Oct 17, 2022

Bill Nelson on How Bank Examiner Preferences are Obstructing Monetary Policy

By pushing heavily for reserve holdings over other liquid assets, bank examiners are creating unnecessary obstructions for monetary policy operations.
David Beckworth Senior Research Fellow , Bill Nelson

Hosted by David Beckworth of the Mercatus Center, Macro Musings is a new podcast which pulls back the curtain on the important macroeconomic issues of the past, present, and future.

Bill Nelson is the chief economist and executive vice president at the Bank Policy Institute. He previously worked as a deputy director of the Division of Monetary Affairs at the Federal Reserve Board, where his responsibilities included monetary policy analysis, discount window policy analysis, and financial institution supervision. Bill has also worked closely with the BIS working groups on the design of liquidity regulations and is a returning guest of the podcast. He rejoins David on Macro Musings to talk about his new note that is titled, *Bank Examiner Preferences are Obstructing Monetary Policy*. David and Bill also discuss how the Fed’s forward guidance is affecting recent market turmoil, how to change the mindset of bank examiners and the public, why the Fed should look into establishing a committed liquidity facility, and more.

Read the full episode transcript:

Note: While transcripts are lightly edited, they are not rigorously proofed for accuracy. If you notice an error, please reach out to [email protected].

David Beckworth: Bill, welcome back to the show.

Bill Nelson: Thanks for having me, David. It's always a pleasure.

Beckworth: Well, it's great to have you on the show. As listeners will know, a few weeks ago we had a paper of yours on the show, your co-author Andy Levin came on and discussed it and I think listeners will recall that, and we'll provide a link to it if they haven't. Great discussion on the Fed's balance sheet, potential losses, and something we may come back to later in the show. We want to talk about your really fascinating new note on how bank examiners may be making life difficult for the FOMC. It's an important point and you have firsthand experience as a former Fed official. Interestingly, there were some people who were really surprised to see this, but it's an important fact, I think, because of that to get it out there. Now, before we get into this really interesting note though, Bill, I want to step back because today you had another note come out. I want to touch on that real briefly. Two things I want to mention from that note, the first one, you spoke to the Fed’s forward guidance as it relates to all the financial market turmoil we see now. Can you speak to that briefly?

Forward Guidance, Interest Rates, and Market Turmoil

Nelson: Sure. Let me quickly mention first, David, that you're referring today to one of the emails that I send out once in a while on monetary policy. Those are a bit more of my own and not so much BPI publications, but nevertheless, and I'd like to encourage anyone who would like to be added to that list, they're free. Just shoot me an email, I'd be happy to add you. But thank you for reading it, David. I appreciate it. Yes, one of the things that I observed in the email this morning was that even though the FOMC participants’ recent catch phrase that they're all saying is, "We're going to get policy to a restrictive stance and we are going to leave it in that stance for some time,” that doesn't mean that the committee members are insensitive to the financial market turmoil that goes on right now. Jay Powell often mentions that the FOMC's objective is to influence financial market conditions. Financial market conditions have tightened a lot with the reduction in equity prices and the increased value of the dollar and the widening of credit spreads. That means that whatever your personal sausage machine for formulating your own guess as to what monetary policy should be, you should be including the turmoil in the financial market right now into that sausage machine.

Beckworth: Yeah, and I saw Lael Brainard, the vice chair of the Board of Governors, she recently had some comments along those lines that they're aware, they're cognizant of what's happening around the world. What was interesting to me is she said very similar things, I believe, back in 2015 when the dollar was previously appreciating quite a bit. Very similar periods, we have the Fed, at least back then they were talking up rate hikes, today they are actually doing them. But the dollar appreciated dramatically 20% this time, I think a little over 20% last time. The Fed did have to step back from the original plans in 2015, 2016. Maybe we're seeing a little bit of that now.

Beckworth: It's interesting to see the Fed be aware of the role it has in the global economy. This, Bill, speaks to something that really fascinates me, we've discussed in the show a lot, and that is this role is the Fed as the monetary superpower of sorts. It sets monetary policy for the US economy. That's its mandate, but it inevitably affects the world economy given the reserve currency of the dollar and the linkages and just the dominance of the dollar.

Nelson: That's right, that's right. I do want to emphasize, David, that even though I'm saying that the committee's language about keeping rates tight for some time should be interpreted in light of the financial market turmoil, I'm in no way questioning their commitment to fight inflation. The financial conditions necessary… the Fed's objective is to slow the growth of the US economy, and that objective can now be met with a slightly lower path for the federal funds rate, given these other developments. Sorry that wasn't very responsive to the point that you just made, but it's certainly true that the Fed's actions have consequences throughout the world. As we're all reading, the incredible strength of the dollar is putting a lot of strains on the rest of the world, especially developing countries.

I do want to emphasize...that even though I'm saying that the committee's language about keeping rates tight for some time should be interpreted in light of the financial market turmoil, I'm in no way questioning their commitment to fight inflation.

Share this

Beckworth: Yes. This brings to mind two people, Helene Rey's, *Global Financial Cycle,* where she talks about the key factor in global financial movements and crises is US monetary policy and the dollar. But the other person that comes to mind is John Connally's famous remark in 1971, "The dollar is our currency, but your problem." I think that's very true today and it's just so fascinating to see this. To me, at least, it speaks to the underappreciated hold that the dollar system has on the global economy. When people talk about, "Oh, the dollar's days are numbered, or the US is going to go into a period of fiscal dominance." In fact, right now there's a lot of conversations about fiscal dominance. I do think we are going to see some losses as your paper with Andy Levin showed. But I believe, and maybe you can respond to this, that rates will eventually go back down and we will be in a world very similar to what we were pre-pandemic, low interest rates, low financing costs over the longer run, back to that world, and we won't have fiscal dominance. But I would be interested to hear what you think about that.

Nelson: I agree with you. I guess I still think that rates could have quite a ways to go up, just because real interest rates are still currently negative. The federal fund rate minus the underlying rate of inflation is still negative, and they need to get that positive, and positive by some significant margin. It's just not clear that markets are pricing that in. But I do have every confidence that the Fed will get inflation rates back down to their target of 2%, and that interest rates will eventually move back down to the configuration that they were in before the pandemic.

Beckworth: Yeah, if you looked at the Summary of Economic Projections that came out this last meeting in September, they have the, I believe, it was longer run federal funds rate at 2.5%. Of course, they have their long run inflation forecast at 2%, which means the real federal funds rate over the long run, the equilibrium, the neutral rate, is half a percent in real terms. That's very similar to the pre-pandemic environment. Okay, let's move on from that because we do want to get to your paper, but one last thing in your note this morning is, you had a section titled Here Comes The Losses. Speak to that briefly too.

I still think that rates could have quite a ways to go up, just because real interest rates are still currently negative...and they need to get that positive, and positive by some significant margin. It's just not clear that markets are pricing that in. But I do have every confidence that the Fed will get inflation rates back down to their target of 2%, and that interest rates will eventually move back down to the configuration that they were in before the pandemic.

Share this

Updates on the Fed’s Balance Sheet

Nelson: That's right. Interestingly, the Fed's weekly balance sheet… I only just learned this, which is a bit humbling because, as you mentioned, I was at the Fed for a couple of decades in the division that creates the Fed's weekly balance sheet. I've studied it all the time as a Fed watcher. But I only recently learned because Lou Crandall at Wrightson ICAP, who writes the Money Market Observer, pointed it out to me. There's a line item that shows the money that they owe to Treasury. So when the Fed makes money, they pay it over to Treasury. Each week they accumulate some money and then a week later they make the payment.

Nelson: There's usually a pot of money sitting at the Fed reflecting the money that they had earned that week. You can see that on table six of the H.4.1 statistical release. Over the last few weeks, that number has turned negative and it's probably now accumulating negative losses, because when the Fed makes negative operating income, negative profits, their plan is to just drop remittances to Treasury to zero. All of that money that they're now not making is accumulating as a larger and larger negative item in that H.4.1 item. It's currently negative 2.1 billion, and it went down by negative 1.5 billion over the week ending in September 28th.

Nelson: That suggests, as we've all anticipated, that the Fed's interest expense has risen above its interest income. Now, other things could be going on. They probably made losses on their holding of foreign currency and the TIPS inflation adjustment complicates things. But nevertheless, we appear to be now at the point where the Fed is beginning to make actual operating losses. That number will probably move back up in the next H.4.1 to zero. I'm not sure, because the Fed's plan, as described in its financial statements, is to record an asset on their balance sheet rather than recording negative equity, which is the money that they plan to withhold to fill back any hole made in their balance sheet by losses before they will start remitting money to Treasury again. It's called a deferred asset. If they make those bookings once a quarter, then since September 28th, since the next H.4.1 will be after the end of the quarter, we'll see that number come back down. Then, it will all be revealed in early November when the Fed publishes its Q3 balance sheet.

Beckworth: Just to be clear and make sure I understand this, this figure, and it's also available via FRED, and we'll provide a link to it on the show note webpage. It shows a negative number now. Help me understand where that negative number comes from.

Nelson: Each week the Fed usually earns a positive amount from all of the interest payments on all the securities that they own. It makes a payment of interest expense on its interest bearing liabilities, which are now a much, much bigger part of their balance sheet than it used to be. That accumulated income minus, I assume, some money held back to meet operating expenses is recorded here as money that they will be giving to Treasury. But there's evidently a weak delay before when they decide how much they're going to pay Treasury and they pay Treasury so that each week what shows up in there is the money that they accumulated but haven't yet paid over to Treasury. That raises the question of, "Well, what are they going to do about negative operating income?"

Nelson: We're all still trying to figure this out on the fly, but what appears to be the case is that when they make losses, that enters, initially, enters into this item as a negative amount that they owe to Treasury. But it's not as if they're going to get money back from… as say happens in some other countries, where the Treasury has indemnified the central bank against losses. Instead, the Fed's going to book itself this magic asset which makes the losses go away by promising to withhold money in the future before it starts giving money over. That should make this item go to zero, as of the next H.4.1. But we're all trying to figure this out because this is new territory.

Beckworth: It is, this is an exciting time to be someone following the Fed's balance sheet, for sure. Something new is always something exciting. One last clarifying question, does the Fed determine the amount it sends to Treasury? There's no law that says you have to send X percent, the Fed has the discretion?

Nelson: No, I think that's not correct. There is a law that limits the amount of the equity that the Fed can maintain, to a very small number, very close to zero. Any money that it's made and doesn't need to cover its expenses, it hands over to Treasury.

Beckworth: Okay, alright. We'll be following this closely and, again, we'll have the link in the show notes. Alright, let's go to your exciting note on bank examiners. Again, the title is, *Bank Examiner Preferences are Obstructing Monetary Policy.* The image that comes to my mind, Bill, is I have this image of the boardroom at the Eccles building, the FOMC has just voted on a decision. Then, there's all these examiners out there just really messing up the plans. They're thwarting the desires of the FOMC. You gave us two key ways that examiners are doing this. Maybe just off the bat, how are examiners getting in the way of monetary policy?

Bank Examiners as an Obstruction to Monetary Policy

Nelson: Even before doing that, I do want to make it clear that everyone here is operating with good intent. I don't want to cast bank examiners as somehow, deliberately, as obstructionist. It's just a matter of entrenched practices and experience that ends up having that effect. Basically, surely everyone listening to the show knows that the Federal Reserve is engaged in quantitative tightening. That means that they're shrinking their balance sheet. When some of their assets mature, they're not reinvesting the proceeds, and this is causing their balance sheet to shrink, their assets to shrink. Now, when the assets shrink, their liabilities also have to shrink. In particular, reserve balances shrink.

Nelson: Those are the deposits of commercial banks and credit unions and thrifts at the Federal Reserve. Just as your deposits at your bank is a funding source for your bank, those deposits by those banks at the Fed, reserve balances, are a funding source, a liability of the Federal Reserve. As assets shrink, those reserve balances also decline. Now, a complication now that all people who teach students about monetary policy should be unhappy about is that the overnight RRP facility makes this story a bit more complicated. That is a facility where the Fed is actually borrowing from money funds and GSEs at a fixed rate. But let me set that aside for the moment.

Nelson: There's a lot of interesting things to talk about in that area as well. But setting that aside, when the Fed's asset shrink, these reserve balances shrink. Now, the Fed is currently conducting monetary policy in using what it calls its abundant reserve framework or a floor system. That means that it's oversupplying reserve balances, it's maintaining a balance sheet that's large enough that the corresponding reserve balances are in excess supply, pushing money market rates basically down to the interest rate that the Fed pays on those reserve balances. Because if you're a bank, you certainly have a lot of liquidity, but you're not going to lend it out to anybody else who needs the liquidity at a rate that's below the interest rate that you earn from the Fed.

Nelson: Now, none of this is exactly true, but I'm painting the picture, I think, in a way to provide the intuition. The Fed is currently shrinking. It can't shrink further than the quantity of reserve balances that banks are demanding right now to meet their structural needs, to meet their liquidity needs, their anticipated outflows, their liquidity requirements, whether they're standardized requirements or examiner preferences. All of those things determine banks’ reserve balance demands. The Fed can't get any smaller. It can't let its assets shrink any longer than any further below the amount necessary to create the reserve balances that banks demand. If they did, interest rates would go up above the rate that the Fed is targeting, because banks would start competing for interest rates and that would push rates up. With that lengthy intro, bank examiners are doing two things that are unnecessarily increasing bank's demand for reserve balances, and therefore shortening the runway the Fed has to continue QT. The first thing that they're doing is that, even though bank regulations recognize that a number of assets can be a source of liquidity, examiners have expressed a clear preference for reserve balances to banks, understandably, because reserve balances provide instant liquidity.

Nelson: But that's a bit overkill because the banks could hold treasuries or agency MBS, those can be converted to liquidity immediately at the Fed's new standing repo facility, for example, or at the discount window or in the market by the next day for many of the largest banks. But still, that preference exists and banks respond to examiner preferences. It's just expensive to make your examiner unhappy. The second thing that they're doing, that examiners are doing, is that they continue to disapprove of banks making use of either the discount window or the standing repo facility to meet very temporary overnight liquidity needs, or even potentially using the collateral that the banks have at the Fed to incur a daylight overdraft.

Nelson: What's important to recognize is that all of the Fed's policies, monetary policy, oriented lending policies towards the discount window, towards the standing repo facility and towards daylight credit, make it clear that they want banks to be willing to use these things. But the examiner raised eyebrow, as it were, or significantly more than a raised eyebrow, is preventing banks from being willing to do so. That being the case, banks have to hold a tremendous amount of cash on hand to be sure that they will never have to end up doing so. Banks are hit with needs for funds or delayed money coming in. They can't know for certain what their balances are going to be. If they want to be certain that they'll never have to borrow from the discount window or use the standing repo facility, or even in current daylight overdraft, then they have to hold a huge quantity of reserve balances in their account all the time to make sure that that won't happen. Both of those things are pushing up the demand for reserve balances and maybe if you want, I can talk about how that demand for reserve balances has evolved over time and why it might have done so.

All of the Fed's policies, monetary policy, oriented lending policies towards the discount window, towards the standing repo facility and towards daylight credit, make it clear that they want banks to be willing to use these things. But the examiner raised eyebrow, as it were...is preventing banks from being willing to do so.

Share this

Beckworth: Yeah, just to summarize real briefly, so the two ways that examiners are obstructing the implementation of monetary policy is, one, they're pushing heavily for reserves over other liquid assets, and secondly, this contingency planning for liquidity needs. They don't want the banks to tap into the discount window, standing repo facility, or this daylight credit borrowing from the Fed. They're making it really hard for banks. You give some interesting examples in this note, I'll encourage the listeners to go check it out, where you talk to treasurers and the treasurers say, "Hey, we got to talk to our examiners before we even think about tapping into these sources." It really creates pressure upon them. You know, Bill, I posted your note on Twitter and I got some response back, which was interesting. Many people were just surprised to see this. They simply weren't aware that this is happening. But in your work at the Fed, did you see it as well since you were in this area?

Nelson: Yeah, I did. I worked in the Division of Monetary Affairs, which is the monetary policy division. As part of that division, part of its responsibility is overseeing the discount window. Starting, I think around 1999 or 2000, I became the staff person responsible from the board side of discount window policy. In that role I was very involved in the 2003 revamp of the discount window that took place. That revamp was importantly intended to make the discount window a more effective tool for monetary policy. Also, for financial stability policy, but importantly for monetary policy. Now, what we did was we raised the discount rate. It had been below market rates since the mid '60s. That's an interesting story in and of itself, that maybe we'll discuss sometime. But in 2003, the Fed re-raised significantly above market rates, but also made it a no questions asked facility. We told banks, "Whatever you want to use it for, you can use it." We published that in Regulation A, the regulation that governs the discount window, and that's still there. We also encouraged banks to include the discount window into their liquidity planning.

Nelson: This was all done because we hoped that it would reduce the stigma, the significant stigma that exists about borrowing. If there's stigma about borrowing, then the discount window isn't an effective cap on money market rates. In theory, it's supposed to be a cap on money market rates because why borrow in the market for more than you can borrow from the Fed at the discount window? Well, if you don't want to borrow from the Fed in the discount window, you're willing to pay quite a bit more. What we used to do as we were monitoring the effectiveness of those changes is we would keep an eye on how banks were bidding for funds in the federal funds rate.

Nelson: If a bank appeared to be consistently paying more than the discount rate, I would make an excuse to talk to them. Maybe go out and visit them, talk to them about the discount window and their preferences. That happened with one large bank, probably in 2004. I reached out to their funding officer and asked them how they felt about the discount window. They said they weren't really interested in borrowing from the discount window because they didn't think they were allowed to. I said, "No, no, we've changed our policy, and it's now perfectly fine." They said, "Oh, great. Okay, thanks." They called me back up and they said, "We talked to our examiners, an OCC examiner, and they said they didn't think we should be using the discount window." I said, "Huh, okay, thanks." Then, I met with the Fed bank supervision side and the OCC and the credit union supervisors and the thrift supervisors and everyone collectively wrote a letter, what's called an SR letter, which is instructions to examiners. This SR letter said, "It's okay to use the discount window. It's a good source of liquidity contingency funding, and banks should be including it in their contingency funding." Now, the credit unions went further and said, "You really need to include it unless you have some kind of other alternative." But nevertheless, here's this official letter out there.

Nelson: I went back to the bank and I said, "Okay, problem solved. There's now an SR letter that makes it clear that you should feel free to use the discount window and that the examiners are okay with that." They said, "Great." Then, they called me back a week later and they said, "We talked to our examiner again and he's still not comfortable with us using the discount window." Just an illustration of the fact that everyone has their own amount of autonomy and they all get used to things. In many ways that examiner might have been prescient because as it happened during the global financial crisis, the banks that did borrow, many of them, they were encouraged to borrow during the crisis. The Fed created a special discount window facility to try to get banks to borrow more of the term auction facility, and all of those loans were repaid. All of the Fed's crisis loans in the global financial crisis were repaid on time with interest, often above market interest. But nevertheless, as you know, banks were castigated after the crisis, called before Congress for borrowing, it was called a bailout. As we discussed in those interviews with treasurers, banks are now very unwilling to use the discount window.

Beckworth: So many questions, Bill, from those stories. Very interesting. You yourself saw, firsthand, that this happened. It's nothing new, but it's definitely on a more pronounced level. That's what I got out of your note. There was a big change after the financial crisis in 2008. But let me step back before we talk about what caused this change in thinking and ask, who could lead change in the current thinking? I know this is near the end of your paper, maybe we're getting ahead of ourselves here, but if you wanted the government, you wanted the examiners to change their perspective… Also, you mentioned Congress, because you just said Congress went after the banks, the public, but who do we start with? Is it Michael Barr? Is it the Vice Chair of Supervision? How do you get this done? There's a number of agencies, as you mentioned, that regulate banks. Where would you even start on this journey of changing the mindset of examiners and the public?

Changing the Mindset of Bank Examiners and the Public

Nelson: I think there's a few things that need to be done that could really help a lot. One is just public communication by the leadership of the Federal Reserve that borrowing is normal. Borrowing is a business decision on the part of banks. That's especially true for the standing repo facility that they've just created, that they intend to look differently than the discount window and be perceived as a normal business decision. Leadership of the Fed needs to explain that to the public and they need to explain it to Congress, and they need to explain it to their supervisors. The second step actually has to do with the way that the Fed is conducting QT.

Nelson: Now, the Fed's current plan is to keep shrinking its balance sheet until it gets to about $300 billion above the structural demand for reserves that banks have. Now, the Fed doesn't know what that structural demand is, so it's actually very difficult to stop $300 billion above that level. They're just going to be looking very carefully for signs of stress, of any kind of indications of liquidity in the money markets to do so. But unfortunately, that plan means that reserves will remain super abundant. The interest rate that the Fed pays on reserves will remain above money market rates, and borrowing will remain extraordinarily rare. Daylight credit will remain extraordinarily rare.

Nelson: All of these things will maintain the stigma associated with using any kind of credit from the Fed, and will provide no financial market incentive for banks to economize on their holdings of reserve balances and therefore get smaller. Ironically, this policy which is designed in part to minimize bank's need to borrow from the Fed actually leaves the banks relying on the Fed as their source of liquidity in huge dollar amounts every day. Economically, there's not very much difference between a reserve balance and borrowing. They're both payments by the Federal Reserve. But anyway, the Bank of England is conducting their QT differently.

Ironically, this policy which is designed in part to minimize bank's need to borrow from the Fed actually leaves the banks relying on the Fed as their source of liquidity in huge dollar amounts every day.

Share this

Nelson: Their plan is to shrink until their reserve balances get low enough that they get a bit below structural demand, and banks have to come in and borrow regularly in order to meet that demand. The Bank of England has created a new facility that it's priced at the rate that it pays for its deposits and it's instructed its examiners that use of that facility should be seen as ordinary, as a business decision. By creating a situation where money market rates will probably be a bit above the interest rate that the Bank of England is paying on its deposits, a situation where borrowing will be more frequent, it's actually set in place all of the necessary ingredients for the Bank of England's balance sheet to get even smaller over time. Those are some very important steps that the Fed could take that I think would help allow the Fed to get, in the end, significantly smaller than it will achieve if it's following its current course.

Beckworth: That is very interesting. The ample reserve regime or the floor system, as we like to call it, itself, is a reason why it's hard to change the minds of examiners and the public in general, because it's ample reserves. On page eight, you note, "Ironically, by designing a regime where the Fed would never have to lend in an emergency, they created a regime in which banks rely on the Fed as their primary source of liquidity every day." Which is, you mentioned that earlier. You have this operating system which creates the incentives for examiners to keep doing what they're doing. You're saying even if you did communicate from above, from the Fed, from Fed officials, you're still up against the wall of this ample reserve system. You need both of those pieces moving before maybe you begin to change minds. Is that right?

Nelson: That's right. You can really see the effect of the growing, increasing reliance on the Federal Reserve as the source of liquidity over time. Before the global financial crisis, reserve balances were measured in the tens of billions of dollars. Banks weren't using reserve balances to meet their liquidity needs. Reserve balances paid zero, and examiners didn't expect them to. One of the first things that I did when I joined the Federal Reserve was to go to examiner school for a month and I learned about how to evaluate a bank's situation, including their liquidity situation. What I was taught was the first thing you look for was for a bank to have healthy access to market sources of funding. Nowhere on the list was holding a lot of reserve balances, that just wasn't contemplated. That situation changed for a number of reasons. All of the QE made reserve balances more abundant.

Nelson: The Fed got the authority to pay interest on reserves, and the post-crisis regulatory regime was designed in this environment, where suddenly it was seen as appropriate and desirable for banks to hold reserve balances. Over time, if you look back to, say April 2008, when the Fed did a study on how to use its new power to pay interest on reserves to conduct monetary policy, it came up with an estimate of, "Well, what quantity of reserves do we need to provide in order to move rates down to the floor? What quantity is needed to be abundant?" The number that we came up with was $35 billion. Just to remind the readers, reserve balances are currently $3 trillion.

Nelson: In March 2016, the Fed revised that estimate, reserve balances had now been plentiful for a while and they raised it to $100 billion. In March 2018, so getting close, their estimate was $600 billion. Then, in September 2019, after the kerfuffle in repo markets, the number was raised to $1.3 trillion. Then, finally, if you look at the most recent New York Fed’s incredibly informative forecast of the balance sheet that they publish each spring, the current number is $2.3 trillion. That's the amount needed to be abundant, up from $35 billion before the global financial crisis.

Beckworth: What's neat about the Bank of England is that it's almost retraining, forming new habits by forcing banks to go to the Bank of England itself. It's helping the examiners maybe learn new ways of thinking as well as the banks learn new ways of thinking. You can't just communicate ideas. It would be nice to have Jay Powell, Michael Barr say, "Hey, view the discount window, view the standing repo facility as normal places of business for banks to do their contingency liquidity needs from." It's great to have that, but you need to have the banks actually go and use them, actually put them in a place where they're using them as a normal course of business. What you're suggesting is that may not happen under the current setup. We look to the Bank of England as an example of what they're doing. I want to have one last question on what the Bank of England is doing in terms of the system. I've already mentioned this ample reserve or floor system. You and I many times have talked about a corridor system, which is what we had before and many places have had in other parts of the world. What would you call the Bank of England system? Is it any one of those two or is it different?

Nelson: It's both in some sense. But before explaining that, let me note that the United Kingdom has a much, much smaller problem with stigma than the United States does. It has had little signs of it, and they will tell you that they have signs of it, but they have nothing compared to what the Federal Reserve has. That really goes back to the founding of the Federal Reserve system, where the Federal Reserve has had a come hither, go away approach to the discount window that's embedded stigma into the system from the founding. Although it got much, much worse during the crisis.

Nelson: In the case of the United Kingdom, and maybe I'll give another example, the silliness in the United States that illustrates the Bank of England approach, but in any case, they have a much smaller stigma problem. What they're planning to do, actually, is to set the amount that they pay on their deposits equal to the amount that they pay on their loans. You could view that then as a floor system or a corridor system, because the two rates are equal. I think that they would describe themselves as still aiming for a floor system, but wishing to take actions that shrink the balance sheet to the fullest extent possible.

Beckworth: You provide some steps for the US to follow and they start with follow the Bank of England's example. Forego plans to keep reserves well above the level banks want. We've touched on that. Then, you've mentioned already that you need to instruct the examiners and educate the public and Congress that borrowing from the discount window is not a bailout and ditto for the standing repo facility. You also had a third innovation here, and I wanted to walk through this in some detail because it's probably new to listeners, it's relatively new to me myself. I know you've mentioned this in some of your earlier work and that's where I first saw it. But you want the assessment of bank's liquidity conditions to include collateralized lines of credit from the central bank as a source of liquidity. This could be part of their contingency liquidity planning. What is that? What is the collateralized line of credit from the central bank and what would it look like in the US?

Contrasts Between the Fed and the Bank of England

Nelson: I'd love to talk about that. This is something I've worked on for years and I will talk about it, but that's very much a reach goal. If you don't mind, before going there I'd like to give just another example of the contrast between the Fed approach and the Bank of England approach that seems achievable. I know that you've had David Andolfatto onto the show talking about the standing repo facility. The standing repo facility is a lending facility that the Fed created just this last July 21st, and what it is, is it's like the discount window, but it provides repos against Treasury securities and agency mortgage backed securities. Two virtually riskless securities that the Fed uses in open market operations, and the Fed lends banks overnight money against that collateral. It does so at a rate that's a bit above market rates. It created the facility for two main purposes, but basically, broadly, to try to create something that was free of the stigma that was associated with the discount window so that it could achieve two objectives.

Nelson: One objective, and this was very much with their eyes on the September 2019 repo market volatility. One objective was to provide an escape valve for pressures in repo markets so that repo rates wouldn't spike up above the standing repo facility, and so that wouldn't spill over to the federal funds market, the objective of the FOMC. The other reason was to make banks confident that they had a way to convert their Treasury or agency MBS into cash on demand if necessary so that banks would be willing to hold smaller quantities of reserve balances. Exactly the process that we've been talking about. This is really especially relevant for banks that aren't global, systemically important banks, banks that don't regularly interact in a high frequency way in the repo market or the money market, because for those banks, those regional banks, they don't operate in that market regularly, so they don't have those relationships. This is an important way for them to convert their OMOs cash, so the Fed created this facility. Now, many readers might be familiar with the liquidity coverage ratio or the net stable funding ratio, these liquidity requirements established in the post global financial regulatory reform, but more importantly, really, for bank's behavior, banks are also required to conduct monthly internal stress tests of their liquidity situation and report the results to their supervisors.

Nelson: Monthly, if they have $250 billion assets or above, quarterly if they have $100 billion assets or above. Those requirements are generally more binding than those standardized requirements that people are more familiar with. More important in determining how much liquidity banks feel that they have to hold. Now, those requirements require banks to conduct the tests at multiple horizons, including overnight. For a regional bank that say holds a lot of treasuries or a lot of agency MBS, extraordinarily liquid assets, they have to be able to demonstrate that they can convert those to cash immediately to meet the overnight horizon. Now, they normally wouldn't be, so they don't have the market connections to do that.

Nelson: But as far as we know, as far as I've been able to learn, the Federal Reserve is not allowing banks to say, "Well, we'll convert this to cash at your new standing repo facility." This is especially ironic because that's precisely what the facility was created for. As bank treasurers have explained to me, it doesn't really make sense for them to sign up to the facility because it's just not part of their business model and it's expensive to do so for them, not for the money that they have to pay the Fed, but other arrangements that they have to do. So, they won't, but if they can't even say, “all these agency MBS we hold are a source of liquidity fed for us because we will use this facility,” then the facility is even less valuable.

Nelson: Now, in order to pass their tests, what does that mean? That means that the banks, instead of holding treasuries or funding home mortgages through agency MBS, they really have to hold more reserve balances at the Fed. These two Fed programs are working in direct opposition to each other. Now, I raise all that in the reference to the Bank of England, because at the Bank of England they have similar tests, not exactly, but they do ask their banks to be able to demonstrate that they can monetize the assets that they're holding for liquidity purposes. At the Bank of England, it's perfectly acceptable. The Prudential Regulatory Authority, the PRA, part of the Bank of England, but the one that supervises banks, the part that supervises banks.

Nelson: It's perfectly acceptable for the banks to say, "Yeah, we're going to convert these to cash at a regular Bank of England standing facility." It seems an achievable and just completely sensible goal for the Federal Reserve to allow regional banks to say, "Well, we're going to sign up for your standing repo facility, making it more effective in achieving the goals that you want, and we plan on using it to convert our assets to cash in a liquidity contingency situation."

Beckworth: When you say the Fed is not allowing them to do that, is that the Fed examiners or the Fed itself?

Nelson: It's really difficult to know exactly what's going on. Because it's difficult for banks to communicate about what they're told from examiners. The Fed hasn't made any official announcements on this policy, but what I've heard is that banks that have asked for this ability have been told, "No, you have to be able to monetize your mortgage backed securities, your agency mortgage backed securities in the market."

Beckworth: You said regional banks, are bigger banks able to access the standing repo facility and use it as a form of liquidity transformation?

Nelson: All of these banks are allowed to access the standing repo facility. The issue is whether or not they can write down in their liquidity stress scenarios, whether they plan to use it. There's no credit risk to the Fed. It's hard to see how there's moral hazard associated with converting the same assets that the Fed accepts in open market operations to cash at the standing repo facility, that the giant primary dealers do all the time. Why shouldn't regional banks be able to do the same thing?

Beckworth: Well, if Vice Chair Barr is listening to the podcast today, please take note of this. We look forward to an important change in the planning of regional banks’ liquidity contingency plans. Let's move on to your big item that you really love, and that's collateralized lines of credit from the central bank as a source of liquidity.

Explaining the Committed Liquidity Facility Approach

Nelson: Sure. Committed liquidity facilities. You mentioned at the outset that I was involved in a lot of BIS working groups to design the liquidity regulations. A lot of that work was actually about committed liquidity facilities. The issue is, when you're evaluating the liquidity situation of a bank, how do you take into account the fact that the bank has the capacity to borrow from the central bank? An incredibly reliable source of funding, and may well have collateral pledged to the central bank to make that borrowing immediately available. Clearly, a bank that has that capacity is more liquid than a bank that doesn't have that capacity. But at the same time, every central bank is different in its willingness to lend.

Nelson: It's very difficult to design an environment that takes that on board in a level playing field way. Moreover, for many people, the whole purpose they see of liquidity requirements is to prevent banks from borrowing from the central bank. I forget the word for domestic blinders, seeing it in your own regional way. What's that word? Provincial. It's a very provincial way of viewing the situation because, of course, banks like the ECB, their main asset is borrowing, or at least it used to be before the banking crisis. Their main and practically only asset was loans to banks. It's perfectly ordinary. Nevertheless, built within the LCR is a method for solving this problem in a manner that's very consistent with maintaining a level playing field across central banks.

Nelson: For jurisdictions that were short of Treasury securities, because they recklessly didn't have a lot of public debt, the BIS built an alternative form of liquidity that banks could use to satisfy their liquidity requirement, which were called committed liquidity facilities. In particular, if banks established a committed line of credit from their central bank, paid a fee for that line, and maintained collateral backing the line, then they could count that as a source of liquidity. Of course, economically, a committed line is basically the same as a deposit. Both are promises by the central bank to pay on demand at some time in the future.

Nelson: Now, my efforts were to get that method allowed for everybody, not just jurisdictions that had minimal, because after all it was created so that the Reserve Bank of Australia and other similar banks wouldn't have to change their balance sheet practices just to create reserves. Any reserve bank can create reserves even if there is no government debt. You just buy things or make more loans so that your assets go up, reversing the process we spoke of at the very beginning of this program, which increases reserve balances. Any level of reserve balances is achievable by a central bank, but the BIS felt that those banks shouldn't be expected to change their central bank balance sheets just to satisfy the LCR.

Nelson: But of course, that's exactly what's happening in the United States now. The Federal Reserve has to maintain this giant balance sheet in order to provide reserve balances that are the asset being required to satisfy liquidity requirements. Ultimately, as the final act of modification of the LCR, the BIS did allow all jurisdictions to count committed liquidity facilities in the LCR. However, they added a requirement that it could only be done for a tiny slice of HQLA, and at a very high fee, far, far above market rates, which basically made the addition unworkable. My encouragement is for the Federal Reserve to offer such lines, to offer them only to sound banks and to require the banks to pay a fee for this line.

Nelson: They're currently getting this backup liquidity support from the Fed for free. This would charge them for that amount, and then allow them to count that as a liquid asset for meeting their requirements. That would have a number of really important benefits. One is, it would allow the Fed to get smaller because the banks would no longer have to hold the reserve balances. They wouldn't have to pre-fund their liquidity needs. They would just have a committed capacity from the Fed to do so. It would be a money maker for the Fed and for taxpayers. It would shrink the level of reserve balances, which would also help fix a number of the ways that capital requirements are now being distorted by this huge quantity of reserve balances, which I think maybe some of your other visitors have gotten into. We've spoken to it before, leverage requirements in particular are currently out of whack because reserve balances are so high. This was a change that I would really love to see because I think it would be very good for public policy and very good for the Fed, and would not in any way impair the liquidity soundness of commercial banks.

My encouragement is for the Federal Reserve to offer such lines, to offer them only to sound banks and to require the banks to pay a fee for this line. They're currently getting this backup liquidity support from the Fed for free. This would charge them for that amount, and then allow them to count that as a liquid asset for meeting their requirements. That would have a number of really important benefits.

Share this

Beckworth: It's called a committed liquidity facility, and it's a collateralized line of credit from the Fed, in the case of the United States. In practice, would banks really have to use it very often or would it just serve as a backup in times of stress?

Nelson: No, they would basically never use it unless they were under stress, because the way it's designed, and I provide a link in the paper that you provided a link to describing it in more detail, but the way it would be designed, when you borrowed, the borrowing would be at a high rate. You'd pay a fee all the time, but if you borrowed, you would additionally have to pay a high interest rate on that lending. It would be designed so that you would only have a strong financial incentive to only use it as a backup source of funding and to repay it quickly.

Beckworth: And only banks, so you wouldn't have shadow banks and non-bank financial firms have access to it or could they as well?

Nelson: Only banks would be able to do this, because the Fed could create these lines under its lending authority to lend to banks, Section 10b. But it wouldn't be able to create it under its emergency lending authority that gives it the authority to lend to non-banks, Section 13(3) or Section 13(13). It could do it under Section 13(13), but that's another story. But anyway, so no, I'm proposing that this be done as a natural extension of the no question to ask primary credit facility that the Fed is currently offering.

Beckworth: This would really help with confidence. It would minimize the chance of a bank run because you would know banks already have in place this line of credit from the Fed. Would this have made a difference say in 2008?

Nelson: I wish I could claim that it would make counterparties more confident in banks than they would be if the banks held an equal amount of reserve balances. But I do think that it would be equal. The improved liquidity situation of banks has improved counterparty confidence in them and bankers have relayed to me that that was actually an important reason why the spring 2020 events became, in the end, a relatively benign event without much intervention from the Fed in the moment. But one thing, I think perhaps the most important thing that it would do is it would really help economic growth.

Nelson: At a time when we're looking at increasing risk of going into a recession, that seems even more important. Why is that? Well, right now, banks are basically allowed to hold reserve balances, which are loans to the Fed, which in turn are being used to buy Treasury securities or hold Treasury securities to meet their liquidity requirements. Basically, banks can either lend to the government directly or lend to the government indirectly in order to satisfy liquidity requirements. Under this arrangement, banks instead would be able to continue to make loans to businesses and households and pledge those as collateral to back this line of credit.

Nelson: A significant part of the bank's balance sheet, which is now devoted to making loans to the government, could be converted to making loans to businesses and households. The BIS estimates that complying with liquidity requirements reduces lending by something like up to 26%, significantly reducing GDP. Now, banks would still comply with liquidity requirements and they would do so holding a range of assets, but nevertheless this would help reduce that cost to the real economy of having banks holding these sterile assets, assets that I think as we've discussed before, it is not at all clear that they would use. Instead, making loans to small businesses, to new homeowners, and then just pledging that as collateral and having the same liquidity support. It's the same liquidity strength because that collateral would be backing up a committed line from the Fed.

Beckworth: Well, Bill, we are getting near the end of the show and I want to give you the final word. Any more stories or anecdotes you want to share before we wrap things up?

Nelson: Sure, David, let me tell one more story.

Beckworth: Absolutely.

Nelson: I appreciate you [inaudible] my war stories, but during the last five years or so of my work at the Fed, I was wearing two hats. I remained in Monetary Affairs and I remained over the discount window. But because of all my work on liquidity requirements, I ended up becoming Monetary Affairs liaison to bank supervision for supervisory activities. I sat on the LISCC, the committee that oversaw the stress tests and the supervision of the largest banks. I was on the steering committee of what was then called the CLAR, which was the horizontal liquidity assessments. In those dual roles, attending FOMC meetings and sitting on these supervisory committees, I learned that just as the FOMC was planning to shrink its portfolio, the supervisors were telling banks they had to maintain their levels of reserve balances.

Nelson: For all the reasons that we have been discussing, those two things were in significant conflict. I spent a lot of time going between the various principles in those, the board and the officers in BS&R to explain that these two things were in conflict. But I left before the issue was resolved, and then about a year later I actually asked Vice Chair Quarles at the Hoover Institute, and this is all on record, whether it might be a problem for their plans to shrink, weren't supervisors telling banks they had to hold reserve balances. Then, Vice Chair Quarles said, "Yeah, we do understand that that message has been delivered, but we're trying to fix it."

Nelson: Now, of course, the Fed ran into those high reserve balance demands in September 2019, when reserve balances were far above what they expected to get down to. Recently, on your excellent show with Randy Quarles a few weeks ago, you asked him what caused September 2019, and he said, really, part of it was that examiners were, not so much the regulations, but the examiners were putting their thumb pretty heavily on the holding of reserve balances rather than other assets to meet liquidity assessments. That's really right on target with what the Fed is potentially up against today.

Beckworth: Well, we look forward to the Fed addressing this problem and if they have any questions, they can reach out to Bill Nelson. Bill, thank you once again for coming on the show.

Nelson: Thank you, David.

Photo by Stefani Reynolds via Getty Images

Support Mercatus

Your support allows us to continue bridging the gap between academic ideas and real-world policy solutions.Donate