Loretta Mester on How to Improve the Fed’s Operating Framework

Loretta Mester was president and CEO of the Federal Reserve Bank of Cleveland from 2014 through June of 2024, and she is a 39-year veteran of the Federal Reserve System. Loretta is also currently an adjunct professor of finance at the Wharton School at the University of Pennsylvania. She joins David on Macro Musings to talk about her time as Fed president and a recent paper she delivered on the Fed’s operating system. David and Loretta also discuss the ongoing battle against inflation, what to expect from the upcoming Fed framework review, and much more.

Check out our new AI chatbot: the Macro Musebot!

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].

Register now for Building a Better Fed Framework: The AIER Monetary Conference.

David Beckworth: Loretta, welcome to the show.

Loretta Mester: Thanks very much for having me.

Beckworth: It's great to have you on. Now, I ran into you at The Shadow Open Market Committee's 50th anniversary conference at Hoover, and it was a great time, great conference. You had a fascinating paper that we're going to get to in a minute. And I was thinking, Loretta, the first time I met you was at that same facility at a Hoover Monetary Policy conference. And as you know, they have these dinners at the end of the day, and they make sure you sit with people that you don't know. I guess that's the intent, at least. And so, I got to meet you, which was great. But as it turned out, you had me to the left of yourself and then Evan Koenig to your right, and we both happened to be nominal GDP targeting advocates. I believe this was like in 2019, sometime around May. And so, we were just peppering you with nominal GDP targeting advocacy.

Mester: I felt that I had a target on myself.

Beckworth: Literally, yes. So, you were very gracious, and you tolerated me, and we had a good time. But it was great to see you at this conference just a few weeks ago at Hoover, again. You gave a very interesting, and, I think, timely talk— the challenges that the Fed has with its operating system, many central banks in advanced economies. And there's been a lot of developments there, so we'll come back to that. But you are also a long-time veteran, as I mentioned, of the Federal Reserve System. It runs through your blood. You know the ins and the outs of that system, and I want to talk about that for a while. Before we do that, though, tell us, how did you get into economics and get on a career path where you ended up at the Fed?

Loretta’s Career Path and Tenure at the Cleveland Fed

Mester: It was kind of a fluke, actually, because I went to undergrad at Barnard College, which is part of Columbia, and I was a math major, but it was very easy to add economics as a second major, so I did that. Then, I applied to graduate school in mathematics. I never applied to a graduate school in economics, but I ended up getting letters from two professors at Princeton who said, "Our program here in economics is really mathematical. Why don't you come and do that?" So, I did.

Beckworth: Really?

Mester: That's how I ended up doing economics. I know. It seems unheard of now that you would just be like, “Okay” But it was great. It was a great program. I met my husband there, who was also a grad student, so that was good. Then, once you get your PhD, the last year you're there, you go on the job market. It's a big deal. All of the PhDs coming out go on this market and look for a job. And so, my husband ended up getting a job at the University of Pennsylvania in their economics department, and I ended up getting an offer from the Philadelphia Fed.

Mester: It wasn't my intention to go to the Fed, because you're trained to go to a university, but they made me a nice offer and offered me a lot of research time and policy issues. And so, I ended up there, and that's how it started. Then, the Fed was such a fascinating place, and it allowed me to do academic research but also really be involved in policy work, and it ended up being a great place. Then, when the position opened up at Cleveland for the presidency, I threw my hat in the ring and ended up getting that job, so I went to Cleveland.

Beckworth: Well, that's a great story. So, you rose through the ranks of the Philadelphia Fed. If I understand correctly, you were the executive vice president.

Mester: When I left there, I was executive vice president and director of research, so I oversaw the research department. At that time, it was called the Payment Cards Center and the statistics department was part of the management team.

Beckworth: One thing that I've always liked about the Philadelphia Fed is its data, particularly its real-time data center and also the Survey of Professional Forecasters.

Mester: Exactly.

Beckworth: That's been very useful in my own research.

Mester: Great. That's wonderful.

Beckworth: No, it's absolutely a wonderful thing. I'm glad that there are organizations like the Philadelphia Fed that take the time and effort. You have that data, for example. The St. Louis Fed has FRED. So, we're all providing public services.

Mester: Right. Exactly. And so, it was interesting, because when I was at Philly running the research department, that was when we established the real-time data center. So then, when I went to Cleveland, one of the things that I wanted to do there was establish a center for inflation research, which we did. Because I really think that having a center can really propagate the work and make sure that it's there for the long run. So, we established that, and again, we're providing data on some of our measures of inflation that everyone looks at that come out of the Cleveland Fed, and research as well.

Beckworth: Well, Loretta, I didn't realize that you have impacted my research, because you're responsible for the real-time data at the Philadelphia Fed, and I've also used the inflation data from Cleveland.

Mester: Excellent. That's good to hear.

Beckworth: In fact, I was going to ask you, what is the Cleveland Fed known for? In my mind, it's known for its inflation data— expectations as well as these nowcast measures that you do.

Mester: Yes, we do nowcasting. The median PCE now and the median CPI come out of there, and they're doing a lot of really good research that's basically aimed for policymakers, but then they do academic-style research, real technical research. But then, they also have, if you go out to their website, a really good set of things that talk to regular people about, what is inflation? What isn't inflation? So, I think that that's very helpful too, that they've aimed for the three audiences— very technical people in the field doing research, policymakers who have to take things that aren't necessarily exactly what's happening in the world, but then apply it to make policy decisions, and then, also, to regular people who are really interested in economics and inflation in general.

Beckworth: That's a great legacy that you've left behind at the Federal Reserve System.

Mester: Thank you.

Beckworth: So, you literally bleed the Federal Reserve in your veins, and you have left a footprint that's affected many people, including myself. Now, let's talk about your role as president, because you voted on the FOMC. And people want to know, what is it like to be a president of a Federal Reserve Bank? I know that there's no typical day, but walk us through your routine.

Mester: It really depends on what you're doing that day. So, a lot of people know that the Fed is in charge of monetary policy, and you’re right, we go to FOMC meetings. We deliberate. We come up with a decision. But monetary policy is only part of what you do when you're a Fed president, because you're running an organization. So, the Cleveland Fed had over 1,000 people doing various things. And so, the responsibilities of the Fed include monetary policy, but they also include things like financial stability, supervision of banks.

Mester: It's a delegated function from the Board of Governors, but the examiners for the banks in each district really lay within the Reserve banks. You have that responsibility. The Cleveland Fed was one of, I think, four or five Reserve banks that actually do payment services for the US Treasury. So, we're running payment services for them as their fiscal agents. So, that was another important duty that Cleveland has that not every Reserve bank has.

Mester: Then, of course, very important is some of the community development work that we do, because it's important for policymakers to really understand what's happening in their districts. So, that's partly understanding the economics of the region, and the manufacturing sectors, and what's happening in services, and being tied to that. Also, really understanding what's happening in communities, because as we saw during the high inflation, I think that people forgot— there was so much focus on the jobs, making sure people have jobs.

Mester: I think that people forgot about the pain that high prices and rising inflation, well above 2%, can inflict on regular people. And I think that was something that the community development group within each Reserve bank— and I think that Cleveland has a particularly good set of people doing that work. A strength of the Cleveland Fed is their community development work. That really helped us, as policymakers, to really understand what was actually happening on the ground, in the district. So, I think that's another thing that people forget, is that we're not just sitting in offices looking at mathematical equations. We're actually going out and talking to businesses, and talking to labor market representatives, and talking to community development people, so that we really understand the impact of our policy on the economy and on real people and communities.

Beckworth: That's interesting. The inflation surge, where people were affected, and your community connections were important [and it] really complemented your other center on inflation, because it was an inflationary experience. And we've talked a lot on the show about how all of us have been reminded how much people dislike inflation once it gets past some threshold. There's some point where we suddenly begin to pay attention.

Mester: Exactly.

Beckworth: And different people are affected differently. Low-income people may be more adversely [affected], and maybe upper-income [people] can be insulated better from the shocks. But your bank was very engaged in that conversation.

Mester: Yes, because I think that it is important, and I think what you pointed out is correct, especially in this particular episode. If you think about it, a lot of the things that were most affected by price rises, because of the supply chain, were really essentials. And if you think about, what's the consumption basket of lower-income versus higher-income households? A higher percent of their consumption basket is, by necessity, essentials.

Mester: So, they particularly felt the high inflation, especially when it started rising. I would say that the impact was higher— was a deeper impact on the low-income households than on the higher-income households who, of course— many, many, high-income households recognize, “Wow, look how much a dozen eggs cost now,” but they have the wherewithal to be able to afford it. The lower-income households were really adversely impacted by that.

Beckworth: So, part of the puzzle or ongoing conversation right now is, why are people still upset about this? An easy answer would be that, well, they're confusing inflation rates with permanent price level increases. It's hard for them to get over this permanent increase, which is really steep. Others have said that there's conflicts associated with inflation, or that maybe we're not fully measuring it. What is your sense of why people have been so bitter about the rise in prices?

The Ongoing Battle Against Inflation

Mester: Well, it is definitely true that you're paying higher prices. Even if inflation comes back to 2%, and it's on its way back to 2%, the price level is higher. But I think the question is, okay, but have their wages gone up as well? And so, I started looking at that, because you got this sort of pushback of like, “Wow, it's a really hard market out there and I don't like it.” And they were right in the sense that, even though real wages today are positive— in other words, wages here are higher than price increases, wage increases. But that was a more recent phenomenon.

Mester: I would say that inflation really started picking up in April 2021. If you take that as a baseline and say, if I look at how much the price level has risen, and I look at how much wages have risen, are we back to that trend line? No, there's still a gap. In other words, you haven't made up for all of the time when price rises were outpacing wage rises. If you look at a chart that just looks at the price level and the wage level, you haven't totally closed the gap yet. So, they're right to still be upset, because their wages haven't fully closed the gap with prices.

Mester: Eventually that will happen. Then, I think that people will begin feeling like, “Okay, yes, I'm paying higher prices for this basket of goods, but I'm also earning enough to make up for that.” So, it's almost like the cost of living has changed, and you're really going to need to see that gap close before you can say, "Okay, we've gotten past that episode." It's closed over time, and it's closed for certain income groups, but it hasn't totally closed, and I think that's part of it.

Beckworth: So, part of that journey to closing that gap is the Fed lowering the inflation rate. It's shot up. There are different stories we tell for why it shot up— supply side, demand side stories. But now, it's pretty close to 2%. So, you're a former FOMC member. You're outside of it now, so you have a little more freedom. Could you comfortably say “mission accomplished?”

Mester: As a former Fed person, and as someone who is very appreciative of the Fed as an institution, I don't think they should ever say “mission accomplished,” because it's an ongoing mission. Yes, inflation has come down well off its peak. If you look back at where it peaked, depending on the measure, between 6% and 9%, that's a high inflation rate, and it's well on its way back to 2%, I would say, in a sustainable way. But there are risks out there.

Mester: There's some upward pressures that you can think about on inflation. Supply chain issues could come up again. The geopolitical risks have potential for raising oil prices, which might have knock-on effects on the prices of other things. So, I don't think it's ever right to say mission accomplished. But I'm very grateful that inflation has come back down, and what's remarkable is that it's come back down, and still, labor markets are strong.

Mester: They've moderated. They're not overheating anymore and adding to inflationary pressures, which was happening earlier. But that's been a remarkable accomplishment. But still, you've got to understand the risks. There's risks to both the maximum employment part of the mandate now, and there's risks to the inflation part of the mandate. So, they have to stay on the job.

Beckworth: Okay, so, the mission is not accomplished, but major progress has been made.

Mester: Correct.

Beckworth: And it's great that we got here without a recession. So, at that same conference that I mentioned earlier that we both attended, I got to ask Chris Waller, during his lunch address, “How did we get here? Reflect back on this big debate that you had with a few other prominent economists that will remain unnamed.” Of course, he tells the Beveridge curve story. Were you surprised by how relatively painless this return to low inflation was?

Mester: Well, it depends on what you mean by painless. I think that a lot of people suffered from the high inflation, so I would never want to say it was painless. I think one thing that people forget is how unprecedented the pandemic was, and the policy actions taken to address the pandemic. It was very hard during that period to sort of say, "Well, it worked this way the last time, and therefore it'll work this way." So, I think that this was a case where you had to be very humble about what you knew and how much you could extrapolate the past to the future.

Mester: But I was gratified that we were able to bring inflation down while maintaining pretty healthy labor markets. But at every point in time, you weren't assured of that. And so, I think that was the thing that was the difficulty. You know now, in retrospect, what it looks like, but in the moment, you were always concerned that, okay, we haven't seen this labor market really deteriorate, but it could. We were helped by the fact that there was a high level of immigration coming in to actually ease some of the problems in the labor market, and we are helped now, because if you look at what they call prime age—

Mester: I hate that word because I'm like beyond prime age. In any case, 25 to 55-year-olds— they've come back into the labor market. In fact, there was a big dip, of course, when the pandemic was raging, in terms of labor force participation. But at least for those people in the prime working ages, that surpassed where it was before the pandemic. So, there's been a lot of progress on the labor market side as well. I'm just grateful that we were able to do that and not inflict a much higher unemployment rate. The labor market looks really good right now.

Beckworth: Yes, and I want to be clear, too. I'm not saying it was painless, but it could have been a whole lot worse. It could have been a far worse scenario than many were anticipating.

Mester: Well, I think that was the other thing that was— when the pandemic first hit, if you look back, there were some very dire scenarios— very dire— which I think was the spur to action, to go in large. And I think [those] actions, both the fiscal actions and the monetary policy actions, mitigated some of the downside risks. But I also think that we were lucky, in some sense, that things that could have happened did not happen. So, it's a situation that I hope we don't have to live through again, but it was a good situation in terms of what we learned about how the economy works and the underlying strength in the US economy.

Beckworth: Yes. Something else we learned through this ordeal was, how do frameworks operate? How do they manage big supply shocks [and] inflation? So, it's a nice segue into the fact that we are going to soon start the Fed framework review. You were part of the previous one five years ago, and I just would like to know, before we get to your paper that we were talking about earlier, what do you think will happen in this framework review— minor changes, major changes?

Evaluating FAIT and What to Expect from the 2024-25 Fed Framework Review

Mester: Well, I think they're going to actually take a step back and really do a thoughtful review. And so, what I hope they would do is start with a review of what was learned during this episode of high inflation and what, given this episode, we should think about in terms of the framework. When we talk about the monetary policy framework from the point of view of what is supposed to be announced sometime later this year starting, it's really about the strategy for hitting the Fed's two goals of maximum employment and price stability, but also the tools for doing so.

Mester: So, the Fed has a number of tools— the fed funds rate, the interest rate tool, and then the balance sheet tool. Then, the third part of it is, how do you communicate it in a way that makes sure that people understand your strategy [and] understand your rationale for your policy decision, which is very important for making sure that monetary policy actually has the effect that you're hoping it has on the economy. People have to understand what you're doing, and they formulate expectations about inflation, and that's something that you're trying to control so that you can actually hit your goals.

Mester: So, those are the three parts that I expect that they'll review— each of them, in turn, as they did last time. But I hope they do take a step back and look at this episode and take a little critical eye on what could we have done differently, what could the Fed have maybe [learned] from that, and then have that inform— Not that they should be fine-tuning to the latest episode. In fact, I think that's one thing that we do learn from this episode, is that it's really not a good idea to over-index on recent events.

Mester: You should really take a step back when you're doing a framework and really think about other things that can happen. You want the framework to be able to be applied to different types of economies and different things happening in the economy. So, I think that was a lesson from the last framework review that hopefully will be helping to guide the next one. I also think that they need to be bringing back a more symmetric view of things, rather than just—

Beckworth: -The asymmetries that are released across the framework. That's interesting because a lot of commentary has been made, written, and spoken about the framework recently, and some of the views I see that are common across these observations— One, like you said, be robust to different scenarios, whereas FAIT, as it's on paper at least, is very much designed [for] a lower-bound, low-inflation-type world, kind of what we saw in the 2010s.

Mester: Flexible average inflation target, for your listeners here.

Beckworth: Yes, so from FIT to FAIT, flexible inflation target to flexible average inflation target. But Gauti Eggertsson and Don Kohn had a paper, and they said, "Look, this was great for that world, but it's not robust to all worlds,” and they want to see something that's more robust and symmetric, as you mentioned as well. So, maybe, I don't know, drop the asymmetry on the maximum employment. Would you go back to deviations or stick to shortfalls?

Mester: It's interesting that deviations and shortfalls— Most people think, well, that's just a word. It kind of means the same thing. There was a lot attributed to that word as being a difference in strategy. So, when they changed to shortfalls, people thought, "Oh, they only care about when employment is below a certain level. They won't care if employment seems to be above what is sustainable." It turned out that people interpreted that as meaning that the Fed wasn't going to infer anything about inflation from labor market conditions.

Mester: Now, that's not how the Fed acted, but that was the inference from that. So, my preference would be that they go back to deviations, but also that they actually explain what maximum employment is, and I think that maximum employment is the maximum level of employment consistent with price stability. I think that linking of the employment part of the mandate to price stability would be a huge help, because then, they would be able to use conditions in labor markets to help them forecast inflation and have nobody think that it's not consistent with the framework.

Mester: So, I'm hoping that they at least do that in that statement, so that people have a better context of what we mean by maximum employment. I would not want them to put a numerical value on maximum employment, because one, we don't know what it is, and two, the Fed doesn't affect that. There's a lot of other structural things in the economy that affect that. But we still have to try to achieve that in a way that's consistent with price stability.

Beckworth: Yes, and even the statement, the longer-run goals statement— It says that this is something that's dynamic and changing. So, even if we did have a number, it might change from time to time. So, we've got to be humble about it, I guess, is what they're saying. One last question on the framework before we move on to your paper, and that is, FAIT called for these two asymmetries, the shortfalls and then also the making up for below-target inflation, but not above.

Beckworth: And again, I think that there's big communication challenges, as you noted, with both of those. But here's my reading of what happened. I want you to correct me if I'm wrong on this. So, FAIT was implemented as designed, on paper, from its announcement date in, I guess, August or September 2020 up until really when the Fed started to raise rates in early 2022. It effectively abandoned FAIT. The FOMC went almost to a “focus on the price stability” mandate. They were no longer following the shortfalls. They were going back to more of a FIT approach. Is that a fair assessment, or was FAIT always there?

Mester: I personally think that the framework wasn't a big problem, if you will. I know that some people thought that that was the reason that the Fed didn't raise rates earlier, et cetera, et cetera. I think it was more the forward guidance that was actually in the statements, and that forward guidance, of course, was trying to be consistent with— But that forward guidance was so restrictive in terms of, what were the conditions? Because it had both inflation being a little bit— getting it above 2% and labor markets being at maximum employment. That ‘and’ was significant there. I think that was partly why— Of course, there was uncertainty around all of that—

Beckworth: Sure.

Mester: -what was happening in the economy. But I think that was part of why. Perhaps we wanted to be consistent with that forward guidance. Because when the Fed puts something in a statement, it doesn't want to, the next time something happens, go against what it said earlier. But that also means that part of the framework review should really think about, what's the proper way of constructing guidance, and what's a way that you avoid feeling that you're maybe trapped a little bit?

Mester: Or not even trapped, because I know the Fed very well, in the sense that if they thought that something was the right thing to do, they'd do it regardless of what they might have thought in the forward guidance. But when you're in that nebulous world where you could go either way, there's uncertainty, [then] you're going to try to be consistent with the forward guidance. And I think that was part of what happened. But that means that you can learn from that and then construct—

Beckworth: Have a debate or some considerations about how to use forward guidance with the framework.

Mester: Yes, and that is a tool that you use at the lower bounds. So, that should be part of the discussion of the framework review.

Beckworth: So, let's move on to your paper, which was a great paper that you presented at the 50th anniversary conference of the Shadow Open Market Committee. The title of the paper is, *The Fed’s Ample Reserves Monetary Policy Operating Framework: It Isn’t as Simple as It Looks.* So, maybe, let's step back for those who aren't familiar. I think that most listeners know the difference between a floor and corridor system, but just maybe walk us briefly through the history as a way to illustrate the change from the corridor to the floor.

Corridor vs. Floor: The Evolution of the Fed’s Operating System and its Policy Implications

Mester: So, since we were just talking about the framework, let's just make sure that everyone understands that we're talking about the operating framework. So, this isn't about a policy decision [like] “Where is the fed funds rate target, where should that be?” This is, once the FOMC is decided on the target, how do you make sure that you can hit the target, that you're implementing policy? It's about policy implementation as opposed to what the policy is.

Mester: If you think about prior to the great financial crisis, the Fed was operating in, what we call at the Fed, a scarce reserve system, which a lot of people call a corridor system. What that meant was that banks need reserves for a number of purposes. At that time, there were actually requirements. They were required to hold a certain level of reserves, but they also would hold them anyway. They'd hold, a lot of times, above the requirement to ease payments, to make sure that they have enough for actually executing payments, and also for liquidity purposes.

Mester: You wanted to make sure that if you were a commercial bank, you had enough reserves— which, of course, are in master accounts at the Federal Reserve— to be able to run your bank. So, at that point, there weren't a lot of excess reserves in the system. It was basically that banks were sort of using [those] reserves to do their business, and what the Fed would do to actually hit a fed funds rate target— which is the overnight interbank interest rate when banks are lending each other reserves— is that they would just figure out the demand curve of reserves, and it was downward sloping because there weren't a lot of excess reserves.

Mester: And they'd supply the amount of reserves that would be required to hit a target, and the target is the intersection between the demand for reserves and the supply of reserves. Well, what happened? In the great financial crisis, the Fed's balance sheet grew, and we bought a lot of assets to make sure that markets continued to function. But also, we bought a lot of assets to lower long-term interest rates once we had brought down the fed funds rate, which is a short-term interest rate, to zero. We wanted to add more accommodation because the economy needed it. And so, the Fed's balance sheet grew, and it grew substantially. In fact, it grew so much that there were a lot of reserves in the system, because the reserves are a liability, [and] assets are on the balance sheet. Well, at that point, small changes in the supply of reserves by the Fed— that doesn't change the fed funds rate at all.

Mester: So, you're operating on a very flat portion of the demand curve. You can add a lot of reserves or not, and you're not going to change the interest rate. That's called a floor system. So, at that point, it was like, okay, how is the Fed going to implement its monetary policy if, by changing the supply of reserves, you're not affecting short-term interest rates? At the same time, Congress gave the Fed the authority to start paying interest on reserves. That allowed the Fed to be able to control short-term money market rates by setting the interest rate that it was going to pay on bank reserves.

Mester: Once it started doing that, that's the way it was controlling interest rates, because banks had a lot of excess reserves. If the interest rate that they could get by lending those reserves in the market was higher, they would do that. If the interest rate— They could invest in other assets and still short-term assets, they would do that. So, basically, by controlling that interest rate on overnight reserves, they can control the fed funds rate, and then the fed funds rate is related to other overnight interest rates.

Mester: So, that interest rate on reserves, reserve balances, is how the Fed is implementing policy. Now, it still communicates policy via the fed funds rate, because that's a very well-established— people understand what a 5% fed funds rate means or what a 3% fed funds rate means. And so, even though there was discussion about, “Maybe we should change that target to something more related to other short-term interest rates,” the decision was, “No, we'll continue to communicate our policy by the fed funds rate.”

Mester: So, that seems like a very elegant thing. You don't have to spend a lot of time, [if you’re] the Fed, to understand the downward-sloping part of the bank demand curve for reserves, because they'd have to estimate that every day before they'd know how much to supply in the old system. In the new system, you don't need to know that. As long as you know that you're operating where the reserve demand curve is flat— In other words, you're beyond the scarcity of reserves and banks don't really need a lot more or a lot [less]— You can just set anything for the supply so that they can control the interest rate on reserves, and that basically makes things work.

Mester: However, what it turned out being is, in the US, of course, as you know, we have segmented markets, and not every financial institution that operates in the short-term money markets can actually earn interest on reserve balances at the Fed. So, for example, the vast majority of lending right now in the fed funds rate market is from the Federal Home Loan Banks. The Federal Home Loan Banks are not authorized to earn interest on reserve balances in their accounts at the Fed. So, what that means is that they're willing to lend in the market at any rate above zero, because what else are you going to do? And so, what happened was that, that floor, the interest on reserve balances, that interest rate, wasn't really firm, because you had a lot of transactions that could occur below that floor.

Mester: So, that's one thing that wasn't as simple as it was on paper, in a diagram, because you had these other entities and even money market accounts— same thing. They don't earn interest on their reserve balances, but they're active in the short-term money markets. And so, that can affect the fed funds rate market. So, what the Fed did is they set up an overnight reverse repo facility. Essentially, it puts a firmer floor under short-term interest rates, because that is another rate the Fed can control, and if it looks like there are a lot of trades going on below the interest rate on reserve balances, [then] they can offer things on the ON RRP, and it'd sop up those excess reserves. So, that's a way that they're better off— They have a firmer floor. At the same time, if you think about— well, what's the ceiling on rates? The ceiling is the primary credit rate, because if you know that you can borrow at the primary credit window—

Beckworth: That's the discount window.

Mester: -That's the Fed's discount window, exactly— then you won't borrow a lot in the market. You'll go to the Fed's primary credit facility. However, it turns out that banks don't like to come to the window, because there's stigma attached to that. Even though it's not announced who's coming into the discount window to borrow— and it's reported with a two-year lag— banks are concerned that that'll get out and it'll look like they're having problems, rather than just going there because of the pricing— there's arbitrage in the market that they could do.

Mester: So, that meant that that ceiling wasn't as firm as it should be either, and so the Fed then set up another facility to have a stronger ceiling, and that's the standing repo facility. So, again, it's a way of making sure that they have tighter control on that fed funds rate target so that they hit that and, therefore, can control the short end of the money markets, which is what they're hoping to do with implementing policy. Well, that is another complication that made the nice, simple ample reserves regime not as simple as it looks, because you're running two extra facilities that the Fed is in, doing transactions in.

Beckworth: So, it's not as simple as the textbook story and maybe as simple as was articulated when the FOMC officially voted this into being. And I want to go back to that history just briefly. So, you were at the Fed before you became president, but in 2014, you take on the role. So, you're a part of the discussions, and again, it wasn't until 2019 that, officially, this system was endorsed or recognized by the FOMC. Is that right?

Mester: Yes, and after the great financial crisis, of course, our balance sheet rose. We had already started trying to bring the balance sheet down in 2017, but we wanted it to run in the background, because we don't think of the balance sheet as the main tool of policy, and we also had never done it before. We had brought— The balance sheet grew, but we wanted to be cautious about it.

Beckworth: Sure.

Mester: So, rather than selling assets out of the portfolio, it was basically going to be shrinking the balance sheet by runoff. When assets ran off, you weren't replacing them. So, that started in 2017. By 2019, it was determined that there are some good things about ample reserves. We don't have to estimate the demand curve. It's pretty simple. And so, it was announced that that was what the intention was, to actually do ample reserves.

Beckworth: Before you get to that point, though, you come in, it was 2014. Were you thinking, "Oh, we're going to eventually get back to the pre-corridor system”? When did it hit the FOMC, “Well, maybe we won't”? I bring this up because Bill Nelson, who's been on the podcast, tells a story. I think [in] 2014, 2015, when the supplemental leverage ratio was written, the rule was written, they had no idea that it would become a binding constraint. They were estimating far fewer reserves in the banking system. And so, at least at that point, people were expecting a much smaller balance sheet. So, when did it hit you that, "Wow, maybe we're not going to go back to smaller?"

Mester: Well, when the discussion happened around 2017 about the shrinking, I think that it was becoming clear that it takes a long time to get a balance sheet back to what it would have been. And of course, it would never have gone back totally to what it was before, because the economy is growing.

Beckworth: As a percentage of GDP.

Mester: Yes, exactly. So, I think that the feeling then was— as those discussions were occurring— part of the discussion was, well, why go back? And so, that was when some of the other issues about a large balance sheet came up. Charles Plosser, who was at the conference that we were both at recently— he's written a lot about and spoken a lot about the fact that, if the operating framework— which is what we're talking about now— puts no constraint on the size of the Fed's balance sheet, then it could be used— the Fed's balance sheet could be used for other political purposes to finance different things. And it has, over time, been used that way, or, at least, it has been asked that the Fed use its balance sheet to do some things.

Mester: It takes, then, the Fed chair saying, "Yes, we're not doing that." There were things about bailing out the car industry and things. So, I think that Charles makes the point, and I think it's legitimate, that it puts more political pressure on the Fed, because there's no constraint. Remember, in the scarce reserves [system], there was a constraint, because you needed to be operating on the downward-sloping part of the demand curve to have that operating framework work. That limited you. So, you, as Fed chair, could go in [and say], "Sorry, Congress, we can't do that, because we would not be able to implement our monetary policy." Now, a Fed chair can't do that. It has to point out that, "No, sorry, we can't do that. That's not in our mandate." That's a little harder [of an] argument.

Beckworth: Yes, so, it undermines the Fed's independence, right?

Mester: Yes, so, potentially, that could underline if people think that they can start using the balance sheet for things.

Beckworth: The way I think about it is that the Fed's balance sheet is a temptation that, on the surface at least, looks like it's a free lunch. You can use it, but deeper down we know that it's using up fiscal space that should be saved for Congress's choices or fiscal policy.

Mester: Also, it's a fiscal policy decision, not a monetary policy decision. That's why the Fed's principles that are around its balance sheet [are] that it wants to get back to a balance sheet that's primarily Treasuries, and that is because, when it does buy other assets, one could argue that that's credit allocation, and the Fed does not want to be in the business of allocating credit to different sectors of the economy. It wants to basically be neutral in that respect.

Mester: Now, the interesting thing about that is if you look at projections of the runoff of the agency mortgage-backed securities, it's going to take many, many years to get back to, primarily, Treasuries, unless they're willing to sell some of that portfolio. And I would think that they would want to sell some of that portfolio. That would be very consistent with the principles, because the principles allowed for that at some point. I don't think it's imminent at all, but I think that would be something that— I would like them to have a plan of how they're going to get back to, primarily, Treasuries. I think that that would be a good thing to be working on as well, when you're thinking about the operating framework, et cetera.

Beckworth: We had Peter Stella on the show, and he's worked on central banks at the IMF, and he's seen what's happening in the US and other countries long ago. And he made this great point, that when the Fed buys Treasuries, it's not really a transfer, because it's just one side of the government offsetting another side of the government. But when it bought mortgage-backed securities, the transfers were to households. David Beckworth, who refinanced his mortgage in 2021, is a recipient of that. And so, it is very different in nature. So, it would be nice to see a return to Treasuries only, is the idea.

Mester: It's interesting, because the great financial crisis, of course, the center of that crisis was the housing market. So, you could make an argument that that was an important thing, because that market wasn't working well. In this episode, it was harder to make that argument. Because we had done that before and we knew those markets, it was determined that we would implement the same kind of program that we did before, which is buy Treasuries and agency mortgage-backed securities.

Beckworth: So, you bring up a lot of the challenges that make this system not so simple. You already touched on the leaky floor. I love how you put it in the paper. It's a cork floor. It's not a marble floor— nice image there of the water seeping through the floor. So, you're trying to catch it below with the overnight reverse repo [facility], and then above, you have the standing repo facility. So, the Fed has to reinforce its operating system. But you also bring up some other concerns. One is that the argument for— well, you don't need to know where you are on the demand curve.

Beckworth: So, in, 2019, I think that one could tell the story that the Fed accidentally fell back onto the downward sloping part, because a lot of things were happening and it was just an accident that you got there. But you raise, I think, a more fundamental point. You’ve got to estimate where the curve is. Yes, you don't want to get on a certain portion of it, but you have no idea where the curve is, because there's structural growth, and that reserve curve is actually shifting. It's not just moving along the curve, but it shifts in the curve. So, speak to that.

Estimating the Demand for Bank Reserves

Mester: Yes, so, one thing that we learned, and I don't think that we fully— I certainly didn't fully appreciate it at the beginning, was just knowing how much reserves are in the system doesn't inform you about scarcity, because the distribution of those reserves across banks matters, and banks, as you know, are not monolithic. Different banks are different sizes— one thing— but they also have different strategies. They have different risk preferences over liquidity. So, even though there's liquidity regulation, how they manage their liquidity differs across banks. So, you have that going on.

Mester: You have this heterogeneous banking system. And so, just knowing the aggregate amount of reserves isn't enough. You’ve got to make sure that if there are pressures on the interest rate, that reserves will get from the banks that want reserves, to those banks, from banks that seemingly have an excess of reserves. But you might be running a bank, I may be running a bank. We might look like we have the same level of reserves, but you consider it excess and I don't, because I know what I'm going to be doing with those reserves, and I may want to hang on to them.

Mester: What that means is that the demand curve is moving around. And so, what used to be considered maybe that inflection point— that isn't necessarily there. Even if the Fed's giving the same [amount of] aggregate reserves, it really depends on what stresses are happening in the market. And so, that's one thing. The other thing that Viral Acharya and others have worked on is that they've noticed that during episodes of quantitative easing— that's when the Fed is in there really supplying a lot of reserves— banks actually increase their demands on those reserves— demand on deposits and credit lines.

Mester: And then, when the Fed begins quantitative tightening— letting balance sheet assets roll off and, therefore, reserves potentially coming down— they don't reduce those demands on reserves. So, what you have happening is that the actual level of reserves that banks need and want goes up, but it doesn't come back down. And so, you could be mistaken about where you think that demand curve is because of that. That's what's going on. And they argue that that would have been part of the reason that, [in September 2019], it wasn't just that the Fed was in a scarce portion, but the demand curve had shifted as well.

Mester: And so, you get this, what they've called, demand for liquidity, but a lot of people call it a ratchet effect— that the more reserves that are put into the system, the more demand for those reserves goes up, and they don't come back down. And so, you get this sort of ratchet effect. So, that, I think, is also something that— even though the Fed doesn't have to estimate demand every day, they have to be really watching and monitoring carefully that they don't fall into a scarce reserves situation. And that takes a lot of effort. Roberto Perli, who is the director in charge of the Open Market Desk at the New York Fed, did a nice talk recently on, "Here are the panoply of indicators that we're looking at to assess whether we're getting near a scarce level of reserves." And it's not as simple as it looks, because you have to do a lot of monitoring, even if you don't have to estimate the demand curve.

Beckworth: Yes, the New York Fed recently put out a tool where they're measuring that slope, and it's a cool graph. [If] you go online, you can look at it. We'll provide a link to it in the show notes. The slope, right now, looks close to zero, but we really don't know. It's just an unobservable, and that's the challenge that you raise, and one other challenge, and then maybe we'll talk about potential fixes or directions the Fed could take [with] its operating system, and that is the over-reliance in the interbank market on the Fed to solve all of your problems.

Beckworth: And so, this is something that Claudio Borio has mentioned in his work, and many others have noted— Bill Nelson, too— is that there's a missing market. So, the federal funds market used to be a healthy market. There was price discovery. When you have unsecured overnight lending, you have to know what the creditworthiness of your counterparty is, and there's no need to do that. You just rely on the Fed. And so, I hate to say the word lazy, but you lose your muscle memory. How important do you think that is?

Addressing Over-reliance on the Fed in the Interbank Market

Mester: So, I think that's significant in the sense that you do want banks to be managing their liquidity, and you do want them to be active in the market. You don't want a central bank to substitute for the market. We're there to support the market when it needs support, but you'd rather not crowd out, and right now, you're right, there's not that much trading in the fed funds market, and the main lender in that market is the Federal Home Loan Bank System.

Mester: So, again, this is something that is concerning in the sense that you want healthy financial markets, and you want banks to have the responsibility for managing their liquidity, and not assuming that, if things become stressful or volatile, that the Fed will necessarily step in and add even more liquidity. And I think that is a legitimate thing. In fact, the Norges Bank, which is the central bank in Norway, and the Swiss National Bank— they both were operating a simple floor system.

Mester: And they, because [they] really wanted to encourage more bank activity in their overnight bank lending market— they stepped back a little bit from that. So, they do tiering in terms of what interest rate they'll offer on reserves with less interest at higher levels, precisely because they want to have a more active overnight bank lending market. And I'm interested to see how that's going to work out, because I think that that would be something that the Fed could imagine doing as well. When we were doing all of this, of course, we looked at all of the central banks and looked at how they were doing it. 

Mester: But I think that's something that the Fed could consider if they thought that making sure that the fed funds rate market remains a robust market was important. Right now, we're in a little bit of an interesting situation where we're communicating policy with the fed funds rate target, which makes perfect sense to me in terms of the communication, because people understand it. But it's not a robust market in the sense that there aren't that many trades. So, there's going to be volatility in that market. Then, the question is, do you want to take that on, or do you want to expand to another—

Beckworth: Like SOFR?

Mester: -or an overnight bank funding rate, which they publish now. That's a legitimate thing to be talking about is that maybe that— And that would be a step into broadening the number of trades that you're using to estimate that rate.

Beckworth: Just to flesh this out— A tiered operating system would be one where only some of the reserves are compensated with—?

Mester: Yes, I think the way that they do it is that it's basically by the amount of reserves. The interest rate varies. Then, there's some that are actually quota systems, that if you get above a certain level, then you're not going to get interest.

Beckworth: So, it incentivizes banks to economize on the reserves that aren't compensated at the central bank.

Mester: Correct. Yes.

Beckworth: Let's say that we were to go [in] that direction. Let's say that the Fed, at some point in the future, heads down that path. Would that revive the federal funds market? Would that get us back to where we were?

Mester: Well, it's hard to know whether we'd get back to where we were, but it would probably increase the number of transactions in that market from where we are today, which, again, is something that I think is worth considering. I would think that we would want to try to estimate whether that's a good thing or not. This is part of the issue about— and I said, specifically at the talk, that this shouldn't be read as, “We should go back to scarce reserves.” I personally think that the transition costs back to scarce reserves may not justify the benefits.

Mester: But I do think that it's wise to be very astute to the fact that the ample reserves regime doesn't give you everything that it's often touted to. It doesn't necessarily increase liquidity, for example. If banks are hoarding reserves or keeping the reserves in stress conditions because their demands on those reserves have gone up, it doesn't necessarily mean that you have a firm floor unless you add these other facilities. And it does open yourself up to more potential political pressures to do things that are not appropriate for a central bank to be doing. So, again, I just wanted people to know that things aren't as simple as they look— at least what the theory would tell you— and to be astute to these issues and challenges.

Beckworth: Bill Nelson, I believe, at the conference, and definitely in some of his writings, brings up a quote from Jay Powell, where he's thinking about the size of the balance sheet [and] where it could go. He goes, "Man, I hope I don't have buyer's remorse." So, Bill Nelson says, “I bet he has buyer's remorse now.” But what you're saying is, look, let's just make a modest change, and let's see what happens. That's what Norges Bank did [and what the] Swiss National Bank [did]. And even, I think, the other banks— more recently, the ECB, Bank of England, the RBA— what they call it is a demand-driven operating system. But in general, it's a gentle move in the direction towards more healthy interbank activity.

Mester: Exactly. And you're right, the ECB just made an announcement, I guess, I don't know how many months ago— But it's relatively [recently], maybe a year ago, about their new operating [system], and you're right. It's a demand-driven floor system. It's kind of like, let's get closer—

Beckworth: Yes. Baby steps.

Mester: -towards making sure that we're giving the right incentives for banks to actually operate in that market.

Beckworth: Okay, so, we know what Loretta Mester would recommend. Should there ever be an operating system review, that's her suggestion— something like a tiered framework. Let me throw out a couple of recent developments that could also affect the size of the Fed's balance sheet, and we'll end there, and I want to get your thoughts on this. There's been an increased push to use the collateral parked at the discount window to count towards liquidity regulations.

Beckworth: The hope would be that, one, there'd be less stigma. That's the primary motive, I think, from the 2023 ordeal. But it could also lower the structural demand for reserves. If you know that you have the ability to draw from the discount window and you're comfortable, it might make it more likely that you don't need to sit on a bunch of excess reserves. That's what, at least, Bill Nelson has argued. Bill Nelson has also argued [that we should] do that along with [bringing] back some version of the term auction facility, so that it's a regular part of business— maybe some collateralized lines of credit. Make it seem normal.

Beckworth: So, the discount window increased usage— might that lower structural demand for reserves? Then, the other suggestion— and this actually came up in your panel at the conference. Darrell Duffie has suggested this, and Lorie Logan had a speech this week where she brought the same point up, is to open up the standing repo facility to more counterparties via central clearing. So, the Fed's footprint wouldn't be dealing with everybody, but via central clearing, it would be able to maybe get more activity, which could also lower the demand. What do you see as the potential for those changes?

Loretta’s Thoughts on Central Clearing and Increased Use of the Discount Window

Mester: Well, I don't know about the potential for those. I think that the idea of the standing repo facility was to try to make it more accessible than the current— people that don't have necessary access to the Fed could use that facility. And so, that same thing with the ON RRP was basically allowing other institutions to effectively get interest on reserves. So, those things make sense to me. You'd have to see the details of exactly what they want to do. I think that the discount window stigma problem is going to be hard to solve.

Mester: It's been around for so long, and when you actually go out and talk to commercial banks, bankers, they'd much rather borrow in the Federal Home Loan Bank system. It's easy for them. It's better pricing, too, in some sense, that they get there. So, they'd beloathe, I think, to want to come back to the discount window. I've seen other proposals, which are like, let's not call it the discount window anymore. We can have a facility that's for banks that need the liquidity because of [the] issues that they're having.

Mester: Then, we can have this regular standing facility that could be called something different and [would] require different kinds of collateral, et cetera. That would really be part of the everyday running of the bank. But if you continue to call it a discount window, [then] you're not going to be able to resolve the stigma issue with it. And [with] the bankers I talk to, I think that sounds right. Because even though the Fed doesn't report who's in at the window— for two years, I think, is the lag now— They don't want to use it, even when the pricing is better.

Mester: So, again, you kind of ask yourself, well, why are they behaving that way, and would doing that actually alleviate the stigma, or would they then have to set up something even more arcane to try to get around that? I'm not quite sure it would work, but I do think that it's worth considering, ways of trying to de-stigmatize the discount window. Because you do want banks to come in. Part of the other issue, of course, is that, for the monetary policy purposes and the financial market purposes, we want banks to come in, but banks feel that examiners aren't always aligned with that. So, they also feel that, if they're in at the window, they'll get scrutiny from their examiner staff.

Beckworth: There's a tension, for sure.

Mester: So, you have that going on, too. So, that's why having a new facility, unless it’s something new, I think, might be--

Beckworth: A rebranding might be needed.

Mester: I like that, a rebranding. I hadn't thought of that.

Beckworth: Yes. Well, with that, our time is up. Our guest today has been Loretta Mester. Loretta, thank you so much for coming on the program.

Mester: Thanks very much for having me. It's been fun.