Isabel Schnabel on the ECB and its New Operational Framework

Isabel Schnabel is a Member of the Executive Board of the European Central Bank, and she joins David on Macro Musings to talk about the ECB and its new operational framework. Specifically, David and Isabel also discuss the structure, operations, and monetary policy instruments of the ECB, the history of its operating framework, the details surrounding its new regime, 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: Isabel, welcome to the show.

Isabel Schnabel: Thanks, David, for having me.

Beckworth: It's great to have you on. It's an honor to have you on the show, to have a prominent ECB official. We have audiences around the world, but it's great to get someone who is so important to global central bank policy [to] join us. And you are a fan of all things central bank operating systems, and you were really important in this process, as I understand it, in the ECB's transition to the new operating framework. So, it's great to talk to a kindred spirit who enjoys thinking and processing through the plumbing of the monetary system, so I’m really looking forward to that discussion today. You had a recent speech that we'll turn to, as well as the actual announcement, but before we do that, Isabel, tell us a little bit about yourself. How did you get into economics, your career path, and what do you currently do at the ECB?

Schnabel: Yes, happy to do so, David. Actually, when I graduated from high school, I didn't know much about economics. That was not dealt with at school, so I had, really, very little knowledge about it. And so, I did something that many Germans did at the time, I actually first did an apprenticeship at a bank. So, that is, you work in the bank, and then half the time you are also at school. It's a two-year program. During that program, I actually had this fantastic economics teacher. He was so fascinating that I decided to study economics, and this is then what I did after finishing the apprenticeship. I went to the University of Mannheim. I also studied in Paris at Sorbonne and one year at Berkeley.

Schnabel: It's also in Mannheim where I met Martin Hellwig, who then became my PhD advisor, and I did my PhD with him at Mannheim. I wrote my dissertation about financial crises in economic history, and, actually, economic history is still quite important to me. And, I think, actually, as a policymaker, it's quite useful from time to time to think about what happened a long time ago, because you may be able to learn something about the present.

Schnabel: So then, later, I became a professor of financial economics, first at the University of Mainz, and then at the University of Bonn. I then also moved a bit into policy consulting. I became a member of the German Council of Economic Experts, which is an important independent advisory body of the German government, a bit similar, maybe, to the Council of Economic Advisers in the US, but independent of the government. And actually, then, I was lucky to be proposed by the German government as a member of the Executive Board of the ECB. There is a lengthy process at the European level, but, in the end, the European Council appointed me after having gone through a hearing in the ECON Committee. And so, now, here I am at the ECB. I started, actually, in 2020, just before the pandemic hit, and I'm actually very happy to be here at the ECB.

Beckworth: Well, you joined at an amazing time, the ECB, right during the pandemic. You probably hit the ground running, had a lot of things happening before you. I'm going to go back to that first teacher, though. You had an important teacher who played a pivotal role in your life, in changing the trajectory of your career path. So, that shows the importance of good teachers, right? 

Schnabel: Indeed, I agree.

Beckworth: Now, who was your dissertation advisor? Was it Martin [Hellwig]?

Schnabel: Martin Hellwig. You just had Anat Admati on the show.

Beckworth: Yes, I thought I recognized that name. 

Schnabel: Yes, indeed. He's a great economist, and also a very kind person. I benefited hugely from him.

Beckworth: Yes, so listeners of the show would know, he's been on twice, on the podcast, before. As you mentioned, his co-author of The Bankers’ New Clothes was on the week before we are recording this show. So, interesting to make those connections, and again, [it shows] the importance of good economics teaching. So, to all of the teachers out there, keep up the good work, you may have a future central banker among your students. Alright, Isabel, let's talk about the ECB. And, again, this show is based out of the United States, so this show does a lot of talk on the Federal Reserve, but there are some folks who may not be as familiar with the ECB. Walk us through the structure of it. You mentioned that you're on the Executive Board, there's a Governing Council, there's national central banks. Help us make sense of all of these different parts.

Breaking Down the Structure and Operations of the ECB

Schnabel: Yes, so let me start by saying that the euro area is a currency union of 20 member states, which have a single monetary policy, but 20 different fiscal policies. The euro, as you know, was launched in 1999, so it's 25 years old, and monetary policy in the euro area is conducted by the Eurosystem, and the Eurosystem is the ECB, plus the 20 national central banks of the euro area member states.

Schnabel: So, we have a single primary mandate of price stability that is defined in the EU treaties. And if I go a bit into the structure— so I said that I'm a member of the Executive Board of the ECB. The Executive Board consists of six members, including the President, the Vice President, and four other members. So, each of us has certain responsibilities. I'm responsible for market operations, which is, of course, one of the reasons why I'm so interested in operational frameworks. The other business area is statistics, which produces and disseminates statistics. And the third business area that I'm responsible for is research.

Schnabel: So, the Executive Board manages, if you want, the day-to-day business of the ECB, but it also prepares all of the decisions by the Governing Council, which is the main decision-making body of the Eurosystem. The Governing Council consists of the six Executive Board members, plus the 20 national governors. The main responsibility of the Governing Council is to conduct monetary policy for the euro area. Of course, there are other responsibilities. The ECB is also responsible for the direct supervision of the largest euro area banks through the SSM, the Single Supervisory Mechanism, which has a separate governance structure, and then of course, we are also responsible for banknotes, payments, financial stability, statistics, and many other things.

Schnabel: Let me stress one point that is very important. The Governing Council members do not represent their respective country, but we are all representing Europe, and we're supposed to always take a European view and never a national view. That is, actually, quite important. Finally, then, we have, of course, the 20 national central banks. There is very close collaboration between the ECB and the 20 national central banks, and if you want to describe how it works in just one sentence, I would say that there is centralized decision-making, but there's very often decentralized implementation, and there are various structures that we have in order to collaborate closely.

Beckworth: Okay, so that is a difference between the Federal Reserve System and the European Central Bank. We have one central bank [to] implement everything, the New York Fed, but it sounds like all of the national central banks participate in that process.

Schnabel: Yes, that's precisely right. Even though you could say that, maybe, our market operations department at the ECB is closest to the New York Fed, the implementation of monetary policy is far less centralized. So, actually, most of the operations are conducted at the level of the national central banks, but the market operations department plays a very important coordinating role, and it also conducts a small share of the operations itself.

Beckworth: So, you are a part of the Executive Board, and the President, of course, is well-known, Christine Lagarde. And where are you located? Are you in Frankfurt?

Schnabel: Yes, exactly. We're in Frankfurt.

Beckworth: Do you have branch offices, or are they distinct, like the Bank of France? How do you guys relate to each other, operationally?

Schnabel: The ECB itself has three separate buildings in Frankfurt, but we don't have any further branch offices, except for small representation offices. We have one in Brussels, we have one in Washington. And so, these offices exist, but they are, really, very, very small. But then, we have very close collaboration with the national central banks through what we call a committee structure. Every Governing Council decision is prepared by the committees. So, let's say you have one committee for market operations. It's called the Market Operations Committee. This has the staff from the ECB and all of the national central banks, and they are preparing, then, all of the decisions related to market operations in the committees, and then the governors are already informed, and when people come to the Governing Council, they already have a lot of information.

Beckworth: At the Federal Reserve, as you know, we have all of the regional bank presidents there, but only some of them get to vote. All of the board members get to vote. At your Governing Council, does every national central bank governor get to vote, or do they take turns?

Schnabel: There's actually, also, a quite complicated voting system. There is a rotation of votes, but there are different groups of countries. So, there is a small group of larger countries, and there's a large group of smaller countries, which means that the larger countries have voting rights more frequently than the smaller ones. The Executive Board always votes, but to be very frank, when it comes to the real policy decisions, we normally don't even vote. We're a very consensus-oriented institution, and normally we don't have formal votes.

Beckworth: How long do the meetings last when you come together for a monetary policy decision?

Schnabel: The meetings normally start on the Wednesday morning, and, first, we have certain seminars, which are more educational. Then, the official meeting starts after lunch on Wednesday, and then we have the whole afternoon. We have a dinner together, and then we have the decision-making meeting on the Thursday morning. After that, there is the press conference, which is done by President Lagarde and Vice-President de Guindos.

Beckworth: Okay, so, a very similar two-day process, kind of like the Federal Reserve. Now, I want to circle back briefly to the banking supervision, because the Fed plays a big role in the United States. How do your responsibilities as a banking supervisor interact with national bank regulators and financial stability executives in the government?

Schnabel: Let me first say that, at the ECB, we have a clear separation between the central banking side and the supervisory side. Even though the Governing Council, in the end, is the main decision-making body, they are actually quite separate governance structures. So, we are not closely involved in the supervision ourselves. There are other people who are doing that. There's one member of the Executive Board who's also the Vice Chair of the Single Supervisory Mechanism, so that is actually a close link. But, apart from that, we are a bit separated.

Schnabel: Then, of course, you're asking a very good question. So, in what we call the banking union, there's this principle that only the significant institutions are directly supervised at the European level, while the less significant institutions continue to be supervised at the national level. That means that you have all of the national supervisors. Some of them are at the central bank, others are separate, and so on. So, there is a combination of European and national responsibilities.

Beckworth: That's very interesting, because sometimes, here in the United States, there's discussions that the Federal Reserve has too much on its plate. It's doing too many things. And some have suggested that maybe it's doing too much with financial regulation or bank supervision. Maybe peel that off, have some other regulatory body do more of that, have the Fed focus more just on monetary policy. So, it sounds like that is what's happening at the ECB. You are, really, mostly focused on monetary policy, and then in a separate body, the SSM is doing most of the bank supervision.

Schnabel: Yes, that's correct.

Beckworth: Okay.

Schnabel: And that was done intentionally in order to avoid conflicts of objectives.

Beckworth: Yes, which I think is interesting and arguably a good approach, because it also helps preserve central bank independence. If those two things are at conflict with each other, if it gets politicized, you can easily see how that could undermine central bank independence. Let's talk about the monetary policy instruments that the ECB has. Let's start with your interest rates. You've got three interest rates that are well-known, they kind of form a corridor. So, walk us through those.

The ECB’s Monetary Policy Instruments

Schnabel: Yes, so, we call them our three key policy rates. As you said, they are spanning a kind of corridor. The lowest rate is the DFR, the deposit facility rate, which is the interest rate at which banks can deposit funds overnight at the ECB. Then, the rate in the middle is the MRO rate, so the rate of the main refinancing operations. That's the interest rate at which banks can borrow in our weekly operation. So, these are repos. These are lending operations against collateral. Then, the highest rate is the MLF rate, the rate on the marginal lending facility. So, this is the interest rate at which banks can borrow if they want to borrow overnight between two weekly operations. And that rate, again, is a bit higher.

Beckworth: We'll come back to these rates in a few minutes when we talk about the operating framework, what it was historically, what it is, and what it will be moving forward. But let's spend a little bit more time on your instruments. You also, like the Fed and other central banks, have done asset purchases, and you have an asset purchase programme. You also had a pandemic emergency purchase programme. Describe those for us.

Schnabel: I could talk a lot about all of the asset purchase programs that we've had. So, actually, the first one was a response to the sovereign debt crisis, so it was introduced in 2010. It was the SMP, the securities markets programme. Then, this was replaced by the famous “whatever it takes” program, the OMT, the outright monetary transactions. And then, only in 2014, 2015, we actually started QE. And so, we started what you already mentioned, the APP, the asset purchase programme, which includes purchases of public sector securities, but also of private sector securities. So, we were also purchasing covered bonds, asset-backed securities, and then a bit later, also, corporate bonds.

Schnabel: Then, at the beginning of the pandemic, we introduced a new asset purchase program that you also mentioned, the PEPP, the pandemic emergency purchase programme. And then finally, in 2022, we introduced yet another purchase program, what we call the TPI, the transmission protection instrument, and that is actually quite an important tool in our toolkit. So, this is a program that wants to counter unwarranted market dynamics that could pose a threat to the transmission of our monetary policy across the euro area. And similar to the OMT program, the TPI program so far has not been used. By just being there, it already has had a stabilizing effect.

Beckworth: Okay, so that's kind of the announcement effect. And I wanted to circle back to this observation that you made that you purchase both public securities, private securities, bonds, and I believe I had a conversation with Ulrich Bindseil many years ago on the podcast, and we talked about the different views on what central banks should purchase. And, if I recall correctly, he said there was a German view, and then there's a US-UK view. The US-UK view is that you just buy government securities, because from a consolidated government budget constraint, it's kind of a wash, one government liability for another. But the German view is that the proper neutral approach is to buy a wide variety of assets. So, you're buying what, maybe, the average portfolio would represent. You're not buying too much of one asset from the public. Is that the thinking behind that approach, or is there something else?

Schnabel: Well, I'm not sure I would call it the German view, but what I can say is that there are certainly different sensitivities in the US and in the euro area. And I think that is partly due to the fact that we have this complicated structure of a heterogeneous currency union with sovereign states, which means that buying sovereign bonds is a bit different from a situation where the Fed buys Treasuries, because there is this concern about market discipline, fiscal dominance, and these kinds of issues. Therefore, buying sovereign bonds was always a bit more controversial in the euro area than, I think, it has ever been at the Fed. But, another difference that I also find interesting— and we're going to come to that later as well, I think— is that in the euro area, there is a greater willingness to provide liquidity via lending operations to banks than via asset purchases. So, this is another issue which is a bit different. It's interesting to see that you have these different sensitivities.

Beckworth: So, what you're saying is that there’s a very pragmatic, practical reason why you diversified your assets when you did your asset purchase program.

Schnabel: So, basically, for us, we have a very special interaction between monetary and fiscal policy that we need to take into account.

Beckworth: We also have the targeted longer-term refinancing operations, the TLTROs. Those were also headline-grabbing when they were first introduced. Tell us about those.

Schnabel: Yes, so, of course, there are also the longer-term refinancing operations, so without the T. These are simply operations that have a longer maturity. But then, in 2014, we introduced the targeted longer-term refinancing operations, and what makes them special is that they have an incentive feature. For example, the maximum amount that a bank can borrow, or the interest rate, is linked to the bank's lending to non-financial corporates and households. So, that is the type of operation that we had. They were actually quite important, also, during the pandemic, when the banks could get funding at extremely low rates. So, basically, negative rates that were even lower than our deposit facility rate if they kept up their lending.

Beckworth: And so, the key was that they had to lend to the real economy. They couldn't just reinvest it in some other financial asset.

Schnabel: Yes, so, it's a bit like a funding for lending program.

Beckworth: Yes, that's great. Okay, I think we've covered most of the instruments. [Is there] anything else I should note about the monetary policy instruments?

Schnabel: You could possibly mention forward guidance--

Beckworth: Oh, yes, forward guidance.

Schnabel: -Which also played a certain role at the time when we were near the effective lower bound, and when we wanted to provide further easing by giving some indication on the future path of interest rates.

Beckworth: Okay. Alright, so that covers the tools of the ECB. In many ways [they are] similar to the Fed, [but with] different labels. Sometimes there are meaningful differences. We mentioned what asset purchases are included, [and], also, the emphasis on lending versus asset purchases and funding banks. Let's move on, then, to the ECB's operating system, and we just had a big announcement from the ECB. I understand that it was a 15-month process where you deliberated over what we are going to do. So, maybe kick off this part of our conversation by talking about that 15-month process. What triggered it? Why even have that conversation? What did you learn along the way?

Detailing the ECB’s Framework Deliberation Process

Schnabel: That's a very good question. Let us first, maybe, talk about what an operational framework even is because that's, of course, important. So, the goal of the operational framework is to steer short-term money market rates in line with the Governing Council decision. It’s very important that this is about monetary policy implementation, so it has to be seen separately from the monetary policy stance. That's a crucial point. The second is that this process that we've had has to be seen in the context of the ongoing balance sheet normalization. As many other central banks, our balance sheet, of course, has grown enormously, both during the phase of QE, but then again during the pandemic.

Schnabel: So, then, there was— in response to the surge in inflation— there was this turnaround. And in our case, we first had the repayments of the TLTROs, the targeted longer-term refinancing operations, and we also started, very gradually, our QT. What that implies is, of course, that now, excess liquidity shrinks, and even though we know that, in the aggregate, excess liquidity is still abundant, we know, from the Fed's experience in 2019, that you may actually be surprised that, at some point, you see that there is upward pressure on money market rates, or even tensions in money markets, even though you thought that liquidity was still abundant. Of course, we don't know really well when that is going to happen, when that could happen, because the financial system has changed so much. We know very little about the bank's demand for reserves and so on. The regulation has changed completely after the global financial crisis, we had the new liquidity regulation and so on.

Schnabel: Therefore, we need to be prepared. We have the monetary policy decision that the monetary policy portfolios are going to be run down, but we have to make sure that, on that way, we do not [have] an accident and that this goes smoothly. This is why I always argued that we should be early, and we should prepare for that early, even though some people said, “Well, you still have several years to go.” But I think it's much better to be early and to be prepared to send the proper signals, also, to the banks, [so] that they can prepare for what is going to come.

Beckworth: Now, is this going to be a regularly occurring process, like every five years or every so often you will review how the framework is working?

Schnabel: Well, actually, we said that we are going to review the system already in two years time. And the reason is that just, at the moment, the environment is changing so quickly that we thought it could be useful to already have a look in two years. Of course, if something happens, we could also review the framework earlier. We can always change the parameters of the framework if the framework doesn't do what we want it to do. And we are in a very unusual situation, coming from this situation with huge excess liquidity, shrinking the balance sheet, and getting into kind of a new world. And I think that this is why it's justified to have another look in two years.

Beckworth: Well, I appreciate that humility that you're showing, that the ECB is willing to take a look. I think it's great that central banks evaluate their operating frameworks as well as their strategy, their targets. The Fed is beginning to do a five-year review every five years on what it’s targeting. I know you did one recently at the ECB, and the Bank of Canada, I know, has done this for a long time. They review every five years, I believe, as well. But It's neat to see that same emphasis on the operating framework as well. And I believe that you guys are leading the charge on this. You're the first ones to come out of the gate and look closely at this, so, kudos to the ECB for doing that. Now, when does this new framework actually get implemented? So, it [was] announced in March. When will it actually begin?

Schnabel: There is one change in parameters which is quite relevant, which is the size of the corridor, so, the spread, in particular, between the DFR and the MRO rate, which is going to be narrowed from 50 basis points to 15 basis points. And this is going to be implemented in September of this year. We wanted to make very clear that this is not a monetary policy stance decision, and so, even though this change leads to a cut in interest rates, this is not to be seen as a monetary policy decision. It's really an operational decision, and this is why we thought that it's wise to separate it a bit from the actual announcement.

Beckworth: Okay, let's talk about the history of the operating system at the European Central Bank. You had a speech on March 14th that's titled, *The Eurosystem's Operational Framework.* And it was really fun to read it, because you go through the history in that, and you have some really nice charts that I looked closely at in preparing for the show. I really like the chart that you provided of the corridor of interest rates. You can see where the market rate— when it's between them and then when it's near the floor and such. So, let's walk through and talk about this history. What did the ECB start off with when it first began, in terms of an operating framework?

The History of the ECB’s Operating Framework

Schnabel: Yes, so, by the way, it's really a pity that we can't show the charts, because I also think they're really nice. I like them. 

Beckworth: We'll provide a link to them in the show notes, for sure.

Schnabel: If you have a link on your website, people can have a look.

Beckworth: Yes.

Schnabel: We started out in 1999 with a corridor system with scarce liquidity. So, that was, by the way, the common system at the time. How that system works is that the central bank basically plugs the liquidity deficit coming from the autonomous factors, which is mainly banknotes, and minimum reserve requirements by offering reserves in the regular refinancing operation. So, the central bank takes a view on how much liquidity is needed in the aggregate, and then, it provides exactly that amount of liquidity, and the banks would regularly access the weekly operations. By doing that, the central bank would be able to steer the money market rates to the middle of the corridor. I think that it's important to note that there was, of course, some volatility around that middle of the corridor that was tolerated. And remember that, at that time, the corridor was actually quite large, so it was like +/- 100 basis points, which sounds pretty large from today's perspective.

Beckworth: Right. Now, tolerated these wide swings, and you note in your speech that it's large relative to the Federal Reserve or the Bank of Japan, what they tolerated in their corridor systems prior to 2008. But you said that was a means to an end, because the goal was to, one, promote interbank market activity, interbank lending, have a robust market between banks, but also a small footprint for the ECB. Tell us about that first point, the emphasis on interbank activity. Why is it important, do you think, to have a healthy market there?

Schnabel: Actually, I should say immediately that this argument also played quite an important role in our more recent discussion. I should mention that there is what we call the open market economy principle, which is actually a principle from the Treaty. So, we are supposed to foster market activity. We are not supposed to crowd out markets unless that is absolutely necessary. Apart from the legal consideration, it's clear that we believe that money market activity is important, that banks regularly have to tap the markets to just see at which rates they are getting the funding. So, this has a disciplining effect, potentially. It can provide useful information for other market players, but also for the supervisors, and so we really believe that this is important.

Beckworth: So, without this interbank market, we don't get all of the price discovery and information we would otherwise get.

Schnabel: Exactly.

Beckworth: So, even things like bank stock prices, or maybe long-term bonds, they're not revealing the information you would get from an overnight interbank market. So, it's literally a missing market. And what you're saying is that the new framework, [which] we'll get to in a minute, you're trying to restore some of that, and we'll come back in a minute to that. Alright, so, that corridor, how long did it last?

Schnabel: Well, it basically lasted until the global financial crisis, which then showed that the corridor system was vulnerable, and the reason is easy to understand. So, what the corridor system does is it provides sufficient liquidity in the aggregate. It really relies on an efficient redistribution of that liquidity in the system, and this is what no longer worked during the global financial crisis. We all remember that money markets froze, and so the redistribution no longer worked, and the interbank funding costs increased quite dramatically. This is why, then, the ECB took a very important decision, which was a quite fundamental change to its operational framework, namely to offer the regular operations at a fixed-rate full allotment. 

Schnabel: That means that banks could get as much liquidity as they liked against the eligible collateral. And at the same time, the collateral framework was also broadened quite a bit. This, then, changed the entire functioning of the operational framework, because the liquidity was no longer determined by the ECB alone, but it also depended on the banks’ demand for liquidity. So, if you want, this could be seen as a shift to a demand-driven system with a relatively wide corridor, because, at least initially, the corridor was still at 150 basis points. Of course, it implied that the ECB could no longer steer money market rates to the middle of the corridor. 

Schnabel: You can see that very clearly when you look at two operations that were introduced in 2011 and 2012. These were longer-term refinancing operations with a three-year maturity, and they actually injected a lot of excess liquidity into the system. That, then, basically pushed down money market rates towards the floor of the corridor, and given that the corridor was so wide, that implied a decline in money market rates of around 70 basis points, which of course is quite a bit.

Beckworth: Yes, I encourage listeners to go look at slide 5 in Isabel's speech. The title is “Key ECB Interest Rates, Overnight Rates, and Excess Liquidity.” As they say, a picture tells a thousand words, and this one is very, very telling, and you see the wide corridor persist for a while after 2008, and then in [2015], when the APP comes in, it really gets compressed. But one thing I wanted to ask about this graph and what you just said— so in 2011, you mentioned that these new tools are introduced, and if I'm looking at this figure appropriately, the overnight market rate, it falls because of the Great Recession, it hits the floor, effectively becomes like a floor system just due to the fact of what was happening, but it struggles, and it gets back up, and it actually crosses the MRO rate, the main refinancing rate. So, that looked like it was trying to get back to a quarter, but then it quickly collapses again in 2011. Is that a correct reading?

Schnabel: Yes, it's correct. So, it basically reflects the fact that that was a time of recurrent crises in the euro area. We came out of the global financial crisis, but then we basically entered the sovereign debt crisis, which was quite a difficult period for the euro area.

Beckworth: So, just for fun, let's do a counterfactual world where there was no sovereign debt crisis. I know that's maybe highly unlikely, but just imagine it happens. Would the ECB today be on a corridor system? Also, maybe assume no pandemic as well. Assume away a lot of history. It looked like the ECB was slowly returning to something like a corridor system, a scarce reserve corridor system in 2009, 2010, and then in 2011, the sovereign debt crisis ended that journey.

Schnabel: Of course, I wasn't there, so I cannot really tell you. You would probably have to ask someone who was there, but I would guess that, at that point, there was still the idea that one would return to a corridor system, but then, of course, there were simply too many crises happening, and, in particular, then, of course, there was the low inflation period, which then gave rise to the next switch in the operational framework.

Beckworth: Now, when did the fixed-rate full allotment tender procedure get introduced?

Schnabel: That was introduced immediately at the time of the global financial crisis, so in October 2008.

Beckworth: Okay, so It's been around for a while, but 2015 is when the asset purchase programme was first introduced, is that right?

Schnabel: So, there were some asset purchases already before, like covered bonds and asset-backed securities, but the real thing, I would say, started in 2015, and of course, that then changed a lot. There was this massive increase in excess liquidity, which then pushed money market rates to the floor of the corridor. And so, we de facto moved into a supply-driven floor system. That was not a decision about operational frameworks, right? That was just happening in the context of these monetary policy decisions.

Beckworth: Right. Out of necessity, you had to do that. And George Selgin has been on the show a lot. He likes to say that a scarce reserve corridor system naturally folds or collapses into a floor system in a crisis, a zero lower bound environment. It just naturally becomes one, just out of necessity, as you mentioned.

Schnabel: Yes, and by the way, I think that is a good thing. So, you know that, if needed, you can move from one system to the other.

Beckworth: Yes.

Schnabel: The way I think about it is that there's not the ideal operational framework, but the best operational framework depends on the specific circumstances. Of course, it also depends on the institutional characteristics that you have in a certain country. The euro area is still very bank-based, much more bank-based than, let's say, the US. We have this heterogeneity, we have this risk of fragmentation. All of these things mean that, possibly, we have a slightly different system than other countries.

Beckworth: Yes, so, maybe instead of saying that we have a floor system, a scarce reserve system, a tiered [reserve system], we need to say that we have a state contingent operating system that depends on when and what is happening, in real time, to the economy. Okay, so again, I recommend taking a look at this chart. Slide 5 really illustrates a lot. So, 2015 to the present— Now, you mentioned that this is where the real asset purchases begin. And, I think earlier, you called it real QE. So, in your mind, what distinguishes this asset purchase program, that makes it truly QE or the real deal, versus the earlier ones?

Schnabel: Well, they had a very different purpose. So, this was really a program that was supposed to deal with inflation being too low. Remember that we had already introduced negative interest rates at that time. That happened in 2014. So, our interest rates were already really low, and the inflation outlook was also much lower than we would have liked. In the euro area, there was quite a bit of opposition towards QE, initially. But then, in 2015, it finally, then, happened. Some people say that was possibly a bit too late, but I don't know.

Beckworth: Yes, we used to call QE, at least here in the US, unconventional monetary policy. Today, it's conventional, right? It's kind of the standard toolkit central banks have. So, let's go to the present. We have a pandemic. What fundamentally changed then, just the PEPP program or anything else?

Schnabel: Yes, but maybe before we go there, let me also say a few things about this de facto supply-driven floor system, and this is also partly related to the pandemic. So, the pandemic programs, they were, if you want, another round of QE under a different name. Of course, as a response to this once-in-a-lifetime crisis, it had special features, and it actually worked pretty well, I would argue. But what we also saw [are] the downsides of this de facto supply-driven floor system, and one of the interesting observations was that, actually, the floor turned out to be quite leaky. 

Schnabel: So, we saw that the money market rate— so, in our case, the €STR, the euro short-term rate— gradually moved down, and, at the moment, it actually stands 10 basis points below the DFR. I think that that is something that has to be recognized. It's related to the fact that not everybody has access to our deposit facility. The non-banks don't have access to our deposit facility. If a non-bank wants to deposit money at a bank, then this gives rise to balance sheet costs for the bank, and so the bank will actually offer only a lower rate than the DFR. And so, if you want, this is a type of segmentation of markets, but it means that the money market rates are not anchored firmly at the floor, but actually below the floor, if you want. And so, that was the first thing.

Schnabel: The second thing is related to what we discussed before. So, with all of this excess liquidity, we saw that there was a crowding out of money market activity, because if there's so much liquidity around, then, of course, there's no need to tap the money market. 

Schnabel: And the third is that we also saw some impairment of market functioning. Remember that the Eurosystem had quite a large footprint. At the peak, we were holding something like one-third of outstanding government bonds, and that implied that collateral became scarce, especially the highest quality collateral like the German Bund. And at some point, actually, in 2022, we saw that repos using German collateral traded 40 basis points below the €STR, which is quite something. We were quite concerned about that, and we also saw that the transmission of our monetary policy decisions was somewhat delayed. So, we did see that this very large market footprint had some undesirable effects.

Beckworth: So, the overnight rate is below the floor. You have this leaky floor in Europe. Now, in the US, we had a similar problem, and the Federal Reserve set up the overnight reverse repo facility. One of the reasons it set it up was to kind of drain that excess liquidity. Did that possibility ever come up in discussions at the ECB, have a way to plug the leaky floor with another facility?

Schnabel: Of course, at some point, everything is discussed, but it never really gained traction, which may also be due to the fact that the non-banks do not play the same role in the euro area that they play in the United States.

Beckworth: Very interesting. Alright, that brings us to the present, then, and this framework review, and again, this decision was announced in March, and it's implemented in September. Talk about what it will be. What will we have when this is all said and done?

The ECB’s New Operating Framework

Schnabel: Okay, so, maybe I [will] start by just summarizing the main parameters, and then I [will] go— as I did in the speech— I [will] go through the three key points, if that is okay with you, David.

Beckworth: Yes.

Schnabel: Okay, so, the first observation is that we will continue to steer monetary policy through the DFR, the deposit facility rate. However, we say that we tolerate some volatility as long as this does not blur the signal about the monetary policy stance. Then, we said that we're going to use a broad mix of instruments in order to provide liquidity; so, of course, our weekly operations, and then our three-month operations. Then, we said that we're going to introduce, at a later stage, structural operations and two types of structural operations, so structural longer-term refinancing operations, similar to the LTROs that we had at some point, and then a structural bond portfolio. So, the shorter-term operations will continue to be conducted at fixed-rate full allotment against broad collateral, and, as I mentioned before, the spread between the DFR and the MRO is going to be reduced from 50 to 15 basis points. These are the main parameters, and now I can go through the logic of the whole system, if you like.

Beckworth: Yes.

Schnabel: There are three key characteristics of the framework. Let me start with the first one, which is that this is a demand-driven system. We're going to move from the current de facto supply-driven system towards a demand-driven system. What does that mean? It means that the marginal unit of reserves is provided through our regular refinancing operations. With fixed-rate full allotment, that means that banks can get as much liquidity as they like against broad collateral. The question is, why did we decide to go in that direction? I would argue that this is the best suited system for the euro area, which, as I said, is quite heterogeneous and potentially prone to fragmentation. One important issue is that if you have a supply-driven system, you cannot ensure that the liquidity is evenly distributed. So, we know, for example, that the liquidity that we injected through our asset purchases— so it basically ended up at large banks, and it mainly ended up in the core countries, not in the periphery countries. So, the question is, how do you get liquidity to all corners of the euro area? Then, a demand-driven system is much better suited.

Schnabel: This also implies that it will lead to a leaner balance sheet, because you only provide the liquidity that is actually demanded, while, when you have a supply-driven system, you always have to make sure that you inject enough liquidity in order to avoid [getting] this unwarranted volatility. And if you look at the current discussion at the Fed, for example, you see that they are starting to get concerned about the point at which you may get tensions in money markets. The problem is that none of us know where this inflection point of this demand curve for reserves actually is. That means that maybe you have to err on the side of caution, and it will be much harder to reduce the balance sheet. 

Schnabel: But we would actually like to have a leaner balance sheet, because that leads to a smaller market footprint that reduces potential financial stability risks, and it also reduces what I find quite important liquidity dependence. So, there are these papers by Viral Acharya and Raghuram Rajan who said that with QE, you have this ratcheting up of the balance sheet. So, whenever there's a shock, you go up, but you never really get it down again, and that's a bit of the problem that I also see. And I personally believe that the demand-driven system is better able to deal with that.

Beckworth: Great. Now, this push towards a demand-driven system— so does that imply that you're going to allow enough volatility above the deposit rate so that the main refinancing operation is utilized, so that banks will start going to it and maybe even have interbank activity again?

Schnabel: Yes. That, actually, already leads to the second key characteristic, which is that we said that we're using a broad set of instruments. Let me explain a bit how that works. It's closely related to your question. So, what is now happening is that we are running down the monetary policy portfolio, so if nothing happens, so no further shock, [then], theoretically, we could run it down to zero. At the same time, the autonomous factors are likely to continue to grow at a certain speed, and so, you will have an absorption of liquidity from two sides. And, at some point, liquidity becomes less ample, and some banks are going to start to access our operations.

Schnabel: And so, over time, more and more banks are going to access our operations and our volume of the MROs is going to grow. Then, at that point in time, we may then ask ourselves whether we want to roll over, every week, huge amounts of operations. Theoretically, of course, you could run down the whole monetary policy portfolio and replace it entirely by weekly operations, let's say. But [when] I talk to my people in the market operations department, they don't like that very much, because that also gives rise to operational risks and so on, and it doesn't seem like a very efficient thing.

Schnabel: And given that the demand for liquidity— part of it is structural— it makes sense to also use structural operations to provide that liquidity. So, part of those operations could then be rolled over into longer-term operations. That means, then, that that liquidity is provided at a longer maturity. None of the details have been decided, but it could be somewhat longer than the operations that we have now. And so then, at that point, of course, we still have quite a big bond portfolio, but once the bond portfolio has run down quite a bit, we could also think about building up a new structural bond portfolio. We would still continue to run down the monetary policy portfolio, but we would, at the same time, then, build up a new structural bond portfolio. 

Schnabel: And this portfolio would look very different from the monetary policy portfolio, because it has a completely different purpose. It's there in order to provide liquidity, whereas the monetary policy portfolio was there to extract duration from the market. That means that it will look different. Potentially, it has shorter maturities, but all of that still has to be decided. But this is how I think about it, and this is going to take years.

Beckworth: Sure.

Schnabel: It's nothing that's going to happen tomorrow. What could happen relatively soon is that the first banks start accessing our operations.

Beckworth: Do all of these things eventually lead to increased interbank lending and interbank market activity?

Schnabel: Yes. That leads me to the third key characteristic of the framework, which is what we call the soft floor with a narrow corridor. So, if you compare our system with, for example, the Bank of England, you will notice that they actually have a zero corridor, right? Of course, they also have different collateral requirements and so on, but they have a zero corridor. We decided against that precisely for the reason that you mentioned, because we believe that interbank market activity is important. So, when thinking about the ideal size of the spread, there's basically a trade-off that you face. On the one hand, you want the spread to be small enough so that you can limit volatility, in particular, in this difficult transition phase, to a system with less excess liquidity. 

Schnabel: On the other hand, you want there to be an opportunity cost of holding liquidity. So, the idea is that if a bank goes to the MRO and asks for too much liquidity, and it doesn't need it, it may actually have to deposit it at the lower deposit facility rate. So, this creates an opportunity cost. So, of course, it will first try to lend this money in the market, which then gives rise to money market activity, but, in any case, the bank will be cautious not to ask for too much liquidity, because there is an opportunity cost.

Schnabel: This means that the bank itself becomes responsible again for its own liquidity management, so it can no longer simply rely on the central bank. So, the central bank is not a lender of first resort, but the bank will first try to get different sources of liquidity before, then, actually going to the MRO. All of this, then, hopefully revives the money market activity. Of course, we don't know exactly. So, we now chose this parameter of 15 basis points. Whether that is sufficient in order to give rise to meaningful money market activity, we don't really know. But this is why we have to monitor the system very carefully in order to see whether that spread may be too small or, also, too large. Several things, of course, can happen.

Beckworth: And that's why you guys will be revisiting this in two years, but it was interesting to see that you had 15 basis points between the deposit rate and the main refinancing rate and then 25 basis points between the main refinancing rate and then the lending facility at the top.

Schnabel: Yes, so that is, of course, deliberate. What it does is that— if you abstract from the interbank market, it makes it more expensive to have too little liquidity than to have too much liquidity, because it's asymmetric, which then helps to anchor the rate more at the floor. It's important. We call it a soft floor. Why do we call it a soft floor? The main reason is that, if the rate actually moves up, which could eventually happen, there's no intervention in order to go against that. So, this is the type of volatility that is accepted. But there are situations where we would say that this is not in line with the spirit of our framework. For example, you could have the rate moving up and more or less anchoring at the MRO rate. Then, of course, we would have to have another look to see whether it's still appropriate to say that we steer monetary policy through the DFR. But it could also happen that we get a lot of volatility that we don't want to have. Then, again, you would have a review, and you would need to think about how to deal with that. But this is part of the learning process that we are going through in the next years, I should say.

Beckworth:  Well, Isabel, I look forward to the progress, as this new framework unfolds starting in September and then over the next year, and I'm looking forward to what happens with the next review that you'll do in two years. Maybe we can have you back on the program to reassess where we are in this journey. But with that, our time is up. Our guest today has been Isabel Schnabel. 

Schnabel: Thank you so much for having me today.