2023 Financial Markets Conference - Old Challenges in New Clothes: Outfitting Finance, Technology, and Regulation for the Mid-2020s

Policy Session 4: Mitigating Risks and Preserving Financial Stability

In response to persistently higher inflation, central banks have raised interest rates at the fastest pace since the 1980s. As the economy weakens and economic and financial uncertainty grows, monetary policy's effect on financial conditions and market volatility has become increasingly important in setting optimal interest rate and balance sheet policies. Panelists Viral V. Acharya, Seth Carpenter, Kathy Jones, and Brad Setser discuss falling house prices, illiquid markets, a strong dollar, refinancing risks, and other financial stability concerns with Dallas Fed president Lorie Logan, followed by closing remarks from Atlanta Fed senior vice president and associate director of research Paula Tkac.


Lorie Logan: Looking forward to the discussion that we'll have today. I thought I'd start just by introducing our panelists. First, Viral Acharya, C.V. Starr Professor of Economics in the Department of Finance at NYU School of Business and former deputy governor at the Reserve Bank of India, where he was in charge of monetary policy, financial markets, financial stability, and research.

I'm also joined by Seth Carpenter, managing director and chief global economist at Morgan Stanley. Seth was previously chief US economist for UBS and head of research for Rokos Capital, and prior to that spent almost 20 years in public service, 15 at the Board of Governors where we were colleagues for several years, and at the US Treasury, where he was nominated to be assistant secretary for financial markets.

I'm also joined by Kathy Jones, managing director and chief fixed income strategist for the Schwab Center for Financial Research, and prior to that was a fixed income strategist at Morgan Stanley and executive vice president of the Debt Capital Markets Division at Prudential.

Lastly, Brad Setser, Whitney Shephardson senior fellow at the Council on Foreign Relations. He served previously as senior adviser to the US Trade Representative, and as deputy assistant secretary for international economic analysis at the Treasury, as well as on the staff of both the NSC and NEC as director for international economics.

Welcome to you all. Thanks for being here. It's a real pleasure to be part of today's panel discussion on mitigating risks and preserving financial stability in an appropriately restrictive policy environment. It's fair to say that this panel is exceptionally timely.

In response to high inflation, global central banks over the past couple of years have raised interest rates at the fastest pace since the 1980s. We've seen repeated financial stresses, including the liability-driven investment fund shock in the UK last year and the recent stresses at some US banks. Central banks will continue to need tools and strategies for restoring price stability while also maintaining financial stability, so I'm hoping we can use this afternoon's discussion to draw out some approaches that may work and identify some challenges.

I'd like to open the panel's discussion highlighting three issues that are top of mind for me. First, how to prevent financial instability in the first place. Second, when central banks must intervene for financial stability reasons, how to avoid working at cross purposes to monetary policy. Third, how to design monetary policy strategies that mitigate financial stability risks, but still achieve the appropriate macroeconomic goals. As always, the views I express are mine and not necessarily those of my Federal Reserve colleagues.

Let me just start with that first issue. Interventions in response to financial stability can have undesired side effects, so prevention is better than cure. There are many aspects to prevention, and it would be valuable for central banks to conduct a thorough review of the many stress episodes that have unfortunately occurred since the global financial crisis. This could help draw out how prevention needs to work in different interest rate environments, and we'll get to some of these issues in our discussion.

The second challenge is how to design the cure when prevention doesn't work. In the current inflationary environment, interventions to support financial stability must not inappropriately ease financial conditions. This isn't easy because central banks use the same basic tools: lending against assets or buying them, both to provide monetary accommodation and to support financial stability.

There can never be a complete separation between monetary policy and financial stability. Central bank tools have multiple effects, and financial stability is necessary to sustain strong macroeconomic outcomes. Still, financial stability is not identical to price stability, and central banks will often be able to perform better against their multiple goals when they do have distinct tools to deploy.

The Federal Reserve's and Bank of England's actions over the past year demonstrate two ways to draw this distinction. One, using loans, and the other with asset purchases. In the Fed's case, we responded to the banking stresses with discount window and emergency lending, while continuing to raise the fed funds target range to address above-target inflation. Lending-based interventions like the discount window or repos are relatively straightforward to separate from monetary policy actions. This is because their duration is relatively short, and in part because a central bank can offer loans at backstop price, relative to market interest rates, while monetary policy continues to influence the overall level of market rates.

For example, the interest rate on our bank term funding program is indexed to 1-year OIS, which in turn reflects market expectations for the fed funds rate. Thus, the program adds liquidity to the banking system, but at a price that shifts with our monetary policy decisions. I would note that in determining the appropriate stance of policy, we need to offset the macroeconomic implications of both the banking stresses themselves and the liquidity we are adding in response.

In the Bank of England's case, pension funds last fall were selling long-dated UK government securities to meet margin calls caused by rising interest rates. Those heavy sales created market dysfunction. Lending wasn't an effective solution because the pension funds lacked the capacity to borrow, so the Bank of England needed to buy securities, but without easing monetary policy.

The Bank of England pioneered several useful tactics informed by the findings of a Bank for International Settlements study group that Andrew Hauser and I led for the market committee. First, the Bank of England separately tracked its asset purchases in response to the LDI shock and sold them in a timely way once market functioning normalized. Second, the bank only purchased assets that were distinctly affected by the shock, not a broad cross-section of securities that would have been appropriate if they had been providing monetary accommodation. Third, the bank used a form of backstop pricing to purchase only as many assets as were required to address the market dysfunction, and the bank's communications explicitly linked the purchases to their financial stability mandate.

The Bank of England's approach does not provide a perfectly replicable blueprint for all cases. That's because the shocks that we see are diverse, and institutional arrangements vary across central banks. For example, the Bank of England has distinct monetary policy and financial policy committees, which helped clarify the purchase's goals. The Bank of England's actions still highlight how key principles, careful and transparent communication and backstop pricing, can help central banks navigate these types of situations. This remains an area of active research, and I hope we'll draw out more ideas in our discussion today.

Lastly, we could think of designing monetary policy strategies to mitigate financial stability side effects. I don't mean examining how monetary policy strategies can trade off achievement of macroeconomic and financial stability goals. Rather, the challenge is to set monetary policy in a way that mitigates financial stability risks but still achieves the macroeconomic goals. This is possible because the restrictiveness of monetary policy comes from the entire policy strategy, how fast rates rise, the level they reach, the time spent at that level, and the factors that determine further increases or decreases. These levers can conceivably be arranged to maintain the restrictiveness of policy while reducing financial stability side effects. In particular, gradual policy adjustments can be helpful. Financial conditions can sometimes deteriorate nonlinearly, doing damage to the broader economy, but the risk of a nonlinear reaction can be mitigated by raising interest rates in smaller or less frequent steps, while using other dimensions of monetary policy to maintain restrictive financial conditions.

In a recent speech, I used the analogy of a road trip in foggy weather. When conditions are uncertain, you may need to travel more slowly, but a slower pace of tightening shouldn't signal any less commitment to achieving the inflation goal, any more than slowing the car would suggest you don't want to get to your destination. With that said, I'm looking forward to hearing the perspectives of our panelists on these and other issues. We'll dive right in with some questions around these three themes, and then we'll open it up to all of you for further questions.

First, I wanted to explore the question around the pace of changes in the policy rate and their effect on financial stability. I thought I'd ask Seth and then Kathy to comment on this theme first, and then open it up for Viral and Brad to comment. I had three questions in this theme for the two of you. If you could first talk about the channels for these effects: the impacts on financial institutions, individual investors, and economic activity that feeds back to financial stability. What are those channels that we should be focused on? How can we distinguish the normal effects of higher interest rates on the financial system and economy from the unexpected financial instability? Then, how should central banks weigh those trade-offs about how quickly to move?

Seth, why don't I start with you, and then turn it to Kathy?

Seth Carpenter: Absolutely. Thank you very much, Lorie, and thank you, Raphael Bostic and the Atlanta Fed, for having me here. Thanks, all of you. One place I'd like to start, Lorie—I personally thought it was helpful when you talked about financial stability and monetary policy—you didn't make it a stark dichotomy that a central bank would have to think about one at the expense of the other.

In fact, I would go even stronger. In general, it's a false dichotomy. When I think about having been an undergrad, having been a grad student, having been a professor, and then working at the Fed, thinking about what monetary policy is supposed to be doing now, and tightening monetary policy, it is supposed to be, in general, raising short-term interest rates. Maybe throw QT in there, with a specific purpose of tightening financial conditions, making funding costs for intermediaries higher not lower, making intermediaries willing to extend credit lower not higher.

When we think about at least what's going on here in the US with some of the bank turmoil, everyone should think very, very carefully: how much of what's going on in the banking system is an exogenous shock of financial instability that is complicating the lives of the FOMC in terms of doing monetary policy, and how much of it is just monetary policy tightening? I hear clients all the time saying, "What's the Fed going to do? The cost of funding for banks has now gone up, and it's gone up pretty substantially. They're much less willing to lend. There's going to be a credit crunch, they're pulling back from funding businesses across the country." My reaction sometimes is, "Yeah, go on. You've described monetary policy."

Now, the challenge is one of calibration, and in some sense then the definition of financial stability or financial instability that we're worrying about is a large, possibly nonlinear, tightening of financial conditions, or breakdown in the transmission of credit through the economy, that results in more slowing of the real economy than is anticipated. That could be a useful working definition of financial instability concerns here. On the other hand, if you ignore financial stability in this context and just think about what the Fed is trying to do with monetary policy: they are trying to raise interest rates, they are trying to tighten policy enough that the economy slows down and inflationary pressures come down, but not so much that they cause a severe recession. So again, for me, the distinction between financial stability concerns in the current context and monetary policy are much more muddy than most of the public discussion has in mind.

That's one starting point. The other point, and very specifically to the narrow question you asked, is about speed, and it does matter. This is one point where it's very helpful to have these sorts of conversations where you have academics, you have policymakers, you have market participants all talking together, because the different perspectives matter a lot. If I think back to my time at the Board working on the Tealbook, there were the different simulations for optimal policy and things like that. The difference in most of the macro models that get used at the Fed, that go to the Board, between waiting for a long time before starting to raise rates and then raising rates aggressively versus a much more measured pace, the difference in macro-outcomes tends to be pretty small. You can sort of replicate the two even by starting later.

The difference in the real world where most of us live can actually be quite severe. For example, if you think about the banking turmoil, a longer period of time to recognize the writing on the wall where asset prices are going to restructure their books: more time could have helped, even if you end up, as you said, getting to the same outcome. The flows within money markets really do matter in terms of the speed here, and one of the things that I know you want to talk about later is monetary policy tools. This hiking cycle has been fast. It's gone to a relatively high peak if we're done. On top of that, it's also using the reverse repo facility in a way that was used in the previous sort of quasi-hiking cycle, but not in previous cycles. Anything anyone thinks they know about the transmission, especially through money markets, of monetary policy tightening, it just has to be somewhat different because of the existence of the reverse repo facility. As a result, faster movements in money market rates create larger dislocations, or result in larger incentives for swings in assets through money markets. Yes, to answer your question, the speed really does matter a lot for the transmission.

Logan: Kathy, maybe turn to you?

Kathy Jones: Yes, I would agree with Seth about the channels. We all know what monetary policy is trying to accomplish. That's pretty clear, but when it comes to the speed, it matters even more than you do. As a market participant, for me the rate of change is a key factor. We look at things like the MOVE index, which is volatility in the bond market. Not only has this been a rapid rate hiking cycle, but we've had a move up in volatility that really surpasses almost everything we've seen in the past, and that's because of the accelerated pace of rate hikes.

One of the things the pace has done is create a sharply higher level of uncertainty about the path of interest rates, and that's really all the volatility is reflecting. You take that down to, then, the real world where people in financial institutions, businesses, and individuals are trying to deal with this rapid rate of change and volatility, and you're going to have some feedback that is potentially pretty destabilizing. In my view, the speed has been ... I've said this publicly before, so I can say the speed has been a bit excessive in recent months, and a more calibrated approach would have been better. It also obviously exacerbates liquidity issues when banks, when individuals, when investors, when mutual funds, have to change very quickly their strategy based on their expectations, then you get a shift in liquidity that can create more problems.

That reverberates all the way through the chain, in terms of the availability of credit and liquidity to businesses and down to individuals. You can't get a mortgage, it's harder to get a car loan. Those are all intended consequences. Whether they're intended to be sudden is the question. The other outcome has been deleveraging. Obviously, that's affected the banks, and other areas as well. Again, a very rapid demand to deleverage, in order to meet liquidity requirements, and then this huge shift in the yield curve that we've seen which, historically, has been pretty significant. The rate of change has been very important in this cycle, and very different from anything most people have experienced. I started in the business during the late '70s, early '80s, so I do remember rapid rate hiking cycles, but our financial infrastructure and markets are very different now, and it moves very, very quickly through the system.

Logan: Viral or Brad, would either of you like to comment on the question about how central banks should think about this trade-off between the destination and the speed?

Viral V. Acharya: I fully agree with what Seth and Kathy have just said. Just one thought I wanted to interject in there is whether we need to think a little bit about the part dependence of these outcomes. We've now gone, since the global financial crisis, into I would say three rounds of monetary easing. First, there was the immediate easing after the financial crisis. We tried to unwind, and we caused a taper tantrum in the emerging markets. Then we postponed, and I'm calling that the second round of easing. Then we tried to raise rates. We succeeded, we did quantitative tightening, and then we ended up with the repo market spike of September '19. Then we started easing again. Then came the pandemic, we eased a lot, we are trying to exit, and now again we have some accidents coming out.

I want to fundamentally question whether the celebrated forward guidance "low for long" that macroeconomics is claiming as the victory of monetary policy after the financial crisis. We need to take a pause and evaluate it. We are ending up with a lot of financial fragility, sometimes outside the country, sometimes inside the country. From what I can see, it's not working. It's not working as planned. It's a little bit sort of thinking on the fly on exit. The Fed balance sheet is becoming permanently larger after each round of this easing. I just want to throw in this question: whether we are easing too much, and then we are trying to do too much on the way out, and whether this combination is leading to a lot of sudden, unexpected consequences for the economy.

Brad Setser: I'll throw in two observations. The first observation is to pick up on the driving through the fog metaphor, which is quite helpful. My sense is that because of its forward guidance, when the Fed started this tightening cycle it felt it had to speed through a very foggy road, just because it felt that it was starting late. For a three-hour trip, the Fed had waited 30 minutes too long to get started and wanted to make up time, which intrinsically adds to the riskiness of the tightening cycle.

The other observation would be, and it's just perhaps something that struck me, is that even though the Fed was tightening really, really rapidly and really aggressively, deposit funding costs for banks were responding with these incredible lags. We had a really fast tightening cycle that was playing through very quickly into the bond market and into the foreign exchange market but playing through very slowly to bank funding costs. It may be that the bank funding cost mechanism of transition was more powerful than we thought, and it played out a little bit more slowly despite a fast-hiking cycle.

Carpenter: Just on that last point, Brad: The point you're making on bank funding costs and how it played out highlights how every cycle can be a little bit different. Lots of people in markets see QT as the reason why bank deposits are going down, by entirely ignoring the fact that deposits started declining at lots of banks well before QT happened because we saw money market rates going up and banks did not start to raise their deposit rates, in no small part because they had a lot of nonoperating deposits that they didn't want and at least initially that runoff was welcome. That's the pattern we see in just about every interest rate hiking cycle, and this time was no different.

What went differently this time versus the other is that it kept going for longer, and partially faster. There probably is something to the story that it's easier now, for all sorts of deposit holders, be they retail or institutional, to shift money into money funds, or at least the instinct to do it might be coming faster with the online communication. Your point about the funding costs starting off rising very slowly, if at all, and then over time picking up much more aggressively, is exactly right. It highlights for me exactly how similar this hiking cycle is for the transmission of monetary policy, but also the ways in which it may be a calibration exercise because things did pick up even faster, and perhaps we've reached the limit or maybe gone just a hair too far.

Logan: I'll take one of the audience questions, which is related to speed. The question is the speed with which rates rose has created a class of trapped homeowners. They're sitting on 2.5 percent mortgage rates, unable or unwilling to sell. This has dampened the supply of homes for sale, and prices are rising again. Is this a problem? I guess the question is that about speed, or is that about level, in the question that's coming in? Any comments on that question, before we move to the global questions?

Acharya: It's probably about both, but I agree, you would get this even if you gradually went to that level. I'm forgetting the name of the authors, but there's a very interesting paper, a research paper from University of Urbana-Champaign, that's finding that these mortgages where the rates are very low relative to the market rates. People are very reluctant to leave homes as a result of that because they would have to close their first mortgage and then take out a new mortgage at the new rate.

There are some signs that this is actually adding to the tightness of the labor market. I thought that was an interesting, somewhat unexpected consequence of the fact that once again, rates are very low, and then we are trying to raise it very fast, and then there it's throwing some wrinkle in the adjustment of the economy to labor shortages.

Carpenter: The labor mobility issue is potentially an unanticipated consequence. For the housing side of things, it's also important to point out that if people don't feel like they can move, just to trade up to a different house, they are in fact taking away some supply from the market, but they're also taking away demand from the market. That that particular phenomenon is responsible for driving up house prices alone is a little bit harder for me to embrace fully, because the same people who are stuck in their home are not demanding a different home in the market.

There is a fact that, relative to a very long run, sort of demographically calibrated sort of level, there's too little housing in the economy. That was a preexisting condition. At the end of the day, the increase in mortgage rates, at least initially, dropped housing activity (new home purchases, existing home sales), not just because the supply was constrained but because the demand got whacked as well. People who wanted to buy their first house, their affordability was worse. The affordability to buy a new house is, if you take the past six months or so, the worst it's been literally in decades. I'm personally skeptical that the particular mechanism is first-order important.

Jones: Yes, I would just add that I do think it's more level now. Speed has something to do with it, but in most housing cycles it isn't just all the way up and all the way down. You kind of freeze a standoff between buyers and sellers, and then it tends to work itself over out over time simply because sometimes people do have to move, sometimes people have to cut the price of their home if they're selling. Sometimes they're cash buyers. A good 40 percent of the existing home sales have been cash purchases, which is keeping prices elevated. I would say that this is playing out in a fairly, maybe rapid, but sort of a normal level. You're not seeing the housing market now continue to roll over and die, so this seems like a normal pattern and a normalization to higher mortgage rates than we've seen for a long time.

Logan: Okay. I'm going to ask one more speed-related question from the audience: Given that the road still is foggy, what's the rationale for the Fed pausing now? Is there evidence that it's arrived at its destination? Sufficiently restrictive? Or is it just being precautionary?

Acharya: My sense is they just can't take the bank failures that lightly. If you don't have financial stability, you can't conduct your monetary policy in any reasonable way. If I was in their shoes I would hope for a few points down as soon as possible and push the pause button. Because you do want the banking sector bleeding to stop as soon as possible. I agree with Kathy, which is you want the normal transmission of the monetary policy, but you don't want it to be sudden and associated with financial fragility. So normal transmission is good, but unexpected and loaded with bank failures, that's a no-no in my view.

Carpenter: My sense is the answer to your couple of questions is "yes" to all of them, for all the reasons that Viral mentioned. We have seen activity in the housing market come off, notably we have seen consumer spending on durable goods turn negative. If you take any smooth measure, nonfarm payrolls are coming down, albeit still just over 200 is pretty good, but it's much less than it was six months ago, which is less than it was six months before that, and so on. If we take seriously most macro models, that the level of interest rates matters a lot, it's a reasonably straightforward argument to say that you're in restrictive territory. As a result, the longer you stay in restrictive territory, you'll continue to impart restraint on the economy.

Is it enough? Will it bring inflation down at the speed with which the Committee wants it to come down? Hard to know, but the Committee has been pretty clear that they're willing to take a long time for inflation to come down. The SEP has core inflation almost to target at the end of 2025. The Committee started raising interest rates in March 2022, so almost four years is the appropriate amount of time the Committee has said it is fine to take to get inflation almost to target. Through that lens, I don't think it's a crazy thing for the Committee to wait now. With two-way risks in the future, it could be that is not enough, and the economy is resilient and reaccelerates, in which case, presumably, there's more hiking to be done.

Logan: Let's broaden out the conversation and turn to some global considerations. Brad and Viral, what have we learned from the tightening cycle about the potential spillovers from policy tightening in one country, to financial stability in other countries?

Setser: I'll start. I guess the first observation that comes to mind is, we actually know a fair amount about international transmission at this point. We know that for a subset of countries US rate hikes, tighter monetary policy in the US, is a positive impulse to other economies through the traditional exchange rate channels and a shift in demand away from the US towards the rest of the world.

We also know, and people have a great deal more confidence in the second component than they had 15 years ago, that "the dollar is a global currency" means that Fed tightening doesn't just tighten conditions in the US. It tightens conditions globally. For important parts of the global economy a Fed tightening cycle is a global tightening cycle, and it plays out globally. We knew that before, and our conviction around the need to think in a differentiated way about monetary transmission from the US to the rest of the world is stronger.

The second observation is that financial markets respond very quickly. Foreign exchange markets respond extremely quickly. In some ways, it certainly seemed for a while last year like the foreign exchange market was responding way more rapidly than domestic transmission mechanisms were kicking in. For a while last year, you could almost imagine a world where, say, there was an exchange crisis around the yen before you had really seen Fed tightening impact the US labor market. We were fortunate that that possibility didn't materialize. I would say one lesson I drew from that, which is perhaps a bit controversial, is that foreign exchange intervention, even in advanced economies with large, open capital markets, can in some cases work. The Bank of Japan was able to set a floor on the yen. The Bank of Korea, also a little more complicated, was able to set a floor under the yuan in part because the market had overshot domestic transmission. By buying a bit of time, producing a bit of two-way risk, the counterparts around the world, central banks, were not following the Fed and bought the US economy time to catch up with the lags in the transmission mechanism.

The third observation is that there is a lot of uncertainty and complexity in international transmission. Kristin Forbes and I were discussing this a little bit. In most classic models, higher interest rates in the US across the curve lead to increases in foreign demand for US bonds because US bonds yield more. That's kind of the standard transmission mechanism. In practice, there are large institutions, globally, that fundamentally run hedge books that borrow short term and buy long term. Probably two-thirds of the demand from Japan was from hedge books, and for hedge books the shape of the curve matters much more than the absolute yield. We saw empirically Japanese institutions reducing the size of their foreign bond holdings in response to Fed tightening and higher bond yields.

Now, both mechanisms play out. It's an empirical question what matters more. There are complexities that come from the fact that the shape of the curve matters, as well as the level that are not necessarily always well captured. Then some of the dynamics around Silicon Valley Bank and the loss of the value of your existing portfolio, the implied mark-to-market on your hold-to-maturity, can be an implicit hit to your capital that can impede risk taking. There are complexities internationally just as there are complexities domestically.

Acharya: Thanks, Lorie. I'll give the perspective from an emerging market central bank, having dealt a little bit with trying to manage the dollar cycle at the Reserve Bank of India. What I saw from my vantage point was that the large, institutionally capable emerging market central banks, they've come a long way in the last 10 years in managing the dollar cycle. They realized that the waxing and the waning of the Fed balance sheet or the liquidity cycle is going to cause huge movements in their volatility.

Of course, first, they have done the traditional thing when your currency is appreciating. Huge capital flows are coming in so build the buffers. At the same time, put some capital controls in place: don't have huge bond market flows, short-term bond flows, et cetera. There was a time when IMF was not terribly supportive of capital controls, but that has shifted fundamentally, notably after the taper tantrum. Large, institutionally capable, emerging market central banks are now holding big fortresses of reserves to stabilize their currencies.

Second, after the taper tantrum they realized that it's the foreign currency debt, the original sin if you want, that exposes you to the sudden stops, so let the foreign investors bear some of the exchange rate risk. There's a lot of local currency-denominated debt now. While they are putting capital controls in place, they are saying, "Okay, local currency debt I can tolerate a bit more. The exchange rate risk is with the investors." Maybe rather than doing a sudden stop, they will start pulling out gradually once they see the tightening cycle. This is exactly how the capital outflows have played out over the last year.

Last, and perhaps the most important thing, is that central banks like Brazil and India did not even have an inflation targeting regime. What they realized after the taper tantrum is that you can use your reserves to fight the currency, et cetera, but ultimately what the foreign investors are looking for is: do you have an inflation targeting credibility in your central bank, or not? These central banks have adopted these regimes that are still works in progress. Reserve Bank of India was 2016, Brazil was about two or three years after that. Interestingly, many of them, not all, tightened in advance of Fed tightening, just in response to the inflation in the US.

To just sum up, my sense is the emerging markets are now much more prepared for the dollar cycle. Of course, COVID was like a very, very different kind of a shock. No one could have been prepared for it. They took the queues from the taper tantrum. They took the queues from the rising inflation in the US, and they proactively prepared for what's actually unfolding in the US dollar cycle.

Logan: In addition to the fact that they have larger reserves, we also have different policy tools, so the swap lines are a key part of the framework, and we now have the FIMA repo facility. What role do you think those two tools played, and do you think there are any changes needed for those tools as we look ahead?

Acharya: What we saw at the time of COVID was that the Treasuries essentially were being run upon, because emerging market exchange rates were crashing. They needed to take out their dollar reserves and sell them in the market to support the currencies. Combine that with some local hedge fund-related leverage issues in the US, that meant that Treasuries, instead of being a safe haven, in that time actually looked a little bit like the asset class under trouble.

The Fed did something very smart and clever. They've kept it on, and I give them tremendous credit for it, which is they basically said, "Why do I need to worry about an emerging market central bank selling the Treasuries in the market? Maybe I just give them a swap line, or a standing repo facility. If you have the Treasury collateral, give it to me. I'll lend dollars to you against that at market prices."

This is very good, because if there are some deleveraged liquidations going on in the Treasury market, you don't want the central bank liquidations adding on top of that. This has given a bit of comfort to the central banks that the kind of COVID situation won't repeat itself as a Treasury market dysfunction. Some of them had been batting for swap lines from before, even before the COVID situation, but the repo facility, the standing repo facility for the central bank reserve managers, pretty much obviates the need for swap lines, in my view. I give very high points to the Fed and the teams that put this in together as an emergency facility and they've kept it on.

Setser: I'll toss in a few thoughts. One observation is that the FIMA repo, this capacity to get dollars against your bonds, it wasn't used during the COVID sell-off in large part because the Fed responded to the COVID sell-off by buying an awful lot of Treasuries, which was an appropriate response at the time because there was no policy reason for Treasuries to be going up. The Fed didn't have to worry about trade-offs in that particular moment in time, whereas FIMA repo lets the Fed provide liquidity without having to step in and buy the bonds, which seems like an efficient outcome.

I also think we discovered ... I'm going to have to go out on a limb because we don't know for sure who used the FIMA repo facility in the past couple of months ... it is quite possible that a central bank, Switzerland, with access to the swap lines used the FIMA repo facility to provide a dollar lender of last resort backed by its own reserves in the context of bank resolution. Again, it is helpful to have more tools, not fewer, and not necessarily to have had to activate the swap lines.

A final thought is that the swap lines serve a particular role, in my view, in the system. They don't solve all problems. They solve one particular problem, which is that there's an awful lot of dollar-based intermediation that takes place outside the United States. The Fed, by lending to other central banks, lets other central banks be dollar lenders of last resort to their own financial intermediaries. That was enormously important in the COVID shock. It hasn't been, for various reasons, as important right now, but it's an enormously important capability.

To my mind, the perimeter or the boundary around who is given eligibility is a little bit dated. I personally would support expanding it to include, in particular, Korea, which on a number of criteria is very analogous to the countries that already have access to the swaps. It is slightly delicate, because for complex reasons I wouldn't extend it to Brazil, which historically has been treated symmetrically with Korea. It ends up being complicated because you have to make hard judgments with consequences for US foreign policy. I absolutely would not extend it to Turkey, even though the odds that Turkey will make a request in the next five months are quite high.

Logan: I can see from the questions coming in that there's a lot of interest in the next section of questions that we were planning on talking about, which is: How has the evolution of monetary policy implementation tools in recent years, including the new facilities and the expansion and contraction of the balance sheet, affected the relationship between monetary policy and financial stability? Kathy, maybe you'll start us off, and then I'll turn to Seth on this one.

Jones: Sure. A number of the tools have been extraordinarily helpful, and we've talked about them here, in allowing the central bank to set monetary policy separate from financial stability. A number of those tools have been very helpful. When it comes to quantitative tightening, I find it a little perplexing in some ways. We know the channels it's supposed to work through, but in real life it doesn't quite work that way. It was Fed chair Bernanke who said it works better in real life than in theory, and that really seems to have been played out. Although we know it shrinks liquidity, on the other hand what we see in real life in the market is during periods of QE we actually saw a steepening of the yield curve and rising long-term interest rates, and vice versa during tapering and during QT. Of all the channels through which that works, it seems like the signaling is the main channel.

I'm not sure that that's been embraced as much by Fed policymakers as it could be, because in simple terms, if you're engaged in lowering interest rates and expanding your balance sheet, you're sending a signal that you want growth and inflation to go up and the long-term rates tend to follow that trajectory, and vice versa. If you're raising rates and signaling that you're tightening policy, especially by reducing the balance sheet, you're sending the opposite signal and we're getting this inverted yield curve. I'm not sure that that as the signaling part of the tightening in monetary policy is, at least from what I've observed from the Fed, as appreciated. Maybe managing that in a different way, as a part of ... there's very little discussion other than, "We're doing QT, and we're set it and forget it. It's just going in the background." Not necessarily a bad thing, but addressing that more as what the objective really is and how the interplay with interest rates would be, would be very helpful for the market to understand and to function in.

Logan: Seth?

Carpenter: One of the interesting parts of QT, in addition to that, does involve the interaction between the level of reserves and the level in the RRP, and then just as I was alluding to before, the effect of the RRP in transmitting short-term interest rates more broadly across money markets is new this time, and it's worth studying. As you well know, from the beginning, creation, and design of the reverse repo facility there was just never a conception that there'd be $2.5 trillion in the reverse repo facility when it was first designed. It was just, "yes, we'll be able to put a hard floor on money market rates," but the size of it is just massively different now than when it was first being discussed and designed, in the before days.

Logan: It's fair to say that the size of asset purchases we did, was also not considered at the design phase.

Carpenter: I don't disagree. Whenever I talk about where we are in the world and a lot of policy, both monetary and fiscal policy, and regulatory and otherwise, I'm reminded of the children's song about the little old lady who swallowed a fly, and then she swallowed a spider to catch the fly, and then she swallowed a bird to catch the spider to catch the fly. I sort of feel like we're a little bit in that sort of circumstance, in the number of different situations. I do think the difference in tools matters a lot. So how does it matter? First, for QT, at least as I read what the FOMC has said, they want to shrink the balance sheet, get to ample reserves, in particular as a liability item. Gosh, at least it's my inference, that it would be really nice if that very large, overnight reverse repo facility actually went to zero as well, so that the overall size of the balance sheet is undetermined by that, and more just by reserves.

So far that hasn't been the case. One natural version of how that could work, though, is as the balance sheet keeps shrinking and banks start to lose reserves, they look around collectively and say, "I kind of liked that really liquid asset on my balance sheet." Then you get a little bit of a bidding war, market rates go up, and the RRP starts to fall because market rates go above what the Fed is paying on RRP. I always assumed that was part of the plan, but it hasn't worked out yet. We're seeing some tension with banks and funding markets more generally, so how that plays out over time is actually very unclear. At some point, we might bring the unfortunate debt limit into all of this, because that's just going to make that whole process that much more difficult. Nevertheless, I do think in terms of tightening policy and shrinking the balance sheet, the "learning by doing" with how the tools interact with each other is quite difficult.

Then again, the transmission to depositors, to deposit rates, to other money market rates, because the reverse repo facility does in fact impose a very effective floor, just means that whatever you could estimate, using all of the data from previous cycles to come up with whatever policy rule you want, whatever model-based estimate of the transmission of monetary policy to, first, financial markets and the rest of the economy, it just has to be different this time because you have this tool that doesn't allow as much flex to the downside for deposit rates and other money market rates. That fact just should not be ignored in doing any model-based simulation or any comparison of "where are we now?" compared to any rule, if any of those parameters are estimated in the before days when you didn't have the tool in place. In this case, the tool matters very dramatically.

Logan: Let me open it up to see if others want to react to that, but also just put in a couple of questions that are related to this that are coming in. On the first one, and this came up in one of the earlier panels, should the Fed evaluate changes to the RRP to mitigate bank deposit outflows if instability exists? This could be in the form of lowering the rate on the overnight RRP facility. Should the Fed do that? Any thoughts for or against that change? We'll start with that one, and then I'm going to come to another QT question. Would anyone like to take that question?

Acharya: I'm not a big fan of letting banks make money just on having a huge market power over deposits. It's a form of financial repression of the savers, at the end of the day, so if there are alternatives like money market funds that can compete with banks to raise the savings rate for the depositors, some of my hunch is that's going to be a better design in the end because that means banks are not always thinking that their deposit betas are going to be very, very low just because they've been in a low interest rate environment. If they know there's a money market fund that can now generate an overnight return, just the way banks do, they know that at least their uninsured deposits ought to be fragile. Maybe they missed it this time, but next time around they will remember.

Net-net, that seems to be a good thing to me. If we don't have the savers earning the rates that the Fed is wanting to pass through to the real economy, if bank funding costs don't adjust to Fed's rate hikes, we won't have a pass-through. Then we'll have to wait for very long for the transmission of monetary policy to bring inflation down. While it is a fundamental challenge to the monopoly over deposits that banks have—let's say it's a sort of soft monopoly—it's needed, because RRP will improve the transmission of monetary policy to the economy, in my view.

Carpenter: If I could take the way you phrased that, Viral, and then sort of twist it just a little bit. I don't think I fundamentally disagree with anything you said, but the way I see it is the RRP then does do exactly what you're saying. It sort of forces that cost onto banks to the extent that they want the deposits as funding, they now have to pay above whatever the RRP is paying the depositors as a competing fund. I, so far, completely agree. The challenging part is one, in previous interest rate cycles, it was not the fact that banks' funding costs didn't go up. They did go up, they just didn't go up quite as much, for a given increase in the federal funds rate, and now with the RRP rate there.

For the same 500-basis point increase in the target range, you're getting more of an increase in funding costs on banks. To the extent that that bank channel is an important part of the transmission of monetary policy, then, for me, it becomes a calibration exercise. If 500 basis points in previous cycles did the amount of tightening that you want, and the RRP is imposing this funding cost on banks so that 500 basis points now is worth more in terms of higher funding costs on banks than it was in previous cycles, then you have to recalibrate and do relatively less than you would have thought if all you looked at was a transmission through previous cycles.

Jones: I would only add that just pulling that off is a pretty big challenge. We put the RRP in place to stabilize the money market funds, so it's the last problem we're solving and the way to shift away from that is potentially fraught with a lot of difficulty. We have to be talking about financial stability. You have to be careful when you want to get to an ideal state, but getting there can be a real problem, especially when we're going from one problem to another problem, to the solution to one problem is now the problem that we've created.

Carpenter: "I don't know why she swallowed the fly; perhaps she'll die." [laughter]

Logan: There's one question here. If we want a smaller overnight RRP program, why not cut the overnight RRP rate? Viral, you made an efficiency argument into money markets more broadly. Maybe another way of reframing that question to the discussion we just had is: If the overnight RRP were 5 basis points lower, would you see a significant difference in the size of the overnight RRP today?

Acharya: Probably. It ought to be sensitive to the gap that the money market funds can offer versus what the banks are able to offer. Just to come back to Kathy's point, because it is so important. We added this wrinkle in and we now have greater competition for deposits. In the end, though, I'm not sure if that is what has caused the fragility of the banks. We were seeing smooth outflows coming out of banks for a few quarters, as my colleague Philip Schnabel's work shows. It was a little bit the fact that there was a lot of duration play on the balance sheets of some banks, and the news that really triggered the financial fragility was actually the lack of capital on these banks if you had really marked them.

It wasn't so much that they had lost $10 billion in deposits in the last quarter, and therefore let's pull out of this. My limited takeaway from the data so far would be that RRP did cause the loss of deposits at banks, but it was playing out all of last year, and that the financial fragility has been caused more by the fact that some banks are low on capital, if you mark their assets to market. That seems to suggest to me that maybe RRP is working okay for its intended purpose. Now, it could be that in the current environment if you want to stabilize banks, maybe one way is to say, "Okay, we'll let you keep the deposits. Don't offer them high rates." Personally, I would resist going down that path. It may be better to restore bank capitalization in some other way rather than actually limit the interest rate on the RRP.

Carpenter: I agree with everything Viral said, but in addition, directly to the way the question was asked, there are a few points. One, even before the Fed started raising short-term interest rates, and with it the RRP rate and everything else, we saw deposit outflows starting because market rates had already risen higher and banks weren't catching up with that for deposit rates. At that point, if we had gone back to the previous cycle, moving the RRP rate up and down when banks have more deposits than they actually want just moves money market rates up and down. There's a version of the world where just lowering the RRP rate just pulls all market rates down with it. The way that simply lowering the RRP rate gets the RRP down is because market rates have already been bid above the RRP rate, and you're able to introduce a gap.

If there actually is in and of itself a desire to rearrange the Fed's balance sheet so that RRP goes down on the liability side and reserves are the beneficiary, then it's not just lowering the RRP rate, but it's presumably simultaneously raising the rate paid on reserves. It probably can't even be one for one. It would have to be much more because the interest on reserves rate has much less pull on rates. If you want to keep the stance of policy the same and keep money market rates roughly where they are for macro policy, you have to do more than just move the RRP rate.

Logan: I'll switch to another one of the tools, which is QT, for one of the questions that came in, which is really about, should the Fed stop the balance sheet run-off when it lowers interest rates? Does it matter the environment in which that takes place, or should it be guided purely by the definition of achieving ample reserves?

Jones: I'll take a shot at that. I would think that cutting interest rates to ease financial conditions, while at the same time doing QT, is kind of running at cross purposes. I'm not sure that that's a logical outcome. They could stop QT, or certainly slow it down, and allow rates to come down, but it seems like you're one foot on the gas and one foot on the brake if you're doing that. That seems to me kind of like an illogical way to go. Over time, if the goal is to get the balance sheet down to some percentage of GDP, you can allow for growth. If you're stimulating economic activity by lowering rates, you can allow for growth to allow that to come down gradually on its own.

Setser: My view is maybe only a little bit different. I would at least differentiate between passive balance sheet roll-off, not reinvesting maturing principle, and active balance sheet reduction. I can imagine in a world with a lot of fog and uncertainty about precisely how you are going to adjust short-term rates that you would allow a certain level of passive balance sheet reduction while still cutting rates, in part because we might be cutting rates in a much less aggressive way than at some points in the past. You might want to maintain optionality to reinforce rate cuts with balance sheet stability, or even move towards balance sheet expansion if you're really concerned. I would think that the realm of choice is quite big.

Carpenter: I agree with Brad, and maybe even further along the circumstances matter a lot. If it were a circumstance where there's an actual recession and the Fed wants to cut rates in order to provide macroeconomic stimulus, or at least get rid of the macroeconomic restraint, I would think at that point it's probably appropriate to stop running off the balance sheet. At least the way I read the dot plot is, you get into next year, if inflation has in fact come down a lot, then as inflation comes down and presumably with it inflation expectations, then for a given level of the nominal rate, you're having real rates going up. You could be just gradually adjusting down the nominal rate to stay more or less restrictive.

Under that tweaking of the funds rate, I don't see an insurmountable and logical inconsistency with letting the balance sheet continue to run off there. The Committee has been reasonably consistent in saying they want to use the funds rate target as its primary tool for macro policy, and the unwind of the balance sheet is taken as a given and adjustments at the margin would be done through rates. I personally don't think it's that big of a deal when you're doing those recalibration exercises.

Setser: Another comment just to keep the dialogue going. We have one more tool that we haven't even discussed, which is composition of balance sheet. The agency MBS portfolio at some point starts to roll off in a material way, and you would have to make a decision about whether to reinvest in agency MBS, reinvest in Treasuries. Is the long run goal to get the balance sheet out of mortgage-backed securities, or not?

I actually think it's a pretty consequential choice, in part because the agency MBS purchases were probably one of the more powerful QE tools that worked, maybe more directly than just through the signaling effect. Choosing, as a matter of policy, not to do one of your more powerful easing mechanisms in the future is a consequential choice. Buying a lot of mortgages also had its significant economic consequences, and having really low-for-long mortgage rates was not entirely unrelated to the asset management decisions made by some banks.

Acharya: I'll just add in one point, which is here I lean a little bit closer to Kathy's view. It's very hard if you are changing your liquidity policy on the fly with interest rate policy, because that means you have to keep recalibrating your transmission mechanism. To know what a 25-basis point cut or hike is going to achieve, you need to know at what pace it's going into the real economy via the financial rates. If all your historical data is based on just, say, rate hikes or rate cuts, and not simultaneously using two tools at the same time, you don't have a good calibration sense in your counterfactuals of what's actually going to happen when you take on the policy. If you want to use both tools, then you need to have a really good handle on how much my liquidity tool is going to amplify the transmission or dampen it further. Otherwise, keep it simple. If you have one tool that you know works in a certain way, stick to that and maybe you'll get to the goal posts in time.

Jones: I would look forward, also, to the press conference where they try to explain that recalibration process in great detail. That would be a challenge.

Logan: I want to come back to this topic of policy tools and communication strategies that central banks can use to create that separation between market functioning and monetary policy. How important is it to create this separation, and what can we learn from the UK experience, in particular, and elsewhere? Brad, I'll start with you on that one.

Setser: I'll just compliment you on the lessons that you drew about how to differentiate asset purchases done for the purposes of financial stability, geared at the particular asset that is under stress, time-limited to the period when the market is malfunctioning, with an exit automatic when the market malfunctioning ends. I thought those were quite useful observations. I guess I've been surprised. I don't follow the Bank of England that closely, but I've been surprised how well it worked. We don't now discuss dysfunctioning in the gilt market, and the Bank of England is once again reducing its balance sheet. It was a brief period of market malfunctioning, not an inability to unwind some of the purchases, and the Bank of England's policy choices during that period maintained the optionality to return to balance sheet reduction. Probably there are lessons to be learned in a positive sense from that episode.

Going back to some of the things Kathy said, when you're in a period of market stress, the volume of your asset purchases actually matters. The willingness of the Fed to buy without limit when there's a panic can matter. It mattered. I don't think we were buying without limit, but maybe we should have been buying more and without limit in 2008, when the agency market stress was so extreme. The willingness to really be aggressive, in my view, in the COVID crisis, was critical. It is quite possible to differentiate more cleanly between interventions done on a massive scale during periods of market stress, and from using the same tool as a monetary policy mechanism. We perhaps haven't always done that as cleanly as we could have, but conceptually that's the way to go.

Logan: How important do you think it was that the Bank of England had this separation between the Financial Stability Committee and the Monetary Policy Committee, and the communications? Do you think that was a key part of it?

Setser: To be honest, you've thought more about that than I have, so I'm going to rely on your answer earlier.

Logan: Then I'll politely turn to Viral for the next comment. [laughter]

Acharya: To me, it seems very crucial that they had different committees, because the clarity of communication comes from the clarity of governance process within the central bank. It goes back to this earlier discussion of, should you be using two tools at the same time that are going in opposite directions? If you have different committees and they have reached, through a set of data and analysis, the conclusion that for financial stability purposes, I need to do a certain lender of last resort, you can communicate that well. That's good, in my view, because it gives the market a certain level of clarity that this is not blending in financial stability, full arm into monetary policy. I like the structure that Bank of England has. I wonder if the Fed would consider having something like that going forward. Just one question on that, Lorie, though, which is: the governor sits on both the committees, right, at Bank of England?

Logan: I'm sorry?

Acharya: The governor of Bank of England is on both committees?

Logan: Yes.

Acharya: Yes. That's the only place where I see confusion possibly creeping in, because there's, of course, no "Chinese wall" inside the governor's mind. Maybe it makes sense to have the governor on both of those.

Logan: Any other comments on tools, and separation of using the tools for different purposes, anyone wants to make?

Carpenter: I might be just slightly contrarian here. I appreciate Viral's point that it gets more difficult if you're using more tools than is necessary, but part of my point about talking about the reverse repo facility is that they're already using more than one tool just for raising interest rates. There's a risk of believing that interest rate increases or decreases are the same old tool that we've used forever, and we understand it well enough and so we should rely on that instead of anything else. That's taking more credit for the field's collective understanding that is actually there.

Similarly, when it comes to these tools, just for stability. If we look in the US, some of the lending to banks that was about separating liquidity for financial stability purposes from the overall stance of monetary policy, conceptually that sounds great and that's exactly what one would like to do if you could. On the other hand, there's a version of the world where being able to have those facilities there and provide that liquidity means that the transmission of higher funding costs to banks gets muted, because they've got this alternate source of liquidity, especially if the market is pricing in more rate cuts than the Committee anticipates it doing.

That means that the 1-year OIS might actually be lower than you think is the right penalty rate for banks to lend. As a result, you could be decreasing the monetary policy transmission of the rate hikes, because you believe there's this separation of financial stability tools from monetary policy tools. The Bank of England, the benefit of having the two different committees is it forces people to think carefully and distinctly. Just doing it very carefully and communicating much more carefully than central banks are used to doing it is a way around it, it's just hard.

Setser: Seth, if I understand you correctly, you think—I'm just being provocative—that banks should move to the term funding facility because the OIS funding rate is going to become attractive relative to the discount window?

Carpenter: Just to be clear, since I work for a regulated institution, I did not advocate for any particular behavior. I just said that it is possible that you're allowing the transmission that's fully intended of higher funding costs for credit intermediaries to be muted, by giving them access to another facility that could actually prove to be cheaper than you really want it to be.

Acharya: I'll just throw in one point, Lorie, on this, which is: the other thing that was interesting about the Bank of England intervention at the time of the gilt crisis was that it was a very short-lived commitment to support the market. I have a feeling that that was actually a bigger communication clarity that they provided, that this is not permanent liquidity injection, this is strictly for lender of last resort or stabilization of a systemically important market. That clarity of exit, in a finite time period, also probably helped with not confusing it with a monetary policy objective.

Setser: To be fair, changes in government and government policy probably made it a lot easier for the Bank of England to live up to its commitment to only intervene for a short period of time.

Logan: I wanted to come back to Seth's comment a little bit, because you were talking about: how do you think about these calibrations of these different tools? Because there's a couple of questions in here about calibration, so let me start with one of them, which is about calibration: How many rate hikes is the intensified credit tightening worth? It's been a popular question, so I don't know if anyone wants to step in to share your perspective. How do you quantify it?

Setser: Put your finger in the wind.

Carpenter: I was going to say, "yes." [laughter] Very, very difficult to know. I don't think it's falsifiable to go along with Brad, and I sort of feel like Chair Powell's one or two rate hikes ... the challenge is, suppose you knew for sure that it was exactly two. Would that tell you then that you have done enough, or done too little? The answer to that question is, just stipulate that you knew with 100 percent certainty that it was worth two 25-basis point rate hikes. You still don't know if you've done enough or not enough hiking.

Logan: Another calibration question: Do you think the fact that the FDIC is rapidly selling the underwater mortgages they inherited and widening mortgage spreads, exacerbates risk in the banking system, and does it substitute for a Fed rate hike or two? Another calibration question with different dynamics taking place right now.

Setser: I will answer that, because I don't think its magnitude is at all comparable to a rate hike, but I also think it is silly. I don't understand why the FDIC can't use its balance sheet to hold on to government-backed assets for a period of time. I don't see why it feels compelled to clean up its balance sheet, so to speak. The FDIC should think more creatively about the length of time it holds assets taken over by failed banks. I don't think there's a good policy reason, despite what the FDIC says, for selling now.

Jones: Yes, I would agree on both points. Its impact is probably not that great. On the other hand, why exacerbate a problem if you don't have to? It's probably on the margin, but it seems to be on the wrong side of the margin for what's going on.

Logan: I'm going to finish up with one last question before we conclude the panel, which is: What do you think policymakers need to see to feel more confident banking stress has dissipated, or have we passed the point and we're now under a constant shadow of inevitable credit tightening coming?

Acharya: There are two or three things that would be important. One is banks don't do anything on their capital side. In that case, we would have to see the runs and the deposit outflows stop. I see that happening only if either inflation comes down so sharply that the Fed pauses and maybe even cuts rates the way investors are expecting, or if we do something on the RRP rates, which then reduces the ability of money market funds to attract the deposits away.

I don't see either of those being the likely outcome. I've been pushing very strongly that we should do an asset quality review of the regional banking system. We should mark capital honestly, where capital can be raised, because a bunch of them are in public markets. Let them go and raise capital where it can be raised, start orchestrating more mergers ahead of time rather than waiting for something like the First Republic situation.

That would restore confidence in the system, and once the deposits are on a well-capitalized banking system, they will stop running. Then there will be some outflows related to the money market fund rates, but they'll stop running on the banks. To me, it's about capital in the end, and I would like to see that getting boosted. It's not going to get boosted through earnings. You have to engineer a balance sheet restructuring on the right-hand side.

Logan: Any final comments from the panelists?

Setser: I'll make an observation. One sign that we've maybe passed the worst would be a bank that gets purchased with a positive equity value rather than banks that are sold at a loss with the FDIC having to use its authority and having to go into resolution. That would suggest that some of the bigger, stronger, well-capitalized banks see positive franchise value in the deposit base. That'd be healthy. Hopefully we're getting there.

Logan: Well, that seems like a great place to conclude. Please join me in thanking our panelists for the conversation this afternoon.

Paula Tkac: I would like to everyone who made these days possible for us. As I said, sitting through these presentations I began to think about how important it probably is for us to balance three different perspectives. One is to keep an eye on the past. Here I'm thinking about interest rate risk, potential for bank runs, moral hazard, too big to fail, the importance of capital. These are old problems that we really haven't solved, and so it makes sense for us to keep an eye out for those, and make sure that we don't get sanguine about the risks that may not have been here for a while but could pop up at any moment.

The second is to look and see through the change of clothes that is referred to in the conference theme. Everything old is new again, but it can be dressed up differently. When facing a policy challenge, how do we look to see what is fundamental? What can we use—our past experience, the lessons learned from history, our expertise—to help unpack and understand what a good policy response is, or how to make decisions when faced with these new opportunities or challenges?

In terms of the conference, I'm thinking about Kristin Forbes and her comments yesterday about how we can go back to some tenants of monetary policy that really are in our history, and not necessarily that long ago, but they can guide us along thinking about monetary policy challenges. Today, we heard about the importance of leverage, and this is something we all saw in the great financial crisis. It's still here, but now we're thinking about it differently when it comes to these nonbank financial institutions and how they might pose systemic risks.

In terms of ferreting out that fundamental part, we're then left with what are the details that are important for us to think about, that are actually different this time? Or the new clothes that haven't been worn? Again, in monetary policy, I'm thinking, "Well, okay, we're fighting inflation, but now we're doing it with a large balance sheet, balance sheet tools, an overnight reverse repo facility, with both opportunities and risks that we have not faced before." Those things are important, and we're still, obviously, sorting out and grappling with how to involve them in policy in the best way.

In terms of financial stability, I was really struck today by the expansion of what we used to call "shadow banks" into this new "nonbank financial institutions." Again, coming out of the great financial crisis, a lot of time, effort, and reform was put in the arena of money market mutual funds. Those are very much like banks in many ways, with their deposits being very liquid and having the systemic component to them as a group, if not as an individual fund.

Now we're talking about private equity, hedge funds, Business Development Corporations. How do we pull apart what is specific about these different kinds of organizational structures and the risks they pose, and then answer the questions if and how we should regulate them, macro prudentially or micro prudentially, in order to, again, balance both the opportunities and the credit and the diversification that they provide, with the risks that they might bear.

Finally, we have to keep an eye on imagining the future. It's really good practice for us to be thinking about scenarios that may or may not occur. So today, the web3 session gave us a whole lot to chew on in that domain. I loved the question of, "Well, how are we going to know who's right?" The answer is, of course, well, time will tell.

It's quite important for us to be thinking and using our imaginations about things that might happen, so that we can start to, again, sort out opportunities and risks, make good decisions, but be ready to adapt and to make good policy when the time comes, when we see how things play out, and we better understand where markets go and what policymakers can do.

Finally, this last panel was great. There is no "optimal policy free lunch," as you guys have said. There's just so much to talk about in terms of balancing risks. We do it in the briefing room every day, thinking about a dynamic risk management framework and really acknowledging that there probably are going to be surprises, whether it's data-driven or market-driven, that we're going to have to deal with in the moment.

Keeping all these things together—eye on the past, looking through the clothes, but also thinking with our imaginations about the future—is something that I'm going to take away from this conference.

Now we'll get to the important part. On behalf of all my colleagues at the Atlanta Fed, I want to thank all of you for everything that you've done to make these a great two days. We're not done yet, right? There's a great dinner program. I want to thank the participants and the panelists for all their presentations, panel discussions, questions, conversations, and all of you for engaging both with the folks on the program and with each other, for a great couple of days.

I also would like to ask you all to thank the folks that have made this possible from the Atlanta Fed. Join me in giving them a little bit of recognition and a hearty, both congratulations for a job well done, but also our appreciation and thanks. I'm going to ask a few people to stand up so that you know who they are.

First is the planning committee. These are the folks that put together the sessions that you have enjoyed here. This is their brainchild. I'm going to ask Larry Wall, Mark Jensen, Kris Gerardi, Camelia Minoiu, and Brian Robertson to stand up. Many of you may have met them already.

I'd also like to thank Sandra Ghizoni and Sherilyn Narker. They are the organizers that herd the cats, meaning all of us, as we try to pull this off. I don't know if they are in the room. There they are, there at the back. Awesome.

Then here's the cast of thousands. You may or may not have seen these folks, but they have gone above and beyond to bring us a conference that has exceeded any expectations. I'm going to call out several teams here from the Atlanta Fed and our Jacksonville branch. The creative services, production, and design teams. These are the guys doing the AV, the photographers who are making us look great up there, our social media team, our events management team who takes care of logistics, all the wonderful food that we've been enjoying, our law enforcement officers who are keeping everybody safe, especially the VIPs, the folks doing transportation and helping us get around, and then the teams supporting us remotely: we've got web teams, outreach teams, and our publishing teams. Please join me in thanking all of them as well.

Okay, now the housekeeping. Here are the things that I've been told to make sure you know: videos, papers, and presentations will be available on our website, atlantafed.org/fmc. You can have access to everything that you've seen here. We would love your feedback, and so please take the survey in the app or by email. We're already going to start planning for next year, probably starting in July, so we would love to hear what we can do better, what you'd be interested in hearing about, and how we can make this conference as valuable as possible.

Finally, save the date. Our dates next year are June 2 to June 5, not Mother's Day. I've been told that our next contract also will not be on Mother's Day, so we'll make it easier. Yay for that. You'll be getting a "save the date," probably in the fall, for that, but keep it on your calendars. I would like all of you to come join all of us at dinner. I know you're all looking forward to the program. It promises to be fun. With that, we are closed out for the day. Thank you so much for coming.