market, particularly given the size of NBFIs. There's also the enhancement of market structure that could evolve with central clearing. I think one that's highly focused on—Dale Duffy talks a lot about this—is the potential for all-to-all trading that could come from this model. What do you see as the prospects for all-to-all, and do you all think it enhances resiliency, or takes it away during periods of stress?

Maybe, Deirdre, I'll start with you.

Dunn: I think we have mostly thus far been talking about mandatory clearing in repo, and the repo market moving to clearing, which is different than the entire Treasury market moving to clearing. But there are many market participants who do clear the Treasury transactions already, broader Treasury transactions, and I think you're asking about all-to-all in the Treasury market specifically.

And while I think having repo clearing could certainly help with that, I don't see it necessarily as a panacea. I feel that there has been ample opportunity, and there will continue to be, for all-to-all clearing to take off even in the current environment; because you can be a clearing member directly as you like and so forth already, and we haven't really seen it take off.

I think there's probably a few different reasons for that, some of which is cost effectiveness, some of which is the scale of the market and your ability to find equal and offsetting risks. Obviously, I work at a primary dealer and intermediary, so my job historically has been to intermediate. But I would say the reason that that job has existed is because the market generally needs it. And we have found that it seems to be in higher demand, especially, in times of significant volatility; and that's where we even see some of our larger transactions move more towards voice trading, or other changes that signify that in times of extreme stress, especially in responding to VaR shocks, where market participants may need to affect certain transactions in order to manage their own, whether it's capital ratios.

Logan: Did you see that in April; did you see more go to voice?

Dunn: Yes; of the larger trades and in the off-the-runs, we see more of it go to voice; and that was true in April, it was very true in COVID. And so, I don't see central clearing . . . it already exists in Treasury, so people could be using it; so, I don't see that really absolving everyone of that intermediation need. But it certainly could pick up; it can't hurt.

Wuerffel: Yes, I agree. I think it'd be a useful innovation. But to your point, Deirdre, if you need another side of a trade, and it doesn't matter if you're going from one asset manager to another asset manager, but you don't want that security at that moment in time, then you still don't have the other side of the transaction. There's a good New York Fed study—or maybe it's a joint study, but—on all-to-all trading and it looks at the time frames in which you have an offsetting purchase or sale. What you want is those things to match up at the same time, and for the most liquid securities—on-the-run securities—that's almost always the case, but as you get further and further off the run, it can be hours or days before you find the offsetting side.

So, at least in the Treasury market, the way it's built today—over-the-counter trading; it's not exchange traded—you need to have dealers as intermediaries that have holding periods, that can hold that security, and then the next day or the next week, be able to sell it to somebody else and manage that risk. So, I don't know; I think all-to-all trading would be great in some form. There's no true all-to-all trading; dealers are still going to be a part of that, if there is such a trading network. And still, you need that holding time that dealers can offer.

Logan: So, we've talked about infrastructure changes in the market, given the growth in NBFIs, so that the market is more resilient. Another key feature, of course, is liquidity provision to nonbanks. Since last fall, the Fed's been experimenting with the standing repo facility—operations, testing early settlement to the standing repo facility. How do you all think the Fed can best provide liquidity that reaches those nonbanks, and what have you learned from those experiments that have informed your own thinking about that facility and that liquidity passing from the primary dealers and into the nonbank part of our financial sector?

Do you want to go this way? We'll start with Nate, and come back.

Wuerffel: I think there are a lot of interesting ideas that have been thrown out around expanding the use of the standing repo facility, in terms of, for example, expanding the number of counterparties you might have to it, or the types of transactions that you might be interested in. For example, some have raised the question: Could the standing repo facility be there to provide an offset to cash-futures basis trading activity?

I think those are really interesting ideas, but it's sort of like tinkering with your prescription when you have like a stick in your eye. I think there are more obvious things that can be done to improve the efficiency of the standing repo facility, and central clearing is at the top of my list because the balance sheet implications of trying to intermediate—you get cash from the Fed, and now you want to turn around to a client that needs the cash, but you don't have the same counterparty on both sides, so that means you can't net it for balance sheet purposes.

So that means it could be very expensive—30, 40, 50 basis points of cost—to move that cash on to your client. So that's a huge intermediation friction to monetary policy, and why would you want to have that?

I think in a future world, especially in a world that is mostly centrally cleared, I think it would be very costly if you have a standing repo facility that's not. So, that would be the first thing that I would do, because once you can do that—and, given the role that intermediaries play in the Treasury market—you should be able to fund lots of different types of transactions, and lots of different types of end users of Treasury securities.

Logan: So, there's been a lot of focus on potential changes in regulation and SLR reform, potentially; if there was SLR reform, do you think that need for netting is still there, or how do you think about the marginal improvement from—or, is it marginal improvement, if there was SLR reform, if the SRF was centrally cleared?

Wuerffel: Yes, I think the two things do slightly different things. I do think that leverage ratio reform—and I would throw into that category also tier one leverage ratio, in addition to the supplementary leverage ratio; they can both be binding constraints on capacity to intermediate. But I think what that does is for the purpose of the leverage ratio, if the leverage ratio itself is your binding constraint, that could help with intermediation in the Treasury market.

But dealers and banks are faced with lots and lots of different constraints, and the leverage ratio isn't always the only one. And so, in that context I think also having the ability to centrally clear transactions could be complementary to this, because it actually allows you to net down on your balance sheet, for accounting purposes, the offsetting transactions. So, I think the two could be more powerful than just one.

Dunn: I agree. I think the SLR is not necessarily the binding constraint for everyone at this point. In certain environments, obviously, it can be. And I also think, in terms of providing intermediation in times of extreme volatility, specifically, it's really more the VaR shock that ends up being the constraint, in terms of how much you can hold, I've found, at least in this past cycle (and also in COVID, and other times of extreme stress)—much more so than it is just the sheer balance sheet implications.

So, the clearing definitely helps you from a capital standpoint, in terms of what the returns look like on the business as well.

Crandall: Obviously, I agree with all of that. I think the financial stability and risk management benefits of a leverage ratio are essentially nil in this respect. I just don't see why we need one, and directionally it can only help. In terms of moving into central clearing, I would make one—all the points about being allowed to net down are critical, and they're the main driver. I would also make the point that I think a very good argument could be made, if you do move the standing repo facility into central clearing, is the one the Fed publishes its transactional counterparty list with a two-year lag—you would just show the FICC as your counterparty. You wouldn't be listing the individual banks that initiated the trade as the institution you were exposed to, from a risk perspective.

I think that's defensible, and for a number of institutions that are very averse to showing up on those lists, it would be a real incentive to participate in the standing repo facility. This is analogous to the way we treat foreign liquidity swaps; we just show the central bank that the Fed faced in the transaction as the counterparty in those regulatory disclosures. We don't have to show the individual banks that they pass the money through to, and this is consistent with lender of last resort practices over the decades.

Originally, people used to talk about a hub and spoke model for discount window lending, that you lent the money to the banks and it was up to them to on-lend to the people who actually needed the cash. The Fed didn't know—didn't need to know—who it was going to, so I would very much argue for simply treating FICC as the disclosure counterparty here.

Logan: I appreciate you raising the question. I think it's still meant to capture the transactions in the spirit of transparency, but I do think it's an important conversation, for sure. Anyone want to comment on the early morning operation—anything else you'd want to add to the value of that?

Crandall: The early morning operations are really critical, because an early morning SRF operation is qualitatively different from an afternoon one. The afternoon SRF operations are essentially bank-by-bank liquidity backstops for individual institutions; it's a surrogate for the discount window, with slightly different terms.

A morning operation, however, is a backstop for the market as a whole, that you are injecting money exactly during the time that you're having severe congestion, during the morning rush hour. And I think that backstopping the market is even more important than backstopping individual institutions, so anything that relieves technical pressure as opposed to simply balancing the books at an individual bank, I think is positive.

Logan: I have two more questions I wanted to touch on, so if others in the audience want to start adding questions into the app, I welcome that. But before jumping to those, Nate, I know this is something that you've talked about: When I think about liquidity, there are three major sources of funding liquidity in the financial sector—more broadly, the Fed, the FHLBs, and then the private sector (particularly, in this context, repo). And I think the system as a whole is more resilient when participants can flexibly move that liquidity between those different segments—and I know this is something I think you talked about in testimony recently on Treasury market resilience.

What do you think the most important opportunities are for better interoperability between these three different segments of funding liquidity? If you were going to bring attention to issues, where would you focus that conversation?

Wuerffel: Yes; I think it's a great topic to be having, because it'll take some work to get to a future state. But ultimately, this is about the—I always think about the Treasury market in these two terms: safety and liquidity. And this really operates on the liquidity dimension of the Treasury market, and that is that Treasury instruments are supposed to be cash-like, so you can convert them to cash easily, they're a high-quality liquid asset.

But of course, your ability to do so depends on the market structure, where you're doing that, what time of day you're doing that; and I think in its perfect state you'd want to say, "I can convert a Treasury security to cash at any time, and in any place, in any location." Because today if you're a bank, let's say, and you have a loan from the FHLB and you have a Treasury security as collateral against that loan, but you're facing liquidity timing mismatches or you're facing severe stress, you may not have cash to return that cash to your lender and get the Treasury security back, and then move that Treasury security, for example, to the discount window. Moving that collateral around is not a simple process today, because those platforms where the collateral sits and can be funded—they don't connect to each other in seamless ways.

And so, I think thinking about how we could move to a world where those platforms interconnect in better ways; whether that's triparty repo, or you're at the standing repo facility, or you're in the FHLB system, or you're at the discount window, you should be able to move that collateral and get financing for that at any time, and any place. We're also faced with the fact that we don't operate in a world of 24/7 Treasury securities settlements and movements; so that's, I think, another dimension of the challenge.

Logan: Deirdre, Lou—anything you'd want to add on interoperability that you think should be the focus of conversation?

Dunn: I think the pipes and plumbing are always way more important than most market participants think, and so I would echo everything Nate said. And I would just say also, as we see the pace of payments picking up, as the market continues to evolve, I think we're saying we need the collateral to be able to move just as quickly, and be able to exchange. And that's probably one of the best ways to also maintain the premier status of the US Treasury collateral.

Logan: Well, I also want to turn to how we measure funding markets, and Lou, I wanted to really ask you: As we think about nonbank financial institutions in this market, what is the best way of measuring money markets and funding conditions, and how do you see that evolving over time?

Crandall: The "evolving over time" part is difficult, but I can tell you how I see it right now. First thing, I'm sure the rest of the panel would agree with this—the last way you want to measure the availability of liquidity in the overnight market is the fed funds market. It's an extraordinarily small, artificial, essentially a regulatory artifact at this point. It works perfectly well as a way of communicating what the current stance of monetary policy is, because it's tracking all the rates that the Fed ought to care about reasonably well. That may not always be the case.

Just two quick things. One is, partly because of the nature of the fed funds market—even more, all overnight money markets tend to have a relationship component to them. Fed funds is dominated by that, which means that it's going to be the last market to reflect changes in reserve conditions. To the extent that a small number of FHLBs are in the habit of placing cash with a small number of foreign banks, it's going to take a lot to move the rate on any given day.

So, the rate will eventually respond to changes in market conditions, but it may very well be the last rate to reflect that. Plus, the reason it is like that is that we have this artificial regulatory structure that makes it rational for the small number of players to transact; but that regulatory structure could change overnight, and if that's the case, the Fed would need to be ready to move to a different featured rate.

So, right now, using the fed funds rate is perfectly viable; you can't count on that indefinitely, but as to the things that will be important at that time, the two main candidates are obviously SOFR and TGCR, the triparty rate. It's between the two of them; I think you need to monitor both of them, because they tell you two different things.

If I'm looking for one rate that monitors funding pressures—well, the state of liquidity—I'm going to pick the triparty rate. Because if you're in a situation where SOFR is rising relative to triparty rates, because dealer funding costs are moving up—if banks are not willing to pay more to cash providers to attract money in the triparty market, that's a sign that this isn't really being driven primarily by reserve availability. You don't solve that particular issue, the particular issue of having general collateral rates for leverage investors go up relative to triparty rates. That's not driven by the Fed's balance sheet, so as far as general liquidity conditions go, I would pick the triparty rate as sort of your base rate.

But of course, the additional funding costs that drive SOFR are also extremely important; and to a very limited extent, I think there's an analogy here to earlier eras where the Fed targeted fed funds, but was very conscious of where LIBOR was. They weren't going to take responsibility for the higher market base rate, but they obviously had to factor it into their assessment of financial conditions. On a much smaller scale, I think the SOFR-TGCR spread has some of the same implications for the way the Fed thinks about the world.

Logan: Nate, anything further—or Dierdre, further from you? Particularly, looking forward to—Lou laid out a way of thinking about fed funds versus triparty or SOFR, but we're talking about more central clearing activity of funding markets that would further change the nature of those repo markets; as you look ahead, any other considerations?

Wuerffel: In the first instance, I largely agree with Lou because I think that when we look at past periods of stress, when either the Fed has intervened or when you see early warning signs of that stress, it's in repo market rates—that $5.5 trillion of activity, compared to fed funds, which is tiny. So really, the repo rates are the thing that most market participants are looking at for signs of stress, so that's important. And it's also where the Fed's existing toolsets operate, is in the repo market; so, that seems like an area that we should be looking at closely, is repo rates.

As central clearing moves in, today SOFR makes up triparty rates and then some centrally cleared activity; as that centrally cleared activity grows and starts to pull in the non-centrally cleared repo trades, SOFR's volumes are going to increase pretty considerably, and then you would have most of the repo market sitting in SOFR. Today, about half of it sits outside of any kind of measured rate, and we can't see really what the rates are on that in the bilateral repo space.

So, I think SOFR will start to reflect the entire market. I think Lou makes a number of good points about the segments of the repo market that should be monitored carefully, too.

Logan: Well, one question that's come in that is, I think, directly relevant to this conversation, is: "As the liquidity day becomes potentially more fragmented—with 24/7 payments, early return, intraday repo, and other developments—do you expect that daily reference rates will continue to be sufficient to capture market pricing? Because you're just really segmenting those features even more."

Wuerffel: I think the early morning SRF operations are a good indication that intraday repo pressures matter. So, that's a great, super interesting question, and I think as you move to more real-time payments and real-time repo, having the ability to track intraday would be important.

I don't know from a reference rate standpoint; I would suspect you don't need reference rate data that's intraday, yet. But in terms of benchmarking and understanding pressures in the market, definitely intraday information is super valuable.

Dunn: Yes, I think we're already really using it and tracking it from a risk management standpoint. I would agree; I don't know that it filters through to changing in reference rates, but I think we're already starting to see it happen just in terms of the bifurcation of liquidity in times of day and things like that.

Logan: So, one of the questions on these money markets that I think comes up a lot is about volatility of money market rates, and how do you think about healthy volatility in money markets that increase resilience, versus unhealthy volatility in markets—and I think this question will probably come up tomorrow, in the policy implementation framework conversation, but I think from a nonbank financial and intermediary perspective, how do you think about this question of volatility in money markets?

Crandall: I think it's healthy; I think it will be healthy to see more volatility as the overhang of surplus reserves melts away in the coming months and quarters. I don't know how much . . . we're always looking at money market rates for potentially seeing early warning signs of changing demand, whereas in many cases these are just technical market developments. But having a reason for banks to be actively involved in managing this on a day-to-day basis is probably a net positive.

Dunn: There's always a balance, I think, between price discovery and what the value of price movement is, and volume that is clearing; so, it's a trade-off between the two, or that you have to think about it holistically between the two, overall, I think. But yes, I would agree with everything Lou said, as well.

Wuerffel: Yes. I think repo rates today reflect supply and demand factors, and they move around; and there's plenty of room for them to move today, and so I think you'd still want that price signal. But to the degree that repo rates started—for example, if they were moving way outside of the Fed's monetary policy target range, that might be unusual because that would probably be a sign of pretty significant stress. A 25-basis point move in repo rates is a pretty big move on an overnight interest rate.

Dunn: We also have certain days that matter more than other days, whether it's quarter ends or tax payments or things like this. So, I think that you need some elasticity to be able to reflect those things going through; to the extent that it's sort of unexplained or not tied to anything else, I think that probably also becomes much more of a concern.

Logan: Yes; I think, Lou, you introduced this. There are different types of volatility in these money markets. Some of it comes from collateral supply, some of it comes from the level of reserves or ampleness of the system, some of it comes from these just different known calendar effects of demand and supply.

And then there's the unknown category, and the unforeseen category, that is more challenging, I think. So, we're talking about the healthy or unhealthy amount of volatility we have to keep: the driver matters, I think is what I'm hearing from you all.

One of the questions that came in relates to some of the topics we were talking about earlier on central clearing: "Given the difficulty of using repo markets to liquefy posted margin when Treasury markets are under stress—[video interrupted]—liquidity risks associated with providing done-away execution of Treasury clearing?"

So, we talked a little bit about the importance of done-away; how do you think about that connection with liquidity?

Wuerffel: One dimension of this question, I think, may relate to the fact that sometimes people will point to CCPs as being procyclical in their margining practices—which, if you think about volatility going up, it makes sense that you would in some cases need to increase the amount of margin that you collect, and so I think margining systems are generally procyclical. But I guess I would point out that an even more procyclical form of margin is to not have any in place, and then introduce margin in times of stress—because then it's like an infinite increase in margining, so that's the most procyclical form.

So having a margin regime in place in the first place is going to be better off than not having one, from a procyclicality standpoint. When it comes to once you get into the cleared world, whether it's done-with or done-away, it's basically a question of who's going to pay that margin—whether it's the executing dealer that is also clearing the trade, or whether it's a third-party agent that would clear the transaction.

And so, the margin payments would come from basically the entity that clears it. I think that's just a matter of economics, right? You can price the value of that margin, and that could be passed along to a client in price terms in repo, or it could be a fee if it was a done-away agent.

Crandall: I would add one more thing to that. Done-away is challenging, but one advantage of it is that done-away clearing will only be done if it is priced economically. Right now, dealers who are bundling that as part of the overall relationship may be pressured to underprice that. We're going to know exactly what it costs with done-away; and that may not be a decisive factor, but I think that's an important step forward.

Logan: There's another question thinking about these issues internationally: "The ECB and Bank of England financial stability reports have both flagged the rising role, and the risks, of nonbank activity in funding banks on a day-to-day basis. Is this a trend also evidenced in the US funding markets that's on your mind, and does this add to potential systemic risk?"

Crandall: Banks are in the business of borrowing money and lending it on, so certainly certain categories of transactions can grow to levels that are problematic. I don't really see that right now.

Logan: Another question, coming back to the Treasury facilities that we have at the Fed; one of the questions is: "Why not have the Fed offer repo financing against Treasuries continuously, under 13(3), which places no limits on counterparties and doesn't require unusual and exigent circumstances." I think the question is, right now there's two operations (if I'm understanding the question correctly): the standing repo facility, two operations a day. Why not just have a window open all day, and with no counterparty limits? How do you all think about that?

Wuerffel: I think it'd be super cool. It's a little bit like the question about real-time payments; if you had 24/7 ability to convert Treasuries to cash, and you had a Fed facility that was there that could do that at any time—there's some operational things that you'd have to consider, like the maximum amount of the funding available, how you would price such intraday things, but in concept I think it's a really interesting idea. It's the same with the discount window; if you could have a perpetually open discount window, that would be an ideal thing because then you don't face timing problems.

Dunn: It would also change, I think, the "first among equals" of US Treasuries, in terms of—not that it has equals, but as collateral and cash substitution; for sure.

Crandall: That's a really critical point. There is no downside to the taxpayer, to the financial markets, to anyone, of improving the moneyness of Treasuries. It makes financial markets more resilient; it makes Treasuries a lower cost financing vehicle for the government.

And, to the extent that you can use the Fed's standing repo facility to enhance that, I think it's a net plus; and I would expand it to a broader range of financial market participants—as long as you have robust regulation about things like margining, and all of the potential leverage concerns of having people be more confident in the moneyness of Treasuries (and all of those regulatory protections are important in their own right). There's really no downside to making Treasuries a more attractive instrument.

Logan: Because there are three components of that conversation; one is, how comfortable are you all with the aggregate size of the facility during periods of stress? Two is, should it be available equal amounts across the counterparties? And three is, should it be available any time of day?

On the any time of day, it reminds me—Lou, you probably remember this—many, many years ago, the securities lending facility was like a window, well before it came into the operations (and, I think, very similar in that structure). But it sounds like you would argue for just making it widely available to those counterparties.

Crandall: I remember producing frictions.

Logan: I want to go to a question on transparency, because we haven't touched on transparency, which I think is another key part of this question about resilience, particularly for the nonbank financial institutions, which we have less data about. The question is: "The OFR just began collecting data on bilateral uncleared repo"—I think we cited this, and how we learned about the zero margins on that. "Is there a chance it could be scaled back? What are the costs to market resilience of losing that data collection; how valuable and important is it? And should a different entity, either public or private, pick up that activity?"

Wuerffel: I think repo data, if you're sitting at the US Treasury Department or the Fed or anywhere in the official sector, and you want to understand the United States Treasury market, then you need to understand what's happening in the repo market. And so, I think the collections that have begun to understand the trading activity in repo are really important. And that's important for the production of reference rates (SOFR is based on repo data), but it's also important to understand where there might be vulnerabilities in the bilateral repo space; so, that seems like an important set of information to have for policymaking.

Logan: Anything else on transparency? If you could have the next initiative on transparency, either in repo or cash, what would you want to be pushing?

Dunn: I would say from—and TBAC has done a fair amount of work on this topic as well, and there tends to be a fairly disparate view of opinions, in terms of the value of transparency and the way that people use them. And I think thus far, the official sector has generally proceeded with some degree of caution—or, more degrees of caution in the Treasury market than in some of the other markets, where we introduce real-time trace data much more quickly and things like that; and I think that has been a benefit to the Treasury market overall.

So, from where I sit, I'm not sure; I think the aggregate information has been helpful (the aggregate information around just number of transactions), and again the system's ability to support that, and see how those things trade. I think it's very relevant. I think it may be worth further study, and more intermarket discourse on some of the volumes that occur around month ends, or rebalancings, or some of the strain that is being put—we've started to see some signs of stress on some of the broader marketplace sort of intersections, almost, in the past month or so.

And so, I think that there would be value and more focus on that and transparency around those issues, which is sort of different than price reporting; but I think it's something that gets hotly debated often.

Logan: If we go back to where we started the conversation, on the April volatility—questions about who was selling, or have holdings changed—it takes quite a long time to get information about that; and even once you get it, it's very hard to directly understand it. One of the ideas has been, should holdings data also be a priority for the official sector? How do you all think about that?

Dunn: It could be a blessing and a curse, depending on what it says—especially if you think about in the time like early April, right? So, I think there's a difference between the official sector having the data, and the information being released broadly to the marketplace. I certainly think the official sector having the data makes a lot of sense. I think that releasing it more transparently, I think you'd have to be very thoughtful around timing, I think, in periods of heightened volatility (and it was especially true in April).

And I commented on this. I didn't feel that financial markets necessarily reflected the aggregate, probability-weighted distribution of outcomes of smart people thinking about what the world was going to look like as much as usual. It reflected the last trade that happened—much more what people perceived to be the last trade that happened. And I think with too much transparency, you can really exacerbate that risk, where people are just following the lead in what's already gone through the market; and that compounds a lot of the intermediation and access to liquidity problems, and in a way, that could be pretty dangerous.

Wuerffel: Yes, that's a good point. I don't know if you would agree with this, Deirdre, but—I think most of the debate in the marketplace is around the public disclosure, the publishing of data as opposed to whether, for example, the official sector should have information on the holdings or the activity in the US Treasury market, which seems a lot more defensible. But to your point, there can be downsides to excessive transparency in public disclosures, and so calibrating that is super important.

I think the Treasury and the official sector have done a good job of just taking that in baby steps, to make sure they're not breaking anything as they move incrementally down that path.

Logan: Well, I want to thank you all for the conversation today. We covered a lot of ground on the growth of NBFIs, what we saw in lessons learned from April, what structural features in the market may need to change, and the role of Fed liquidity and interoperability of Fed liquidity with other parts of the financial sector. Anything else you wanted to make sure to cover on this topic of Treasury market resilience that we didn't cover, that you'd like to say before we conclude the panel?

Crandall: If I have one item on my wish list—and I know that this is one that is difficult to implement—but if I could expand access to the standing repo facility to one group of players, it would be central counterparties. The idea that central counterparties, that may have to manage large amounts of Treasury collateral in the event of a default, you can't go to the Fed and turn that into cash instantly, strikes me as very counterproductive. I understand that there may be legal issues with doing that, but when you talk about trying to minimize the cost of central clearing, one way to do it is to reduce the amount of liquidity support the banks have to commit to the financial market utilities, and I think that's just a no-brainer.

Logan: Anything, Nate or Deirdre, that you want to end with?

Wuerffel: We covered a lot of ground; I agree with Lou on that one. But I just go back to doing things that make the Treasury market safe and liquid, because I think that's the enduring hallmarks of the market; and so, anything we can do to make it more so, the better it's going to be.

Dunn: I would just say, I think the Treasury market, specifically, has a long-standing and enduring private-public partnership—whether through TMPG, TBAC work that we all do, FSOC that we all do together, which I think reflects just the central importance and everyone's appreciation of the underlying importance of the Treasury market, and especially as the scale of the Treasury market continues to increase, I think that's something I think we all look forward to leaning into.

Logan: Well, thank you all very much; and thank you to those of us who joined online, and everyone here in person, for day one of the Financial Markets Conference. We're going to take a break now for this afternoon, both online and in person; and for those who are here in person, please be back at 6:00 p.m. for the reception here, and there'll be an event starting at 7:00 p.m. by our evening keynote speaker, Michael Schwarz, corporate vice president and chief economist at Microsoft. Dinner will be served promptly afterward.

For those online, we hope you'll tune in via YouTube at 7:00 p.m. to hear the presentation, and we look forward to having everyone back at that time. And I just hope all of you would join me in thanking Nate and Lou and Deirdre for all of their insights about the importance of Treasury market resilience.