2026 Financial Markets Conference – Policy Session 1 Transcript - May 18, 2026

May 18, 2026

Policy Session 1: How Should the Federal Reserve Deal with Future Financial Stresses?

In the Federal Reserve Act of 1913, Congress mandated that the central bank mitigate banking panics by "furnishing an elastic currency." The act established the Fed as the lender of last resort, paving the way for unprecedented market interventions during the 2007–08 financial crisis and the 2020 pandemic. To address strains in a significantly evolved financial system, the Fed created new emergency lending and liquidity facilities, and though they proved effective in supporting market functioning, these interventions have been the topic of intense debate. This session explored how the financial system has changed and how the Fed can best meet its responsibilities when new stresses emerge.

Transcript

Cheryl Venable: All right. Good morning, everybody. It's nice to see everyone here this morning. I'm going to go ahead and get us kicked off for our Financial Markets Conference for 2026. I'm Cheryl Venable. I'm the interim president of the Federal Reserve Bank of Atlanta, and my other day job is the first vice president and chief operating officer of the Federal Reserve Bank of Atlanta, so I've had the distinct pleasure to serve in both roles while we're searching for our new president.

It's great to see everybody. It's beautiful here at Amelia Island, and I also want to make sure I'm welcoming those who are joining us online today. The entire conference is being live-streamed, ad-free, and I've been told at no cost, so people can enjoy it as they will.

Boasting is not our style at the Atlanta Fed, and it's really not becoming of a central bank—I think we've got some central bankers in the room, or ex-central bankers in the room—but I'm going to brag for just one minute about the Financial Markets Conference. This is our signature annual event, and we are celebrating our 30th edition. So, I'm just really pleased that I have the opportunity to kick off our 30 years. And I see some applause. I think we should give—all of us—a round of applause for that.

For three decades, we've assembled some of the best thinkers in economics and finance in a relaxed setting that allows us to delve into the timeliest issues confronting the nation's/world's financial markets. Events like this advance important conversations and ultimately equip both the Fed and other policymakers with the sturdiest intellectual materials for building policy.

In fact, the power to convene groups like this is one of the Federal Reserve's less obvious but yet significant features. When we invite people to come together and share ideas, they generally show up—and so I'm pleased that we have a full room today.

They do so because we offer a nonpartisan, independent venue to interrogate issues, research findings, and hold conversations about critical questions of the day. As I noted, we have been at this for decades now, and while it's worth recognizing the longevity of the Financial Markets Conference, I also want to point out that we are not here to reminisce—this is about not looking in the rearview mirror, but really firmly focused on the future, and so our topics are appropriately forward-looking—yet discussions will also be grounded in an appreciation for the lessons of history.

Recent years have produced no shortage of economic and financial shocks—from COVID, to an inflation spike, to a trade war, and now to some shooting wars—so our very first session could hardly be more appropriate. A distinguished panel with deep expertise grappling with crisis at the Fed will explore how the financial system has evolved, and how the central bank has responded and confronted financial stresses, both then and now—and in the future, as they are certain to come.

We will also dig into the role of stablecoins in the financial system, both through a research lens and through a discussion grounded more in the practical role stablecoins might play in the payment system. Another group of experts we've brought together will examine the experiences of central banks in adopting and regulating digital innovations and the financial system.

And of course, no conference these days is complete without a session on artificial intelligence, so we'll explore research on the real economic value of generative AI, and take a comprehensive look at the varied, complex ways AI might transform financial markets, ranging from its effects on volatility and risk management to regulatory considerations.

Then we have lined up also an all-star roster of keynote speakers. First up today, we are going to be pleased to welcome Gita Gopinath, the Gregory and Ania Coffey Professor of Economics at Harvard University. Given Gita's experience as the chief economist of the IMF during one of the most transformative periods in recent economic history, I can't think of a better person to discuss the breadth and depth of change in the global economy as it absorbed a series of shocks.

Our second keynote features Roberto Perli, who is at the center of the Fed's monetary policy machinery. He is the manager of the System Open Market Account at the New York Fed, and he is in charge of all the levers that transmit monetary policy to the real economy, and to financial markets. Appropriately, he will be discussing the critical topic about how the Fed implements policy, and the myriad of challenges associated with managing our balance sheet.

And then finally, Anna Paulson, the president and CEO of the Federal Reserve Bank of Philadelphia, will deliver the closing keynote tomorrow evening. Anna is a 25-year veteran, but she's the newest president in the system (if you don't count me as interim president; I'm probably the newest in the system). She assumed the helm in Philly last July, and she's going to provide her take on the current policy environment, and then she will sit for what promises to be a lively conversation with Colby Smith, who covers the Fed for the New York Times.

So, I gave you a broad overview of what you can expect over today and tomorrow, but I would just also emphasize: this conference is a great way of building network and connections, and having those conversations. And I know we leave plenty of time to extend what you're hearing today and tomorrow into the afternoon, and into the evening. So, I invite all of you to meet somebody new and make a new connection, and further those conversations.

So with that, I hope everything goes well for you. I'm excited about the agenda that we have for today and tomorrow. I expect I'm going to learn quite a bit. And I am now going to go ahead and get it started with the first panel. Mike McKee of Bloomberg Television and Radio is going to be moderating our first panel discussion, and I am so pleased to have them there.

And before he introduces his esteemed panel, I just want to do one call out and "thank you" to two important groups that make this conference actually come together: that is the staff here at Amelia Island Resort and Spa—but in my opinion, most importantly, our esteemed team here at the Federal Reserve who has brought this together. And if you wouldn't mind just allowing me to give them one round of applause for bringing this all together, because without them we wouldn't be here. So, thank you all.

All right. With that, Mike, I'm going to turn it over to you.

Michael McKee: All right. Thank you, Cheryl. We are very lucky to be the first panel this morning, because you all had your coffee so you're awake (for a while). I don't have to do a long introduction of any of these people, because you know them; they've been in central banking for years and years and years—I think we're calling this panel "getting the band back together," after not seeing each other for a few days. But joining us this morning: Charlie Evans (he was the president of the Chicago Federal Reserve Bank); Randy Kroszner, who was a member of the Board of Governors (and as the old saying goes, you can take the boy out of central banking, but you can't take central banking out of the boy. He just was serving on the Financial Policy Committee of the Bank of England, which we'll hear more about in just a moment); and Jeremy Stein, a former governor of the Federal Reserve.

So, gentlemen, thank you all for joining us. Now, what we're going to do this morning is each of them is going to make a short opening statement. I will ask some incredibly intelligent, penetrating questions, but you can all send questions in through the app. Please send to the Q&A, and I will look at them, sort them, and try to combine (if we get a lot of them together), but we'll get a lot of audience participation in this.

And obviously, we have a lot to talk about this morning, because what we have seen, over and over again, is constant financial crises in a way that we hadn't before. The formal recessions that we used to have were economic imbalances, and the Fed had to solve those basically by lowering interest rates.

But now we've had a series of crises where they've had to take additional steps, and it has become a major talking point: How do they do this? How do they help the economy through these things that are now broader than just economic imbalances, but they are financially based—and that takes away some of the weapons that they have. And so, each of them is going to give us some ideas on that, and then we will turn to the Q&A.

We're going to start with Randy, on the end.

Randall Kroszner: Thank you. I'm delighted to be here. I think the last time I spoke here was actually when I was at the Fed, but the sunshine is still as bright as ever; but unfortunately, the clouds are in front of us, because I spoke here before the Global Financial Crisis, and—exactly as Mike had said—we seem to be getting one global shock after another.

Jay Powell has this nice phrase that he kind of summarized, that we may be entering a period of more frequent and potentially more persistent shocks—a difficult challenge for the economy, and in particular, for central banks. And I think that's the context in which we're having this conference; the shocks can come from the supply side shocks, they can be geopolitical shocks, and obviously there'll be potential for technology shocks, which may be positive or may be negative (or maybe a little bit of both).

And so, what I wanted to do is try to make the case for why monetary policymakers, as monetary policymakers, should be very interested in financial stability, and why that should be integrated into any monetary policy framework. I don't think it makes sense to have monetary policy over here, financial stability policy over here, and when something goes wrong maybe you kind of talk to each other. I think it has to be integrated in from the basics.

And exactly as Mike said, I'm very fortunate to serve as an external member of the Financial Policy Committee at the Bank of England. They have a very different structure than the Fed. So, the Fed, the Board of Governors, and the FOMC control both monetary policy and financial stability policy; and of course, all the extraordinary actions that the Fed takes in response to shocks, those are powers of the Board.

The Bank of England, in response to the Global Financial Crisis, structured itself to have two parallel committees that are both chaired by the governor of the Bank of England, Andrew Bailey. One focuses on monetary policy, interest rates, and inflation, and the other focuses on financial stability and financial regulation. And there's a lot of overlap of the governors and the deputy governors on both of those committees, and each of them have a number of outsiders.

And of course, the Bank of England welcomes outsiders; the previous governor of the Bank of England is now the prime minister of Canada, and they also welcome others onto both the Financial Policy Committee and Monetary Policy Committee. And I think this is very, very helpful—and let me give you a couple of examples to illustrate that, of the benefits of having that kind of structure.

So, the so-called "liquidity-driven investment crisis," when interest rates moved up dramatically in the UK a number of years ago, moved up very rapidly. There's a hedging program that insurance companies had been using that involved a lot of risk—well, it involved very little risk in normal times, and that's how things always are in financial stability. When things are going well, everybody's making money and it's brilliant; when things move a little rapidly, then there are a few problems.

And so, there was a major disruption in the gilts market, the UK Treasury securities market—the British have much better words than we have in the US. In the US, it's the Treasury securities market; there, it's the gilts market—because, of course, these gilt-edge securities were of the highest quality. It's just like, who wants to be secretary of treasury? Don't you want to be chancellor of the exchequer? You know, it just sounds so much grander. I learn new words every time I go to these meetings—and also words I shouldn't be using, because of different interpretations.

And so, I think that in that crisis, was exactly the time that the Bank of England was trying to reduce the size of its balance sheet, but it had to go in and buy securities. So, what's going on? Are they doing quantitative tightening, or quantitative easing? Well, it made it very easy; Andrew Bailey could say the Financial Policy Committee has agreed that we need to intervene in the gilts market to provide stability in that market. That's not a monetary policy action; that's a financial stability action.

And even more extreme—which I think was incredibly valuable—he said this is a temporary targeted action, and then he said it's going to end on Friday. Which was really quite shocking, and I think it would be very difficult for the Fed in its current incarnation to be able to do that. I think it would be very valuable to very, very clearly separate out financial stability interventions from monetary policy interventions, and I think one of the challenges that the Fed had was with COVID disruptions in the Treasury securities market—and I think it did exactly the right thing by intervening with incredible force to buy lots and lots of government securities. But it then became very difficult to then slow down its asset purchases, because it didn't really clearly articulate that this is for financial stability reasons, we'll do something else for monetary policy reasons, and it didn't want to be perceived as pulling back on monetary policy.

So, I think the Fed is not going to be able to adopt the kind of structure the Bank of England has, but it could have some elements of it; and one is I think a clear articulation of financial stability actions versus monetary policy actions. Now of course, when the crisis strikes, it's a little bit of both; but as long as you had those clearly articulated, it then makes it easier to say, "Okay, we've solved this particular problem; we haven't solved the longer run macro problem, so we still may need to purchase securities (quantitative easing) for that, but we can actually now reduce our asset purchase speed because we have dealt with this particular problem."

So, I think that—the communication strategy—is super, super important, because I think the Fed got stuck in buying more assets than it otherwise would have because of that. Second, I think that you really have to think about the role of liquidity provision, in both normal times and crises; and I'm of the view that yes, you can always send up facilities ex-post, and so maybe you want to keep that off to the side until the crisis hits. I think it's much better to have things clearly articulated in advance, clearly regularized in advance, have different facilities that are ready there so the markets know what's coming, and how that support can be provided—and then hopefully by knowing that, it makes it much less likely that you have to use those in large quantities. And so, I think that is extremely important.

And then I think, as part of the scenario analyses and robustness that's done in bank regulation, these kinds of scenarios should be worked through, understanding how shocks—including changes in interest rates, so the interaction between monetary policy and financial stability policy—affect the banking system. Certainly, I think that was something that we should have done more of in the run up to the rapid rise of interest rates, and some of the issues surrounding Silicon Valley Bank.

So, I think the US is not going to be able to adopt a UK-style structure, but I think clear communication strategy differentiates the two pieces, a policy framework that does stress testing and scenario analysis that really brings all of this to bear in both monetary policymaking and financial policymaking, and then thinking about the liquidity provision that is crucial to monetary policy transmission, both in normal and other times. And I think that should be part of the basic framework, and I think unfortunately the Fed missed an opportunity to make this much more explicit when it did its framework review. But I think we're soon to have a new chair who seems to be interested in doing some revisiting of some frameworks, so maybe we can get it in there. Thank you.

McKee: Okay; thank you, Randy. Jeremy?

Jeremy Stein: Well, thanks—and thanks so much for having me here. I think I'm going to really just follow very much on the same track as Randy. I'm going to focus my remarks essentially entirely on a specific question, which is: Suppose we get a replay of March 2020, on the eve of the pandemic, when we had all this turmoil in the Treasury market; how should the Fed respond? So, it's all about that.

And just to refresh your memories, this is the beginning of the pandemic; this is the famous dash for cash. There was a tremendous amount of selling of Treasury securities; roughly, think of this selling as being, in a very, very short period, on the order of about $1 trillion, roughly evenly divided among bond mutual funds that basically were experiencing redemptions, foreign central banks, and hedge funds—and the hedge funds are particularly interesting. They were in this so-called basis trade, where they were long Treasuries and short futures, started getting big margin calls because of the volatility, and had to liquidate their positions.

So that was a big thing then; just to give you a little sense that this could be important, you can tell this pretty clearly from CFTC data: the hedge fund basis trade position on the eve of the pandemic, about 500 billion; now, it's well in excess of a trillion (it's closing in on 1.5 trillion). So that at least potential fragility is out there, which I think motivates asking the question: What would we do if we had to do a do over?

What did the Fed do last time? Well, they admirably did lots of stuff; they did some repo lending, they excluded Treasuries and reserves from the calculation of the supplementary leverage ratio (the idea being that that would somehow give the dealers more balance sheet space to absorb the selling); but in the end, they've just brought overwhelming force, bond buying force, to bear.

I think the basic motivation is the dealers had been themselves overwhelmed by all the selling. Their balance sheets were, in some sense, overstressed, so they couldn't do their other jobs of making markets and repo lending. They were just overwhelmed, and the Fed is basically coming in and saying, "Well, we're going to take some of the stuff off your hands."

And the scale of this was just enormous: they bought $162 billion of bonds in a 10-day period. So just to give you a sense of scale, we were doing, in QE3, $85 billion a month; we thought that was a big deal. Now, it's basically two months' worth in 10 days. By the end of May, it was $1.6 trillion.

And so, huge effort; and in the narrow sense of helping to restore function to the Treasury market, it was ultimately successful. It was successful, but arguably, not without cost; and here I'm just going to echo what Randy said: I think it's fair to say it created some confusion, and essentially bled into or morphed into monetary policy, and you can actually see this in the statements.

The first FOMC statement in March, they said, "We're doing this to support smooth functioning of markets." By September of that year, market function was basically fixed—but they're still buying at an enormous clip, and they're saying to sustain smooth market function and help foster accommodative financial conditions, thereby supporting the flow of credit (in other words, half monetary policy—but they're still doing it at a clip that was huge, relative to monetary policy stuff).

And then they had a very hard time backing off, because they were in monetary policy mode and the thought was, "If we start backing off this current clip of purchases, it's going to send some kind of hawkish signal about interest rates." So, at the end of the day, the Fed ends up buying something like $4 trillion of bonds, much of it after the initial market function thing had passed. And that's a lot. It was not clear that it was necessary, for support of the real economy.

The way it ends up, it's a huge intrusion, in some sense, on the traditional role of the Treasury in setting debt maturity; and in fact, ex-post, it ended up being very costly to the taxpayer, because after we had some inflation and rates rose, the Fed ends up losing a huge amount of money on this position. It's a real cost to the taxpayer because it'll show up in reduced remittances to the Treasury. It's essentially as if the Treasury had all of a sudden decided to finance itself on an overnight basis, by issuing all these short-term reserves.

So, it's a problem. It's absolutely the problem that Randy said, and in fact you can imagine—much by like analogy to the UK—it would be even more of a problem if this recurs in an "above target inflation" environment. What are you doing buying bonds, when inflation is above target?

So, how can we do better next time? Again, Randy sort of teed the problem up exactly: we need to somehow have a better way of separating purchases that are done for market function from purchases that are done for monetary policy. And I agree, I think the Bank of England did this much better—and more successfully—during their LDI episode. They were able to be very explicit that it was for market function, and they pledged to end it in a short period, and so they did much better.

I think it's in part because, as Randy alluded to, they had this separate FPC structure, which made it clear that this was coming from the FPC so it was not a monetary policy thing, and part of it was just good policy—and in part, I think I would say they also got a little bit lucky. This commitment to end things on a Friday; that's awesome, as long as the stuff calms down by Friday. If somehow, for whatever reason, there had been market turmoil that had extended beyond Friday, then they'd be a little bit like "Oops, we're going to go till next Friday after that."

So, with that kind of as preamble, what should the Fed do? I have some work with Anil Kashyap, Jonathan Wallen, and Joshua Younger, and we have kind of a different proposal, which is: If the Fed finds itself having to intervene again for market function purposes, they don't have to just outright buy cash Treasuries and take interest rate risk or duration out of the market; they can buy Treasuries, and at the same time hedge these purchases, either by taking a short position in futures or in swaps. And that way there's obviously no confusion with monetary policy, because the whole monetary policy QE is all about taking interest rate risk out of the market and trying to, in principle, affect long-term rates.

So, why would this possibly make sense, and why might it work? So, if you take away nothing else from these remarks, here's one fact that really kind of changed my thinking about a lot of things: We think about the dealers as having been overwhelmed by a flood of selling, and the Fed is essentially trying to take some of that burden off their hands; but exactly on what dimension are they overwhelmed?

Here's the fact: It's not interest rate risk; it's not interest rate risk. So here, there's very kind of cool work by Jonathan Wallen and Lina Lu where they had access to this Volcker rule disclosure data, where you can see—it's really amazing data—you can see on a daily basis, for each of the five large dealer banks, at the individual trading desk level, their exposures. And what you see is they basically don't take interest rate risk.

So, while it is true that during the pandemic they were being sold a ton of Treasury bonds (if you look at their cash holdings of Treasury securities, they're ballooning), but if you look at what the trading desks are doing, they're hedging all of that risk. So, they'll get shelled with cash Treasury securities; they're turning around and hedging that.

So, the problem that these dealers have is not that they're bearing too much interest rate risk; that's not the relief they need. The risk that they're bearing is essentially a form of basis risk—it's that they're long Treasuries and short derivatives, and that's not a perfect hedge, so that's the relief they need. The Fed can provide that relief by just doing the opposite—taking that package, essentially of long cash Treasuries and short derivatives, take that package off of their hands.

Another way to say the same thing is, what are the hedge funds liquidating? They, too, are in a basis trade position; they're long cash Treasuries and short futures. That's what they're selling, and so you just need to take the other side—which, again, would be for the Fed to buy the Treasuries as they did in the past, but to hedge this. If you do that, it's very clear that you're not doing monetary policy, because monetary policy should have, essentially, to a first approximation, no effect on the level of rates, but it should essentially ease dealer balance sheet constraints and thereby allow them to do their thing.

So, that's the proposal; monetary policy involves taking raw interest rate risk, market function does not. So that's the idea. So, to anticipate one of the obvious objections, that if you do this it creates a form of moral hazard—that, well, the hedge funds are going to think, "When I get in trouble, you're going to come to the rescue"—agree, agree. And it's an absolutely fair critique, and so I want to be very careful: I'm not saying this should be the first line of defense.

I'm not saying that; I'm saying something very different, which is: Suppose you're the Fed and you find yourself in the same position as last time, and you can't help yourself; for whatever reason, you're about to ride in there and just buy Treasuries. Relative to that, I think this has less moral hazard. If you just ride in and outright buy Treasuries every time there's problems in the market, that creates moral hazard, too—and I would say it creates more moral hazard, because you're essentially creating a put on the level of rates. Here you're creating a put on the difference between the cash market rate and the derivatives rate.

So, again, this is relatively more surgical; I'm not saying it's problem-free. Another critique is, "You're having the Fed do this thing and they're doing derivatives, and isn't this kind of crazy and intrusive?" If you think about it, economically, it's just repo; it's just repo. You're just buying and committing to resell in the forward market. Economically, it's the most plain, vanilla thing the Fed does—which is repo. The only difference is, unlike standard repo, your counterparties to the two different legs of the trade are different counterparties, because you're buying from one and you're essentially selling in the forward market.

And then finally, to the BOE's commitment to buy, and this is market function so I'm going to sell by Friday; this is like that, in that the sale, the committed sale, you've done on the day by having this forward position. But as a result of that, you don't have to announce when the program will end. You can keep this in place, because you're committed neutral from the get-go, whereas in Randy's thing the way you're neutral in some sense is you're promising to do a reversal of the trade. Here, it's just sort of built in. But it's very much in the same spirit, and I would say it just fits the Fed better, because we just don't have quite the same structure.

So that's the pitch, and I'll stop there.

McKee: All right. Charlie?

Charles Evans: Thanks, Mike. All right. Thanks for inviting me. I'm going to take a slightly longer view, a little bit less on liquidity (which I think matches better for the panel). So, what are the prospects for a smaller Fed balance sheet? I became Chicago Fed president in 2007, and the System Open Market Account (SOMA) was $800 billion; today, it is $6.3 trillion. How did it get to be so large?

In theory, the Fed could have decided right off the bat to shift from the previous scarce reserve system to an ample reserves floor system, and this would be the explanation; but that was not the direct line of causation, it was a feature. Monetary policy was at the zero effective lower bound in 2008; the Fed responded to the Global Financial Crisis with QE1, QE2, monetary expansion program, and open-ended QE3.

Later, with the Fed funds rate at the ELB again, the response to the pandemic generated more QE. So, the SOMA expansion since 2007 was clearly about monetary policy under extraordinary duress.

Separately, consider where today's SOMA size landed after the first quantitative tightening from 2017 to 2019, and the second QT from 2022 to 2025. The FOMC had discussed alternative reserves regimes at the November 2018 FOMC meeting; they ultimately landed on ample reserves, and were aiming for that. One must acknowledge that the US economy is larger, and financial market trading operations are vastly larger. Any longing for the good old days of $800 billion is just completely unrealistic.

So, what are the prospects for a smaller Fed balance sheet? If the effective lower bound is a thing of the past, monetary policy QE would not be necessary, period. The funds rate is the primary tool of monetary policy. Maybe the real interest rate (r*) is higher today; most policymakers have said as much, and perhaps it is much higher now. After all, productivity growth is strong and expected to continue to be that way—and the low funds rate environment, owing to secular stagnation, is behind us, too (also the global savings glut, also the dearth of high-quality assets).

The range for the corresponding long horizon equilibrium funds rate likely includes today's current 3.64 percent rate, as Jay Powell has informed us; and if real GDP growth tests the 3-6 percent range owing to AI investments and deregulation, as some administration experts have suggested, perhaps the equilibrium funds rate is still higher. So, I will repeat: If the Fed, financial markets, and the US economy can avoid returning to the effective lower bound, the SOMA balance sheet will not need to be increased as a tool of monetary policy—so I expect the Fed would continue to refine its ample reserve system for SOMA to be as small as feasible, for efficient and effective bank reserves functioning.

So, what might that look like? And I have a chart at this point; here's the liability side of the Fed balance sheet, and how bank reserves held at the Fed had evolved during these episodes.

In the chart, you can see the Global Financial Crisis and pandemic effects that I mentioned. The chart clearly shows the key Powell strategic policy approach for SOMA and QE—that is, quantitative tightening. Powell and the current FOMC approach has been to round trip the balance sheet to its ample level. Repeat, round trip back to the most efficient and effective size for market functioning. Where are we today? It's growing again, after QT, to be right-sized with the effective market functioning level.

So, how does this fit with incoming Fed Chair Warsh's plans? Can someone go lower? Is a new inflation framework part of the plan? Well, Kevin Warsh had a January 15, 2025, Wall Street Journal op-ed piece where there's a policy proposal providing one outline for potential regime change strategies. In a nutshell, return reserves to their 2019 pre-pandemic level in order to reduce monetarist inflation pressures and reattain Fed credibility. Since this may play a material role in the next Fed framework, let me read the key paragraph, and it's on the chart:

"The Fed must recognize that money matters to the conduct of monetary policy. Its bloated balance sheet has contributed significantly to inflation. By draining as much as $2.5 trillion in excess reserves, the Fed would mitigate inflation and return its size and scope to pre-pandemic levels, hardly a period of restraint or austerity. Call it a practical monetarism to restore Fed credibility."

In offering the title "practical monetarism," Warsh may be signaling his new inflation framework with a smaller balance sheet approach; time will tell. There are sound suggestions for reducing the banking system's substantial demand for reserves. Governor Miran, President Logan, Sam Schulhofer-Wohl, and Darrell Duffie are some contributors here. The proposals range from reducing stigma from borrowing at the Fed's facilities, to re-engineering the financial payment system to net out more crossing payment flows (that's my casual, less expert language).

Personally, I worry that each of these are ambitious, substantial, Manhattan Project-like initiatives. I have my doubts, but I will welcome being educated further on their execution details, and likely effectiveness. In the weeks and months ahead, I expect we'll learn much more about incoming Fed Chair Warsh's new regime strategy for monetary policy. Thank you.

McKee: All right. Thank you, Charlie. I'm going to start with the first question that came in, because it essentially matched what my first question was going to be—and Jeremy touched on it, the idea of moral hazard. With all of these plans and various different thoughts on it, don't we already have a moral hazard out there, in that it contributes to all this because we've got the Fed put?

If anybody wants to take that. You're looking at me, Jeremy.

Stein: Yes. Whether you want to call it moral hazard or Fed put, I think we've absolutely had a ratcheting up of essentially expectations for Fed intervention, and you can see this in lots of ways. For example, in the pandemic—and this is not just the Fed, it's the Treasury as well—in the pandemic we had these bond buying facilities, the corporate bond buying facilities, and I think there was an underlying problem, which is bond funds that hold relatively liquid bonds and offer daily redemption are a financial stability issue. And I think if the Fed hadn't come in with these facilities, we would have seen—and the Fed in some sense ruined the empirical experiment—we were going to see just how dangerous these things could be.

But I can't argue with, that they had to do it; they had to do it. It was really a moment of crisis. But then what happens is that fragility is never really shown for what it is, and that sector just kind of continues to go on and grow. So I don't know if you call that a moral hazard, but by essentially rescuing various parts of the financial system, we don't stress test them in the way that we otherwise would, and the urgency for any kind of reform is taken away.

So, in that sense, the expansion of the Fed's footprint has brought with it—sort of has danced along with—an increase I think in risks in various parts of the market, and at the same time then you can see why it's hard to unwind. "I'm going to teach somebody a lesson" is easy to say, but hard to do when you're sitting in the chair and the world is coming unglued.

McKee: Yes. Randy?

Kroszner: Yes, this is exactly the issue of thinking about stress testing not as just traditional, "there's some sort of macro shock that could lead the unemployment rate to go to 20 percent and the stock market to fall by 40 percent"—and that's kind of the traditional way we've done stress testing in the US. It should, I think, be focused exactly on this issue, and in particular focused on the non-bank sector—because (and you can think of it as both part of an intended and part of an unintended consequence) a lot of activities moved outside of the banking sector.

But that doesn't mean that the risks disappear; it may just mean that they're a bit more opaque. And so, that's actually one of the things that the Bank of England has really been focusing on are looking more broadly at the non-bank financial institutions, the NBFIs, and how stresses there can come back to the banking system, where some of the underlying interconnection fragilities might be.

And so, I think doing that proactively—trying to think through that as part of a monetary policy exercise, rather than saying, "there are some things that the bank regulators do, and we as monetary policymakers don't want to get our hands dirty with that"—they're intimately related. You can't separate out, particularly in, as Jeremy has discussed (and I had mentioned) in both the UK and the US, when there's a shock, the Fed is not just going to stand by. And so, you need to think about the monetary transmission mechanism; you want to make sure that the monetary policy transmission mechanism is being effective, but you also want to make sure it's not being undermined by a lack of financial stability.

And so for me, that gives the rationale of why you would want then, after doing these stress tests, to figure out what structures you want to have in advance; because really, it's not about moral hazard, it's about effectiveness of monetary policy transmission during a crisis. As Jeremy said, especially after what's happened with the Global Financial Crisis and COVID, the markets know that the Fed is not just going to stand by and let the financial markets melt down, nor should it. But I think it's better to have it regularized in advance rather than sort of a willy-nilly thing that comes in afterwards.

And also, what's been interesting is that after Dodd-Frank, people were very concerned that a lot of the Fed's independence to use the emergency powers—those wings had been clipped, because you had to now ask the Treasury Department for an okay to do a lot of things. Well, in the response to COVID, it wasn't that there was less of a response; there was an even greater response, and a lot of that was driven by congressional action that said now the Fed should be giving money to municipalities, to below investment-grade corporates—things that when I was at the Fed we weren't willing to do, but the Congress said do it. So, I think the moral hazard comes in not only on the Fed side, but also because the Congress can get involved, and they directed the Fed to provide credit in a lot of other areas that it had been reluctant to (at least in the Global Financial Crisis).

McKee: If we were to put in place some kind of structures in advance, how would that happen and what would they look like? I'm sure that the Open Market Committee has at least talked about that, and staff has talked about that.

Evans: Well, the FOMC has put in place a number of—they've got the standing repo facility now (I guess it's the standing repo program now, they've renamed it to make it sound less special, I guess). Dealing with this concern about stigma, moral hazard, and whether or not people actually come and use it, is a real problem.

You can go back and look at how Jay Powell and the FOMC pulled down the special programs that Ben Bernanke had put in place during the Global Financial Crisis, and say, "Well, those are on the shelf;" they're standing, they're ready to be used, they innovated more because of the special programs that were—and the Congress supported it, with Main Street Lending and the corporate credits, and things like that.

Some of the most successful Fed programs are the ones where you kind of put it in place, and it just supports market functioning enough so that markets take off, right? The TALF was talked about a lot, but actually not a lot of take up was there. The Commercial Paper Funding Facility was enormously important, in the grand scheme of things going back.

But on the previous comments, I don't have a lot to add—except that you don't need a large balance sheet to still have the Fed take the blame for a Greenspan put, or something like that. You go back to the early ‘90s, after the savings and loan crisis, and the Fed kept the funds rate at 3 percent—which was viewed as a floor at the time—for quite a long time, and that helped banks rebuild their capital during that time period.

And one way or another, that's an important part of the engine for growth to get out of where you are and keep going. So, you need good market functioning; you're going to favor some type of investment somewhere along the line, but you hope it's not too much—and you hope that, in some sense it's irregular enough so that investors can't bet on it, and take them out of that too early.

McKee: Do you agree, Jeremy, that you don't need a big balance sheet?

Stein: You don't need a big balance sheet. One thing I feel like is the discussion over the size of the Fed balance sheet has been—I feel this communication problem on the part of the Fed—has been overly focused on the nominal dollar size of the balance sheet. So, on the one hand, I absolutely agree, and this was in our previous discussion, that you can criticize the Fed for having too big a footprint in financial markets, and for (in some sense) overstepping their role relative to the Treasury in affecting the consolidated debt maturity of the government. It's the Treasury's job to decide the maturity profile of government debt, and when the Fed takes all the long-term debt out and replaces it with short-term, it's in effect intruding on that.

So, I think there are a bunch of critiques that are legitimate critiques, especially when we're outside of a crisis or away from the zero lower bound, that the Fed should have a smaller footprint—but I wouldn't think of that footprint in terms of just the dollar size. I think a much better measure is, in some sense, how much interest rate risk have they taken out of the system; that is to say, what is the dollar duration of the Fed's portfolio?

So, I think it's perfectly reasonable to think that should be a target of policy, and they should aim to reduce that. And the reason I think it's an important distinction is because the balance sheet has to balance, and the problem of thinking of shrinking the footprint in dollar terms is if you have fewer assets, you're going to have fewer reserves—and then you get into the problem of taking too many reserves out of the system.

So, I would say if they just change the maturity profile of what they hold—imagine a limit case, where they're holding mostly bills, or something like bills. Then even if there's a $6 trillion, $7 trillion, $5 trillion dollar balance sheet, they effectively have almost no economic footprint. They're holding six-month bills, they're turning that into overnight reserves, essentially no interest rate risk, no impact on long-term rates—like what I was saying with the derivatives. In other words, the right metric is: How much are you really impacting market rates?

And then given that, if they can do that, I'm not at all fussed about a large quantity of reserves—or even better, of RRP—in the system. If you think about it that way, what the Fed would be doing with its balance sheet is sort of a last mile function. In other words, maybe really what the Treasury should be doing is issuing a ton of overnight bills, because the market really likes one-day paper. But that's a lot to auction, so they issue 6-month bills, and the Fed turns that into overnight RRP. I think that's better than having a—talk about moral hazard—I think that's better than having a financial system.

Who does that now? Hedge funds. Hedge funds have an enormous carry trade position, where they buy longer-term bonds and repo finance those, thereby giving the financial system the very, very short-term paper it wants. If the Fed took over a little more of that role and left a little bit less for hedge funds, it would probably be a good thing for financial stability.

And again, you could do all that; you could keep a pretty big dollar balance sheet while having effectively no real role in finance. I think this is very much in the spirit of old Milton Friedman and the Friedman rule, that if the Fed can produce money essentially at no social cost, it should do that. And money used to be paper currency, maybe now it's essentially overnight interest-bearing money.

So, sympathetic to the larger dollar balance sheet, also very sympathetic to getting the duration down.

McKee: Composition; yes. Randy, you mentioned the fact that the Bank of England talks much more with the Exchequer than we do in the United States, but Kevin Warsh has talked about a much closer relationship between the Fed and the Treasury. What would be the optimal way of doing that without making the Fed appear that it is subservient to the Treasury?

Kroszner: So, I think actually building on what Jeremy was saying, to the extent that Kevin has talked about this, it seems to be on this issue, on the maturity structure should be a Treasury function; it shouldn't be a Fed function. And so, to have very clear delineation of what's in the Fed's lane and what's in the Treasury's lane. So, a coordination on that, I think along the lines of what Jeremy was saying, I don't see that as in any way impinging upon or reducing Fed independence or Fed credibility. It's just a different function.

But again, it has to be clearly articulated; if you just say, "There are a lot of meetings between the Fed and the Treasury, and we're not telling you what they're about," that's not very helpful. But if it's very clear this is about maturity of the debt, we are going to have a very clear protocol that the Treasury is responsible for this, which I think most—along the lines of what Jeremy was saying—I think most economists would be very comfortable with that. I don't think people would say that's impinging upon the Fed's ability to do monetary policy.

And also, on what Jeremy was saying, when Charlie was at the Fed (and when I got to the Fed even a little bit earlier than that) that $800 billion, that was basically all short-term Treasury securities, precisely for this reason that it was—obviously, there needed to be some liquidity management of the system, but to try to have this small of a footprint on the functioning of the financial system, that was up to the financial system, and the Treasury securities market and maturity was up to the Treasury.

So, I think you can do it in that way. I think the Bank of England and the Exchequer have been able to get a good balance without people saying, "The Bank of England is not independent," because they're talking with them about these issues, that there are the discussions about financial regulatory policy and such. So, I think you can manage it, but you just have to be super clear; much like I was saying, you have to be clear about financial stability objectives and monetary policy objectives. You just have to be super clear about what these objectives are. If you're not clear, then it raises all these other kinds of questions.

McKee: Well, Charlie, if you have—well, the Open Market Committee has said, basically, they want to shorten the maturity of their holdings, and Kevin Warsh supports the idea of moving much closer to bills; but then what we've seen in crises is that you've got to buy everything, because that's what's freezing up—especially the longer maturity stuff. So how do you avoid having a balance sheet that is weighted heavily towards longer maturities? And then, I guess the question everybody always asks is: How do you get out of this once you have it? Ideally you want to do what Jeremy said and have almost all bills, but can you do that?

Evans: I believe that the FOMC's greatest discomfort with the long duration of the balance sheet is on the MBS. And of course, that's not just the Treasury managing them, but that's something completely different. And the Fed got into that because they were at the effective lower bound, and the first time around inflation was too low and you were trying to get the economy going. If you find yourself in a situation where you don't hit the effective lower bound, I'm sitting here listening to, "Well, we need to be concerned about markets, and fire sales, and things like that."

And absolutely, I agree with that; but the Fed, prior to the Global Financial Crisis—I mean, the balance sheet was smaller but it was all interest rates. And if you could address the recession and low inflation with just lowering the funds rate but not going to zero, you didn't have to access all of this. You didn't have to buy the long-duration assets that were necessary, but if you find yourself in that situation, then you're trying to do more.

If the funds rate ends up being at 2 percent, I don't think you're going to be buying those kinds of assets, if you're in a recession. I think you're going to hope that you can lower the funds rate and deal with it, and Congress will come along with government spending and things like that.

There are these two distinct functions, and financial markets are bigger now, so I think that speaks greatly to having some type of standing facility and some kind of discipline, like Jeremy is talking about, with futures when you're intervening. But something would preserve you from growing the balance sheet once again into a larger footprint, I guess.

McKee: Jeremy, did you have a comment on that?

Stein: Yes, I think that it's what these guys have been saying. I think the discipline you want to be very clear about is in normal times monetary policy, you have a low-duration balance sheet. There are two potential exceptions to this. One is if we hit the zero lower bound again; but away from the zero lower bound, this is what we're going to do, and you really telegraph that. If you hit the zero lower bound, then I think you may have to start taking on duration.

The other is this market function issue; but then again, my proposal would be: there you intervene, but you still don't increase your duration footprint, because you're hedging. So really, that says if you don't look at the nominal size and you look at duration, and you're very clear in your communications up front—"this is the metric I want you to hold me to"—you don't stray from that, other than at the zero lower bound. Even crises, I think you can basically manage not to take too much interest rate risk.

But a lot of this is a communication thing; I think it was a failure of the Fed, when they first started doing this, to not communicate more forcefully about the difference between just the size of the balance sheet. So now, unfortunately, I think the size of the balance sheet just in dollars has become a bit of an optical political football, and so Kevin has got himself a bit in the position of having to shrink that just to show some progress. I think it'll be a test of his communication, whether he can move the conversation away from that and towards this more economic concept.

Evans: You invoked "Friedman rule" kind of reasoning, and as you grow the balance sheet—now I mean, I agree; I wish we had more experience, and understood better the size of the balance sheet as we're going along and when to curtail it. But, large balance sheet—what are the costs that are associated with that for the economy and financial markets?

One of them is a political economy concern, which is you can point to remittances that nobody's ever paid attention to, until all of a sudden they're negative and you kind of go, "Oh, that's a big deal." Now, that was in the last three years that we saw that, but Ben Bernanke dealt with that question back when the balance sheet grew with all those complaints, and his argument was, "Well, over 25 years we are going to return on average $25 billion a year more, even if we end up having negative remittances for a few of those years."

Now, can you explain to the American people and politicians that a 25-year plan like that is actually good, or are you always going to get castigated in those two or three years when it's easy to point to? Now, Bernanke never ran a negative remittance; you look at those, it's during the pandemic. And you could have avoided the negative remittances by not raising rates—but you had to raise rates, of course, right?

Stein: But you never get into that situation, again, if you take the interest rate risk out. The negative remittances are because—

Evans: At the very beginning, and don't grow into it; that's right.

Stein: That's right, because they had an enormous exposure coming into the—

Evans: Let's remember, when the ECB and Europeans are doing QE, and somebody asked them about, "How are you doing QT?" They go, "Well, we're looking at the Fed, because they've learned how to do this;" right? Bank of England, everybody was learning; and I think that during the pandemic more lessons were learned, the market functioning versus helping the economy. So, we know more; we know to avoid your proposals, things like that. I think those are important. I'd rather not do them, but yes.

Kroszner: Yes, exactly; but the Fed is going to do that, and central banks are set up to do that. They are an insurer of last resort, and so it's very difficult for them to just say... Mike Mussa, one of my mentors, always used to give this great example of, on that moonless night in 1912, when the Titanic hit the iceberg and went down, do we just say, "Well, we've got to teach them a lesson so we can get better ships going forward," or in the time of crisis do you send the other ship to pick up the people who are in the icy waters who are in those lifeboats, and then deal with it afterwards?

My view is Mike's view; you can't just let everyone just go down in the icy waters. You would prefer them not to have gotten in that situation to begin with, but you can't make the world as you want.

But on this point that we've been talking about—it's a bigger picture thing about, how can you get something that's an automatic runoff of a new program that you do? So, in some sense, as Jeremy is having, it's automatic; when you're getting in, you're getting out at the same time. When we were doing the programs that—because I was there from 2006–2009, during the crisis, we were pioneering these programs, we thought very much about: How can we get out without having to explicitly get out?

So, like those lending programs, what we did is, we didn't set any quantitative limits; we just said, "We know that the commercial paper market is not functioning, we know that both financial services firms as well as non-financial services firms are not able to get this funding, which is crucial for what they do, and so we're going to lend against good collateral." We only took the highest rated commercial paper at the time, the A1P1 paper (but that had knock-on consequences for the rest of the market), and with no quantity limits.

And when you look at the two programs together, they went up to a total of a half-trillion dollars; and the Fed's balance sheet, as Charlie had said, was only $800 billion before that. So, that was just an enormous increase in the size of our activity, and doing something we had never done before.

But the way we structured it to get out of it is that we set the cost of the market, the so-called haircuts and other fees, to not be at an enormously high level, because we wanted people to use it. But we set it at a substantial level above what it had normally been, because what we were hoping is that as we intervened and did these various programs, it would calm the markets and then they would naturally decide, "Well, the Fed's not a very good counterparty right now, they're very expensive; let's go back to others."

So, we wanted it to run off naturally, and that's how all these things did run off. And I think it's super important, as much as possible, whether it's the particulars of what Jeremy is proposing or in other programs—there's also the "it's Friday," which is a very risky thing to do. Andrew took an enormous amount of risk. His whole credibility was on the line for it. Fortunately, it worked for him, but that was very, very risky. So, if you just want to, as a natural part of the program when you're doing it, how do you get out of it—and hopefully without having to take an explicit action, that it sort of self-liquidates.

McKee: Let me switch gears a little bit from the balance sheet to another tool that people have talked about for years, and that's ending the stigma on the discount window. As long as I've been covering the Fed, people have been trying to get rid of the stigma of the discount window. Jeremy, can that be done in any way? There is a lot of talk about prepositioning at the discount window, but is it ever really going to go away?

Stein: That's a good question. It's obviously a challenging thing. Some of this maybe you can get at with pricing; in other words, there's the famous Badgett principle that you lend freely at a—I forget what it is—at a penalty rate. So of course, if you do that, then the only people who would ever want to borrow at a penalty rate must kind of be, not great. Maybe the way to lubricate this thing is to have it be somewhat on more favorable terms. There's a real challenge; that's a challenge here. We like the penalty rate for exactly the reasons that Randy said—that it's self-liquidating.

On the other hand, that creates an endogenous stigma, so whether with some of these standing facilities you sometimes want to have them used in normal times to take away the stigma, I'm not sure. I think that's a tricky question.

Another thing you could do—this is sort of a backdoor trick, I think, which is—there are some conversations now around the liquidity coverage ratio, so right now the only way you can satisfy the liquidity coverage ratio is by holding either Treasury securities or reserves. There's been some conversation about how you should be able to get some credit, just by virtue of the fact that you have assets, that you would be able to borrow from the Fed. I'm not sure I'm crazy about the idea that you get liquidity credit for a bunch of loans that you have plans to borrow from the Fed at some future time, but if you wanted to give people 20 percent credit for that in exchange for saying, "I'm going to preposition this with the Fed," take it away.

Part of the problem is the Federal Home Loan Banks. People basically use them as the lender of, I don't know, second-to-last resort—and it's cheap with them, so there's no stigma, and then that interferes with the Fed. Short of just taking the Federal Home Loan Banks out and kind of shooting them, one way you could do this is by saying we're going to give you regulatory credit for prepositioning your collateral with the Fed, and basically giving you credit for the idea that you will come to the discount window. And then in stress situations, essentially amping up that credit so the discount window becomes more attractive.

Maybe you can do a little with that, but that is a long-standing and kind of tricky problem.

McKee: Well, Randy you were there at the beginning...

Kroszner: It wasn't all my fault.

McKee: The decision was made to put mortgages into the QE package. When you look back now, was that a mistake? Would you do that again?

Kroszner: So I think at the time, the idea was—going back to this monetary policy transmission mechanism idea—that the concern was that if we just bought government securities, that's great because it's bringing down the taxpayers' borrowing costs; but obviously, the ground zero of the crisis was the mortgage market, and so we wanted to make sure that the monetary policy was being transmitted very rapidly to that market. And so, I think it did make sense, in that context, to do that. The question then is, how quickly do you get out of it? Do you continue to do that? How much of that do you continue to do?

One of the good things with mortgage-backed securities is they naturally self-liquidate, as people repay their mortgages, as they move, or—in some cases—default. The mortgage-backed securities naturally shrink on their own, and so that's a way that they can self-liquidate. With a 10-year treasury, you've got to wait 10 years before it liquidates.

So I think, given that crisis and given that we were super keen to make sure that monetary policy was being transmitted to the heart of the crisis, because the traditional banking system was not intermediating, I think it was the right decision. But I do think that there would have been a way to get out a bit more quickly, subsequently.

McKee: Well, Charlie, if we get another crisis, would you buy mortgages as well? Or is that proven that it doesn't need to be done now, if it's not a housing-based crisis?

Evans: It depends on the situation that you're facing; it depends on your attitude towards the dual mandate. What are you facing? Are you facing high unemployment? With high unemployment, you're going to want to work to reduce that. With low inflation, I don't think people care so much about missing on the downside. I fought against that for a long period of time.

We've recently seen inflation that's very high, and you look back on this and in retrospect, people really don't mind lower inflation, everything else equal. I think that nominal wages reflect something different, and people do have money illusion, in my opinion; and so, I think that's a little disadvantage for folks.

But if you find yourself in a situation where you need to do something, and you're at the zero lower bound, the Fed is incredibly restricted in what they can buy; and when you look at Treasuries you just say, "That only gets us so far." And MBS is federally insured, it's within the allowable asset class to buy, it is the closest proximity to the private assets you'd rather have some ability to influence those borrowing costs, and so that's why you do that.

Taking duration out away from the market was an important feature there, but that locks you into a particular maturity structure that's difficult to unwind unless you're willing to sell at some point, and that's anathema for central banks. So, it's a very difficult one; I wouldn't be surprised in the future at committees saying they'll do a little bit, but not as much as they did before.

Stein: Yes. I have to say, I can't see the big fuss about MBS, given that you're already buying long-term Treasuries. As Charlie said, they're also insured, so you're not taking credit risk. In either case you're taking interest rate risk. You're arguably doing that for a very specific reason, which is you're at the zero lower bound; and they're going to be more effective in some cases because it's going to directly move the mortgage rate, and the market is sometimes just thinner so you get more bang for your buck. I don't see the problem. Maybe some people say, "But now you're dabbling in credit allocation, and you're favoring housing;" man, you lost that innocence a long time ago.

When you move interest rates, there are interest rate sensitive and less sensitive sectors, and housing has always been kind of a key vector of transmission because it's interest rate sensitive. So, there's no monetary policy that is neutral across sectors of the economy.

So, again, I totally respect the idea that you don't want too much interest rate risk; but if you're doing it, the idea that you're doing it through mortgages versus Treasuries, I don't see that that's something to be hugely fussed about. At that point, you're just trying to help, and as Randy said, you're trying to help in the most effective way.

Evans: Another feature in the current environment is, we are now in a global environment. There's a lot of geopolitical risk, there's a lot of volatility, we've had these supply shocks, there's the interconnectedness. And that was just a huge game changer, in my opinion, in terms of the relative price changes through the pandemic that found their way into inflation, that were sticky, that stayed there—and also the labor shortages and whatnot.

In the earlier day, during the Great Financial Crisis, part of this idea was that it's going to take a long time to get inflation up, because if it's not a supply shock, the Phillips curve is really flat and just generating those inflationary pressures, it was difficult to appreciate that they could rise very quickly (unless the Phillips curve got really steep, and it needed the supply shock with the added "oomph" of labor shortages in the pandemic, and other factors as well). We're probably more likely in that environment now going forward than we were before, so it's another reason to be more concerned about all these risks.

McKee: Well, we've got all these programs on the shelf, as Charlie said, that were brought back out again during COVID. They're still there. But what also happened during COVID was there were additional programs put into place. I'm wondering now—this is sort of a culture and practice and possibility question, since there are constraints on what the Fed theoretically can buy—but you did do ETFs, so are there more programs as you look forward?

I'll ask this question of everybody; we can start with Charlie and go down the row. Is there something else that may need to be done that you could see possibly being done?

Evans: I really don't know. I think you have to talk to the lawyers. It was breathtaking that as many programs that were rolled out, and somehow the Treasury providing funding for first-loss provision was critical in all of that. Unless you have that—and that's the off-the-shelf nature. You can do it if you've got the support of the administration and all of that, and if you don't, you can't do that under the current law (unless the lawyers are more clever than—

McKee: They're always lawyers.

Stein: Well, I think this is where your question about the Fed put kind of comes back to bite, which is—think about what happened during the early days of the pandemic, when there were these bond buying facilities. At some point, Jay Powell gave some remarks in an interview or in a press conference where he said we've crossed a lot of red lines, and this is the time when you do that you figure out the niceties afterwards. Triple C rated bonds rallied massively on those words.

Now what was interesting is, they didn't have the authority even then to buy triple C. These were facilities, essentially, to buy investment-grade bonds with a little bit of an exception for fallen angels, but the market's reaction was like, "the Fed has our back." And if you looked at the pricing after that thing, it was remarkable. I mean, credit spreads were narrower than during normal times, and it was still like before we had a vaccine or anything.

So again, I think the ratchet on the Fed put has gotten very high, but what was extraordinary—and Randy alluded to this—what was extraordinary about the pandemic was the extent of bipartisan support. None of the bond buying facilities are possible without congressional and Treasury support. Those things were capitalized via the Treasury and via the fiscal support, and that was a very unusual thing because it was essentially an alien invasion. It was like the one time where we're like, "Okay, there's this bad thing that's happening, and it's nobody's fault," and so it's very, very likely the next time we have a crisis it's not going to be of that nature and we're not going to have the bipartisan stuff.

And at that point, the worry would be that the Fed is swimming a little naked, which is, whatever they would like to do, they just won't have the legal authority to do it. They can't buy credit-risky bonds on their own without a fiscal backstop, but I'm not sure how much the market appreciates that. And so, you worry that there's a little bit of an attitude towards risk that we won't be able to make good on next time there's a problem.

Kroszner: The one thing that it seems that Republicans and Democrats agree on is to spend. And so, whether it was during the pandemic, or when President Trump was first in office, Republicans and Democrats don't get along; they could never possibly do anything. The one thing that they agreed to do is spend, in both the Trump administration and the Biden administration, for a two-year period, 25 percent of GDP.

So, I hear what you're saying, but if it's just spending—and this is what's very interesting. As I said, the concern post Dodd-Frank was that the Fed's wings had been clipped, so they would be hemmed in and not be able to respond quickly to a crisis. Well, they responded in many new ways, because Congress discovered leverage. They said, "Well, we can give $100 billion to the Fed and they can then—we'll be in the first loss position, but then they can lever up and do a trillion-dollar program on top of that."

And that, I think, is a little bit dangerous because it got the Fed into a lot of areas that, when I was at the Fed, we weren't willing to go. So municipal securities, for example; those are not especially liquid, and many of them are not especially high quality. I come from Chicago and from Illinois, where we've had some challenges over time. Things have gotten better; the credit ratings have gone up, but still maybe not as high as a property owner would like, in Illinois and in Chicago. But that becomes very political, and then you also get the log rolling that, well, I need my muni to get this and you need your industry to get that, and then they agree to just say, "Okay, we'll do a program that will be broader."

And so, I worry that that gets us much more into the political allocation of capital, and so I think it's really an unintended consequence of trying to restrict what the Fed can do that's actually made the credit allocation and the crisis potentially much more political. And I think that we did what we did in 2008, 2009, then in COVID much more, and I think the next time it's going to be, "Wow, there are all these great policies and we can lever it up, so let's do something specific for agriculture, and let's do something specific for airlines, and this and that." I worry that that's the next step.

McKee: Obviously, we've seen the Treasury market expand exponentially over the years, and the financial system has expanded exponentially over the years, into all kinds of shadow market activities. Is there something that the Fed could do to reduce the fragility of the Treasury market at this point? You mentioned stress testing. What would you stress test, and how? But, what else could the Fed do in advance to try to bring down some of the potential problems out there?

Kroszner: This is precisely what the Bank of England is doing in the so-called systemwide scenario analyses, that they are specifically saying, "Let's look at the non-bank financial sector, let's look at private credit markets, let's look at the government securities markets and try to understand better those fragile interconnections." So, I think that's a first step, because we know a lot about banks, and they disclose an awful lot; and for the major banks you have Federal Reserve supervisors who are sitting in those institutions, but they're not in these other places—and so, there's a lot of uncertainty.

Also, one of the challenges with that is that when a crisis strikes and you're not sure, that might lead the central bank to do more than it needs to do. And so, I think having a much clearer view through the stress testing is the necessary first step, because I think we don't really understand all the channels. We know that there's some potential ones out there; we don't know empirically which ones are important, and also there may be other channels we haven't thought about.

McKee: Charlie, do you want to say something?

Evans: I'm not quite sure what you have in mind when you say what the Fed can do to ensure... the Treasury has got to fund the debt. Congress and the president make the choices of the spending and the taxes, and then there's deficits that have to be financed. You've got debt, and that's what they have to do. Markets like those assets, and they create derivatives on the basis of that, and they put on trades with leverage and all of that. The Fed doesn't have regulatory oversight for those markets and exchanges.

As a former central banker, I'd want to step back. It is the case that the Fed needs to be mindful of financial stability issues and provide as much support as makes sense within our dual mandate responsibilities for the good of the country, but making things, facilitating those trades so that more can be done—and I mean, this is just... you want to make sure that the debt gets financed in the way it's supposed to. And I don't really have much more beyond that.

McKee: Okay. Well, let me ask this—and I'll go to Jeremy on it first: What should the Fed's role be, going forward, in terms of the dollar and swap lines? The administration's talking about extending them. We don't know whether they mean through the exchange fund or through the Fed, but does that play an important enough role that the Fed wants to keep doing it, and would they want to expand it?

Stein: So let me actually just back up, because I think it relates. On your previous thing, I think there's some things they can do, and have already done. They have, for example, these standing repo facilities, where they stand ready to provide liquidity against Treasuries, both to banks and (quite importantly, actually) to—and this is why it's going to come back—to foreign official accounts. So, they have this, I think it's called the FIMA facility, where they—and if you think about sort of the counterfactual, one reason I think that foreign central banks were selling their Treasuries is they wanted basically cash.

But did they need the cash at that moment, or was it precautionary? If they knew that they could always monetize it at the Fed whenever they needed to, maybe they don't have to sell immediately, so that's helpful. I think the swap line is a very similar kind of idea. In other words, really, in either case, as the Fed, you're just lending (in the one case, the repo facility; you're lending to a foreign official account, and they're posting Treasury collateral).

What is a swap line? You're lending to a foreign central bank and they're posting own currency collateral. But it's basically providing liquidity, in an essentially riskless thing, on a collateralized basis to another foreign central bank.

I think that's part of just making the plumbing work effectively, and in some sense it's the flip side. People talk about the exorbitant privilege that everything is in dollars. Well, then there comes a little responsibility to go with the exorbitant privilege, which is these guys have banking systems that in many cases are very, very dollarized, and so when they run into a problem, when the banks in Turkey or whatever all of a sudden can't come up with enough dollar funding to fund the dollar loans that they've made, in many cases in syndicates that involve US borrowers, you're just trying, it's part of keeping the monetary transmission. You don't want borrowing costs to spike, and you don't want to lend directly to a bank in Turkey, so you lend—actually, I shouldn't use them because they're maybe not eligible for this one, but—you lend to a central bank that will then turn around and—

Kroszner: South Korea. They have a big currency mismatch, so South Korea is a good example.

Stein: Yes; okay. But you turn around and lend to a central bank that can then turn around and lend to one of the others, so I think it's part of just the responsibility of keeping funding markets working smoothly. It ultimately redounds, in part, to the benefit of US borrowers and dollars, because these are integrated markets and some of the US borrowing is coming from banks in Europe and other places, so I basically think that if they're well collateralized and they're good credits, it makes sense. Obviously, they can become politicized, and you want to do everything you can to avoid that. But I think it's much like repo lending, it's a core function of a central bank.

Kroszner: It's interesting because I think it gets back to your earlier point about Fed-Treasury accord, because clearly, making central banks and other institutions around the world comfortable with holding large holdings of US Treasury securities is something that the Secretary of Treasury is very focused on. We're also going to be hearing about stablecoins next, and that's another area that it's focused on.

So, I think the Treasury sees this as part of that broad demand for the dollar, but also in particular something that could be helpful in reducing the borrowing costs for the Treasury. And obviously, providing these insurance mechanisms is something that will make people be more comfortable to hold more in general, and then be less likely to fire sale in challenging times.

McKee: Well, you do a nice transition to the next question. The administration is very interested in—and obviously this group is, because the next panel's on stablecoins—advancing the crypto industry; what should the Fed's role be in helping out non-banking institutions who deal in crypto in the next dash for cash?

Stein: Well, that's a big question.

Evans: He's the finance expert.

Stein: Here's one thing: so right now, there are two leading stablecoins; there's Circle and Tether. Circle is the GENIUS Act-compliant one; it's basically based in the US, and it has submitted itself to some of that—and there's Tether. And on the one hand, there has been a thread of the discussion about making stablecoin essentially more crisis resilient, because right now they don't have—neither of these has—access to the lender of last resort and they hold—Circle does hold—mostly Treasury securities. Tether does not; they've got some meaningful fraction in like gold and Bitcoin.

So, you could imagine trying to kind of bulletproof them more, or you could try to bulletproof Circle more, by saying, "Look, you're kind of under the umbrella already. Why don't you submit yourself even more so to bank-like regulation, and then we'll give you access to the lender of last resort?" I think that's a reasonable kind of impulse.

What I worry about is, that's not the use case. The real use case for a lot of this stuff is like evasion things. Most of the large, large majority of people who hold stablecoin hold them not through exchanges, they hold them in their self-custody wallets. At least anecdotally, a lot of this—or not a lot, but a substantial portion—is for illicit activity. And so that stuff doesn't want to be brought under the umbrella.

So, it's a case of, you can build it, but it's not clear. You can build a piece of this that is in some sense closer to the regulatory umbrella, therefore the Fed gets comfortable with giving it access to lender of last resort, and you make that piece of it safer.

But I think the dominant use case piece doesn't want to be... right now, the market value of Tether is multiples of that of Circle. So that's in some sense telling you it wants to be in the less regulated and less overseen thing, and it kind of wants to be—for reasons I can't fully understand—partly in Bitcoin and partly in gold, so I worry about that stuff but I don't know that you really have... you can maybe safen up the stuff that is already on the safer side, but I don't know that it's ever going to be the dominant bit of the market.

McKee: Randy, have you ever thought—I want to throw out, I just thought, when Jeremy was talking—that Morgan Stanley and Goldman Sachs didn't want to be banks either, until they needed to be.

Kroszner: Yes, for sure. The long arm of the Fed is quite long. Well, I want to give the example of the UK. So, the original proposal the UK put out about two or three years ago for stablecoins was addressing this issue, and so they said, "Okay, we're going to give the stablecoin issuers access to Bank of England accounts." And of course, from a central banker's point of view, and traditional finance point of view, that's the golden ticket.

Of course, in this realm, not everyone trusts central banks very much, and not everyone wants a central bank monitoring all of the assets that they have. But in turn forgetting that, they said all of the banking assets would have to be held at the Bank of England.

So, the positive about that is that, tick by tick, the Bank of England could see what the holdings were, and if there were any issues that might untether the stablecoin from the pound, they could immediately step in. And so, you would effectively eliminate the run risk.

The challenge was that the proposal said that there would be no remuneration on these accounts. And so, of course there's no take up; it's a beautiful structure that no one wants to use, so the bank is revisiting that and thinking of having a smaller fraction. But that's one potential way to try to do it; if you can get the balance right, if you have... some of the assets will be held at the central bank, so the central bank will be able to monitor that because you're concerned about financial stability issues, and so you have insight into that.

The challenge is, it leads to a meltdown from the banks, because the banks say now you've given them the golden ticket, they're going to be disintermediated, and you're going to have a really big problem on your hands. And obviously, the CLARITY Act right now is the battle after the GENIUS Act over who is going to be more able to try to get funding from the depositors. Will it be traditional deposits, or will it be the stablecoin issuers?

McKee: We've got time for just one last question, that I'll put to each one of you in a row; we'll start with Charlie. What is it that you worry would be the next proximate cause of a crisis, and what tools would you use with that?

Kroszner: That's not a quick question.

Evans: It's a hard one; it requires a lot of imagination. Obviously, geopolitical risk, I think, is a big concern at the moment. And I don't know if it's such a big crisis, in the sense that something quickly happens, but just this long-running supply chain disruption through the Strait of Hormuz would maintain relative prices increasing in certain ways that are very unattractive, find their way into inflation indices; and the Fed has been hoping to get inflation back to 2 percent for over 5 years now and they're getting to the point where I think there's a lot of impatience on the part of policymakers for that, and so that might require some type of attention. But you'd want to be trading that off against any demand destruction and harm to the economy as you think about adjusting interest rates. But that's certainly a possibility going forward.

McKee: Jeremy?

Stein: Well, there's an obvious one that I'm sure everybody has thought about, which is just our trajectory of government borrowing combined with the fact that, as we've been talking about, real interest rates are likely to be much higher going forward than they have been in the past. So, I think that's sort of an obvious one.

One that I don't think is nearly as likely to be important, but I think is more underappreciated, is leverage in the equity market. I think the received wisdom is, equity market bubbles are much less threatening than credit bubbles. There has been enormous growth in the past couple of years of dealer firms providing leverage to hedge funds against equity investments, in the form of something called a total return swap—which, it used to be, you margin borrowed and you put up a 20 percent margin. Now, you're putting up basically 5 percent margin, and so people are taking much more leverage in equities.

Combine that with the high valuation of the market now, and the fact that the same firms are in private credit that are leading the stock market boom, I could imagine something going there. I don't think it's the number one, but it's maybe a little less appreciated.

McKee: Randy?

Kroszner: It just seems like there's a lot of hidden leverage. You gave one example: people are concerned that maybe they're in private credit, maybe they're in the stock market, maybe elsewhere. We just don't have a good handle on that because you've pushed so much outside of the traditional banking sector. Because if we don't know, the markets also don't know.

And so, then there's a pocket somewhere where there's a significant revaluation of the private equity pricing, so suddenly everyone marks all of these assets down by 60 percent. Well, that could have a—even if the private equity firms themselves are gated. And so, if people can't take money out there, they may take money out from somewhere else.

So, there may be all these knock-on consequences, and—exactly as Jeremy said—in the context of very high debt and deficits, not only the US but globally, and we're seeing interest rates around the world move up very rapidly in a context of high government spending and concerns about inflation. And then getting back to what Charlie said, Governor Waller had a great title for a speech that he gave: One Transitory Shock After Another. And it seems that that's what we're having now, which gets back to that original quotation that I had from Jay Powell: "the world just seems like there are a lot more shocks, and often on the supply side." But when they constantly come, it's not something that necessarily the Fed can look through—and doesn't have the tools to deal with.

McKee: Well, we've certainly had one transitory shock after another—bank shocks, viral shocks, and now with hantavirus and the geopolitical stuff we could maybe have two at once. Thank you to the panel for joining us this morning, and thank all of you for your great questions.