Monetary Policy and Central Bank Targets with David Beckworth (Mercatus Senior Research Fellow)

>> Jon Hartley: This is the Capitalism and Freedom the 21st Century podcast, an official podcast of the Hoover Institution Economic Policy Working Group where we talk about economics, markets, and public policy. I'm Jon Hartley, your host today. My guest is David Beckwourth, who is a senior research fellow at the Mercatus center and the host of the Macro Museums podcast.

Welcome, David.

>> David Beckworth: Well, thank you for having me on your program.

>> Jon Hartley: So, David, it's really fun to have you on here. I recently was on your podcast. You're a really big figure in my mind in the sort of macro policy space. I'm curious, how did you first get interested in economics?

And how did you get interested in economics podcasting? How did that all happen?

>> David Beckworth: Well, I took some econ courses in undergrad. It never dawned on me to become an economist. This didn't strike me as something I would do as a career. But later on, working on an mba, I took some more coursework in economics and it really began to click like, I love this.

And so at that point, I went back, took some more coursework, some more math, got into a PhD program, went from there. Now, how I got into podcasting. Long story short, I started blogging back in the glory days of blogging the right after the great financial crisis and great time to be alive as a macroeconomist blogging.

Terrible time for humanity suffering through the slow recovery from that period. But the blogging opened up doors, op eds. I actually got some books out of it. And eventually that opened up more doors to where I got offered a job at the Mercatus center to be a scholar there in their policy group.

And then when I got there, I just thought, there's a missing market for podcasts that touch on macro and regulatory issues in the financial policy space. And so I pitched the idea. It was a little reluctant at first. Folks didn't think there'd be a market or appetite for such things.

But it's gone well. We've been running it since 2016. You've been a guest, a number of guests. We're getting close to their fifth hundredth episode, so so far, so good.

>> Jon Hartley: It's amazing. I enjoy listening to it and I know many, many others do, and I think it's fantastic, too.

You spent a good amount of time working at treasury previously. I guess you've always been interested in macro issues. I know you did your PhD at University of Georgia, and I think George Seljen was on the faculty then. I feel like this is maybe what you were called to do, and it's so wonderful to see all these wonderful shows being produced.

I want to talk about, I think some of the themes that I think are consistent in your w some of the things you've advocated for throughout the podcast and in your academic writing as well. I think one of the big things that you're very early to was NGDP targeting.

And I'm just curious, could you explain what NGDP or nominal GDP targeting is to our audience as you see it?

>> David Beckworth: Yeah, let me provide a little history how I got into nominal GDP targeting, because I think that's useful. So as you mentioned, I went to the University of Georgia and George Selen was there and he's the one who actually first introduced me to nominal GDP targeting.

And at the time he was actually selling it as a way to deal with big positive productivity shocks. So if we have massive productivity surge like we had in the early 2000s, late late 1990s, early 2000s, what's the best way for Monte policy to deal with that? And he argued he had a specific form called a productivity norm, but basically it was a form of nominal GDP targeting.

So it was a totally different set of problems, but that's where I first saw it. And then it became popular again after the 2008, 2009 period. But here's what it is. Nominal GDP targeting is an approach where the central bank would target total dollar spending or equivalently total dollar income instead of targeting the inflation rate.

Now over the medium to long run, you would get a inflation rate with a fallout of that. Because by targeting total dollar spending or nominal GDP, you're implicitly targeting real GDP and the price level as well. And if you're targeting some kind of growth rate, you're targeting the growth rate of real GDP and an inflation rate.

So what's different about a nominal GDP target versus an inflation target is over the short term, that's the big difference. Over the med long run, you're probably going to be very similar in the outcomes. In fact, if you look at US GDP nominal GDP measures between crisis, they look pretty straight.

It looks like the Fed is implicitly targeting something like a nominal GDP target, but where they differs in the short run, the short run, the Fed is much more sensitive to movements in inflation. In the short run if they were targeting nominal GDP, they would keep their eyes focused on the growth path of total dollar spending.

And I've mentioned nominal GDP. Some would measure with final sales domestic purchasers. There's even a proposal to target just PCE a component of nominal GDP. But effectively you're aiming to stabilize aggregate demand directly as opposed to a symptom of it. The Feds, when it targets inflation, it's ideally trying to target inflation being driven by demand and demand shocks and it tries to see through supply shocks that cause inflation.

So what we're saying is cut to the chase, target directly aggregate demand and do so by looking at nominal GDP.

>> Jon Hartley: So for the viewers that aren't as familiar, so you've got nominal GDP, you've got real GDP. Real GDP is basically nominal GDP minus inflation. So in other words, nominal GDP is real GDP plus inflation.

And if you look at the post World War II US economic era, you'll see that real GDP I think is growing at 4% for a long time, 3% and 2% more recent years or more recent decades. And then NGDP or nominal GDP is roughly depending on, you know, the inflationary period.

You can add say 2% to that perhaps obviously much higher during periods of higher inflation. So I'm just curious like what is the exact. So we talk about it as like this is a target the Central Bank since the early 1990s have been targeting inflation and prior to that for a brief period of time in the 80s.

They were targeting monetary aggregates like M1 and m2 o how much money is there in the economy. But for the past say 30 years or so, central bank's been targeting inflation targets. And I think some people, including myself, would argue that it's been a very successful program in terms of anchoring long term inflation expectations.

And also inflation has been relatively speaking much lower in the past few decades, minus the past few years or so. And we could talk a little bit about that as well. But I'm just curious like in your mind, what is the exact policy tool that allows the central bank to achieve an NGDP target?

I mean we had like interest rates at 0% for a long time in the 2010s and we did lots of QE as say Japan is getting like. I think getting NGP to a certain rate or level target, say a 4% or 5% trend path. I feel like couldn't, couldn't have been possible in the 2000 and tens.

Even if you tried harder. How could you have gotten NGDP to a higher level? Say when you're at the lower bound. I'm curious, how do you think an NGP target is really achievable in your mind?

>> David Beckworth: If you look at nominal GDP from the end of the Great Recession up until 2020, the beginning of the pandemic, it looks like a stable 4% growth path.

If you didn't know those two periods, the before and after, you would think the Fed was effectively. So the Fed did do something like this. Now, you could argue it should have been faster 5%, that's far fine. But I would say that concern is a concern that happens no matter what.

You're at the zero lower bound, whether you're targeting inflation, a price level target, a nominal GDP level target, the same issues are going to be there. And I actually did write something in response to this concern and it would apply to an inflation target as well. This was at a, I presented this at a Cato monetary policy conference in 2019, and I recommended kind of a dual approach.

When you're at the zero lower boundary, you follow something like a McCollum Rule. A McCollum Rule would actually you would, instead of thinking about interest rates, you would adjust the monetary base. And a McCollum Rule, by the way, was a very popular rule up, really up until the Taylor rule comes and displaces it and the adoption of inflation targeting and interest rates.

But the McCollum rule, which was well known in the 1980s early 90s, says that the central bank should target nominal GDP but use the monetary base as an instrument. And a lot of people came to that same conclusion once we hit the zero lower bound, yeah, maybe we should rely more on qe, the balance sheet.

Now we can raise questions about Wallace neutrality policy and effectiveness proposition, things like that. But in my paper, I argued what you'd want to do is you want to do something more like a McCollum rule where you think about adjusting the base monetary base. When you're above zero, you're outside the zero lower bound, you go to a more standard Taylor rule, but a Taylor rule that has nominal GDP in it instead of the regular arguments.

I also argued in that piece, when you fall into a McCollum rule world where you're at the zero lower bound, you want to have a fiscal facility to give it the umps, the very thing that you're concerned about. So I argued, you got to have fiscal policy hand in hand with monetary policy at the zero lower bound.

That's what the Fed needs now there are all kinds of problems that come with that because once you give the central banks extra fiscal ability, then you get into questions of delegation of authority. Is this democratically accountable? But I will leave it at that. That the same issues apply to the inflation target, would apply to a nominal GDP target when it comes to zero lower bound.

>> Jon Hartley: Well, I do think one difference though is that you're talking about a level target versus say an inflation rate target, right? And so I think something that the Feds, I guess maybe thought a little bit about in the past. But the thing about a level growth target is you can get these issues, things like hysteresis.

For example, after the Great Recession we had a massive decline in output or in real GDP and nominal GDP and it never got back to its pre trend like that, pre, say 4% or 5% NGDP trend. Now after having the big collapse, there was a pretty straight 4 or 5% target.

How do you think about that? Do you reset the, like what is the target exactly? Or what's the reset the target after a big drop like that or do you try to get back to the original trend? How do you think about issues like hysteresis and things like that?

>> David Beckworth: That's a great question and really two responses and it's driven by this question. Was that collapse something that could have been fixed by policy, monetary, macroeconomic policy? If your answer is yes, then I would have argued more aggressive monetary and fiscal policy could have brought us back up some kind of helicopter drop.

Something like we saw in 2021 where we saw a complete collapse of the dollar size, the economy. But we saw a quick bounce back and I think that partly had to do with the fact that it was supply shock. So there's going to be some quick recovery, but then also the incredible excessive amount of fiscal support as well kind of closed that gap and then went above where it should have gone.

So if you believe that the collapse after the Great Recession was something that could have been fixed with policy, then yeah, you use aggressive policy. If you think it was a permanent structural change that wouldn't have mattered, then this is where I would go. And this would be also how would you deal with a nominal GDP level target?

Let's say there's no crisis but potential real GDP changes. And this is often a question I get, well, how do you know? Okay, sure you embed 2% inflation in that, but what if potential real GDP changes? What if it grows from 2% right now? To 2.5 to 3%.

Or in the case you just suggested, what if it goes from 3% down to 1%, some big shock in that case, you have to update your target. Now, generally, changes in potential rules GDP are slow moving. Generally it's a very gradual adjustment of the target. So it's not gonna be something that's terribly hard to think about or to process or even to implement.

And what I would recommend is you would follow some kind of consensus measure where potential rule GDP is going. Now, going back to 2008, my view is that that collapse could have been avoidable. I think we could have returned back to the same growth path or level. I think it would have taken more aggressive macroeconomic policy.

It would have required the Fed being aggressive as it was, but also announcing a level target. I think level target goes a long ways. It's a signal. In fact, level targeting was something really big in the New Keynesian literature, as you know, Michael Woodford, Gut Ergensen, and, and for them it was forward guidance.

Wait for the march,

>> Jon Hartley: I guess, right?

>> David Beckworth: Yeah, except just to get things up. In fact, all of that Princeton School, Scott Sumner has a paper called the Princeton School of Macro. All of them were thinking about Japan. Krugman, Sinson, Gotti, Ergensen, Michael Woodford. They were all there at that time thinking long and hard, how do we get out of Japan?

And all of them were very keen about a permanent increase in the price level, getting the price level back up. And many of them would talk about, well, we need to have a permanent increase in the monetary base. Well, the thing is, what is a permanent increase in the monetary base?

That's effectively. It's a non Ricardian fiscal regime where you promise not to raise taxes in the future. It's a policy move where you do not change what you're doing. What happened after 2008? There was a commitment to shrink the Fed's balance sheet. There wasn't the fiscal support. I think a policy difference could have been different and we could have returned it. But over time, as time goes on you come to realize, okay, that's not going to happen. We're at a new norm. You adjust your target. And that's one reason, Jon, I came up with this measure called the nominal GDP gap, or a neutral measure of nominal GDP. It's based on consensus forecast and it gradually does because if you simply draw trend lines, at some point they become outdated.

Eventually expectations change, people update their views, contracts change. And so you want to have a flexible understanding. And I think that's true with a nominal GDP target. Eventually potential real GDP will change, and so you gradually, slowly update your your nominal GDP target. But again, there normally wouldn't be big swings.

>> Jon Hartley: How would you deal with the fact that like, I guess a GDP is published like 2 months after the end of the quarter. Unlike say, unemployment or CPI, they're published pretty close to the reference month. How would you deal with that fact? I know some people talked about NGDP futures markets and things like that.

Like that. How do you deal with the fact that I guess, you know, you wouldn't even get NGDP estimates in a real time kind of fashion. I mean maybe you want to get the BEA or start publishing it more quickly, but I don't know if how feasible that would be.

>> David Beckworth: Yeah, great question. Probably the two biggest questions I get about nominal GDP targeting, it's that one, the data revisions are huge. One, they come out, like you said, delayed and secondary, sometimes huge. And the other one is the potential world GDP, that one. So I think I've addressed the potential world GDP.

This one I think is probably more of a challenge. But there's I think several responses to it. You, you mentioned what I think would be the gold standard and nominal GDP futures market, but that's nowhere in the works. I don't want to bank on that. I want to bank on reality.

What could we actually do? So what I, what I suggest is again going to a forecast, going to consensus forecast or where nominal GDP is going. And even if the measures are awful over time, a long consensus forecast is very helpful. You mentioned Lars Findsen, he says target the inflation forecast.

So it's useful and we provide some workaround to that. Again, this nominal GDP gap measure I have created, which also implies a neutral level nominal gdp, I have a real time measure versus the ex post. After all the big revisions, even the five year revisions, and they aren't that big, there would have been some periods where there have been differences, but again the big differences wouldn't have made much significance in terms of vast changes in setting the interest rates, things like that, because I do tie it to a policy rule, a reaction function.

So I would say target the forecast is probably the easiest thing to do, number one. Number two though, I think there's also the ability to rely on many multiple real time measures. There's, there's monthly real time measures. You could put that together with a forecast of inflation. This is just one example and I'm sure there's more out there.

But Lars Christensen, who's also an advocate of nominal GDP targeting, he for example has tinkered around with this weekly economic indicator series. Used to be the New York Fed and I believe the Dallas Fed now publishes it. And he takes that, and takes an inflation breakeven forecast and combines those two together and kind of gets a real time sort of forecast look at where he thinks the economy is going.

He does it every month. I mean the Data for the weekly series is weekly on the real side. So there's, I think there's many ways to do that. There's real time indicators you could use as proxies. PCE comes out monthly. That's why some have advocated that there's revisions to that as well.

Of course there's the market based version of PCE you could rely on. But I would say, you know, use the real time data sources available. So number one forecast, number two, more real time. And the third point I would bring up you touched on and I would say that the data is somewhat endogenous to the target.

If you were to adopt a nominal GDP target, I guarantee the Fed would find ways spend resources on getting data that would better inform it on what's happening to the, the current measure. Again, there's different ways to measure it. May I should be very clear about this. I said nominal GDP or total dollar spending.

You could look at final sales, others have just looking at gross labor income. Some have looked at pce. So there's different ways to kind of slice this up but, and each one may have certain advantages in terms of data collection. But I do think the data challenge is surmountable and even with the data we have now we, I think there's ways to use it. We have a framework review coming up, as you know, and we've written a number of policy briefs. And my policy brief, I definitely would love to see the Fed go all in to a nominal GDP level target. I know they won't. So my recommendation has been, many others, is simply use something like forecast nominal GDP as kind of a cross check.

Use your standard indicator, labor market indicators, things like that. But also look at forecast of nominal GDP. Plug them into a reaction function, see what they say. You were at the American Economic association meetings. Jason Furman was there. He recently talked about, he's finding his religion again for monetary policy rules, you could have a rule that has nominal GDP forecast in it. And again you don't have to go all in but you use it as a cross check.

>> Jon Hartley: Interesting. Yeah, I mean it's interesting, I guess, that the proposal on rules is really using them more like a guideline or a benchmark or something like that. That's interesting thought. I mean, correct me if I'm wrong.

I don't think we have any central banks around the world yet doing NGDP like targeting. I don't know how close we've come. I know Bernanke has said some positive things maybe once or twice in the past. But I'm just curious, assuming that we're not adopting NGP targets anytime soon for maybe some of the challenges that we've sort of outlined, I really want to talk about maybe just inflation targeting and its kind of different variants for a little bit.

You've been doing a lot of really great work, along with other scholars that Mercatus getting prepared for the next Fed policy framework review. And I'm curious what your thoughts are on the flexible average inflation targeting regime that we've been in for nearly five years or so. That was sort of really the brainchild of former Fed Vice Chair Rich Clarida.

This was a framework that they put out really right before the big inflationary spiral that occurred in the early 2000s. In part, I think the flexible average targeting was in part responding to this period where the Fed wasn't able to hit its 2% inflation target for many years in the 2010s and was coming in below.

But you know, at some level, I think there's maybe, I don't know, I feel like a little bit of regret in the sense that, you know, I think that the fate or flexible average inflation target was a bit flawed in that, you know, it wasn't really clear, you know, if we were going to take some average over some period of time, it wasn't clear, you know, what period of time or years we were averaging over.

You know, the Fed generally, historically has followed, you know, a year over year or one year look back window. Right. But how many years back would it be? How flexible would it be? So I think we're all sold, or at least I'm sold on inflation targeting in general.

We could debate what exactly the right target should be.2% or. Paul Volcker, for example, was pushing for a 0% inflation target the rest of his life after leaving the Fed chair or leading the Fed board, as did other central bankers. In fact, in the early 90s, John Crow at the Bank of Canada argued for 0% inflation target too. But in terms of, you know, the F and the a, you know, averaging, thinking about levels versus rates of change and where do you fall on all that? If we can't have an NGDP target, we have to stick to inflation at some level. How do you think about that?

Do you think that the Fed's flexible average inflation targeting regime's been successful or do you think that going forward the Fed will abandon it and so will monetary policy scholars?

>> David Beckworth: Well, based on what Jay Powell said to Catherine Rampel at The Washington Post back in November, he had a sitdown with her and she asked him about the review.

And he said his base case scenario is a reaction function, ie, a Taylor rule that does not promise to make up for past misses or promise to overshoot Shoot the target if they've undershot, which I find a little troubling because that to me rules out makeup policy altogether. And I guess the one thing I would encourage FOMC members, if they're listening to this, is we wanna maintain, hold on to makeup policy. I do think fate was not the best way to do it, but fate did introduce it. I will say this, FAIT I think is an historic sea change in terms of monetary policy.

It's the first time a central bank has talked about targeting something like a price level. Now, I should take that back. I believe the Reichsbank in Sweden did price level targeting explicitly for a while, 1930s, maybe 1940s. It's a while back. But modern central banking, the Federal Reserve and FAIT is the first central bank to incorporate several decades of research literature about zero lower bound about makeup policy board guidance in terms of a framework that I think you could also call temporary price level targeting.

The Bernanke, a version of it, Rich Claire to call it a version of it. So I think it's a step, it's progress. I would do things a little bit differently. And my worry is that they're gonna completely throw the baby out with the bath water. At least that's the signal they've sent.

And, and every place I've been, I've heard FOMC members say, yeah, bait was designed for a different period, which is true. There's been several conferences on this at Hoover, multiple events at Brookings. They've talked about this, papers have been written. In fact, the paper that I wrote up, I kinda summarized what I thought was the consensus from all these different talks.

But there were a number of critiques of FAIT. And maybe we should step back and just remind our listeners what actually happened in fate. So prior to FAIT, flexible average inflation targeting, the Federal Reserve and FOMC had FAIT fit flexible inflation targeting, which was officially introduced in 2012.

So there was a consensus statement. In that consensus statement, it outlined 2% PCE inflation. That statement, though, also had in it the other part of the dual mandate, maximum employment. And it talked about deviations, no numbers, but deviations. So it had both sides of the dual mandate. So you can call it a flexible inflation target because it's aiming for 2% inflation. But it also has its eye on avoiding business cycles, full employment. We go to 20.

>> Jon Hartley: It was the first time that the Fed had ever actually announced that it was targeting inflation. Yes, it was very late to the game many of the other big advanced economy central banks had announced this by the 90s. 2012 was actually relatively pretty late compared to the other central banks.

>> David Beckworth: Yeah, and to be fair to the Fed though they were implicitly doing something close to two by the early to mid-90s least some studies have found that they were there, but not officially there. Like it took Ben Bernanke coming on board and really talking them into it. But there, there are discussions in the transcripts back in the 90s about should we go to an official target. But you're right, 2012, but that consensus statement becomes kind of like a constitution for it. And this is again another reason why I'm a little worried about them completely throwing out the makeup part.

So 2020, they changed it quite radically to include there's gonna be makeup policy on the inflation target if it's below, not if above. So if you're below, you make up from below and then once you go above, it's just regular flexible inflation targeting. So that's very much in the spirit of Bernanke's temporary price level target.

You try to maintain the price level path, but you only do so if you're below. If you're above, you let bygones be bygones. The other big change was on the maximum employment part. Instead of having deviations would be something the FOMC would be concerned about. They then said shortfalls.

So only if the economy was, was weak or cold relative to potential would they respond if it was overheating off the table. At least according to the language of FAIT. Of course, the Federal Reserve abandoned those principles in 2022 when labor markets were clearly overheating and they started raising rates.

So the FAIT introduced those two asymmetries. They would only do makeup policy from below 2% and they would only respond to shortfalls from maximum plum. That was a big sea change. Now, the challenges in those one I think a fair critique. A lot of people, Gotti Ergenson and Don Kohn have a paper, a number of people have written along this.

The Roamers had a paper recently and in one sense fate was fighting the last war, the last decade of zero lower bound. You want a framework that can handle any situation, not just the low inflation. That's, that's fair, totally fair. Also there's, there's concerns about what does it mean to be makeup policy doing makeup, how all, how high, how long above before we go back to two what is a shortfall and how much do you respond.

So there's a lot of uncertainty and there wasn't much clarity on that. There's also some concern that this was putting more weight on maximum employment. Both the roamers and again Gotti Ergensen Don Kohn said FAIT effectively put more weight on maximum employment than the priceability part of the dual mandate. So it added an inflationary bias, which is fine if you're in a zero lower bound world, but if you're outside of that, you don't want that. So there's definitely room to recalibrate this, make it more effective. But I do worry if you completely throw out again the baby with the bath water, I'll make a policy that they'll have gone too far.

So what I recommend they do is use nominal GDP level targeting as a benchmark and any incremental change you make, have it move in that direction again. They're not going to go all the way there. But so here's what I would recommend. I would recommend getting rid of the shortfalls part, go back to deviation so it's both above and below. And there's been some other great work, Michael Kiel.

>> Jon Hartley: A symmetric target is one thing that you want.

>> David Beckworth: Yes, yes, a symmetric target, Michael at least on the, on the maximum employment side, Michael Kiel did some work where he showed using models that if you respond systematically to shortfalls, ironically you make more shortfalls in the future because you're kind of not responding to the overheating and you can actually exacerbate the business cycle.

If you respond symmetrically, then things can actually be better. On the inflation side, I don't think it's great that they respond from below only for communication purposes, but I do get why they would do that and I would recommend at this point to keep, keep that part. And let me go back to the original again, motivation idea behind this as I understand, and that's Brandon Bernanke's temporary price level target.

>> Jon Hartley: Which is metric target to some degree.

>> David Beckworth: Yeah, well, it's very similar to FAIT. It just doesn't have all the language about the maximum employment. It also says if you're below 2% below your target, you run the economy hot. You actually have inflation above your 2% until the price level gets back on its original trend path, but you don't.

>> Jon Hartley: The paper for those that are listening, this is a Brookings paper that Ben Bernanke wrote in the late 2010s. It was very influential prior to the Fed.

>> David Beckworth: Yeah.

>> David Beckworth: And the reason he gave for why you would only do it from below, which is what fate is is because he said, look, what if you have inflation above 2% and it's driven by negative supply shocks?

So the economy is contracting already, inflation is going up. Why would you want to undershoot for several peers? You'd add more, more harm to the economy on top of the negative supply shock already. And that I agree with that. And that's exactly why you would want to do something like a nominal GDP target.

I didn't mention this earlier, but one of the big reasons for nominal GDP targeting is it's a way to work around supply shocks. We don't know in real time whether inflation is caused by demand or supply. We maybe we can figure out, look at the news, but it's really hard, it's not clear in the data.

And we know 21, 22, all this talk about is this transitory inflation or permanent? That's another way of saying is this. Driven by temporary idiosyncratic supply shocks or just something more permanent and we simply don't know. So instead of trying to play God and figure that out, look at the total demand in the economy.

Keep total demand, total spending stable and those shocks will work themselves out. And if, and if you're way above the trend path, whatever your target may be for total spending, then maybe you have some issues. So the supply shock was a very reason Bernanke had a below asymmetric price price level target.

And I say look, that's great, why don't you go all the way? If you go all the way, take the logic to its limit, you end up with the nominal GDP level target you mentioned. Bernanke endorsed it. Christina Romer endorsed it in the 2011 New York Times. She even had an op Ed said Ben Bernanke needs his Volcker moment.

And in the op Ed she said he needs to go all in and adopt a nominal GP level target. Michael Woodford A lot of people have endorsed this for a zero lower bound world. There's other arguments more recently that we could talk about if you want for, for doing that as well.

But back to fate. I, I just hope they don't completely throw out everything and go back to fit flexible inflation target because I think there's some things we want to keep interesting.

>> Jon Hartley: Well, I know I, I feel like the Fed certainly, I think before any kind of framework, any meaningful framework change I feel like has to sort of get inflation back to 2% and sort of to kind of ensure its credibility before really thinking too much about maybe further changes.

And I know there's some people out there who advocate for a 3% inflation target. But I think getting inflation back to 2% in a sustainable fashion is kind of first and foremost, my last sort of question or sets of questions here is really about Fed plumbing. And you've written and spoken a lot about this through your podcasts and this gets back to the kind of abundant reserve system and how much 2008 and the first rounds of quantitative easing really changed how the plumbing around the Fed and the short term money markets really work.

So one thing that's happened is we used to be in the system, some might call reserve neutrality or a scarce reserve system. And that was a system where you do open market operations that would move the Fed funds market and that's all the Fed needed to do. But since 2008, because the money supplies has been shifted so far out to the right or we have what's called an abundant reserve system.

Now the Fed, in order to raise interest rates, has to actually use price floors, so things like interest on excess reserves, the reverse repo rates and so forth. And that's how it's able to raise interest rates without unwinding all the quantitative easing that it's done in the past.

And so the Fed's been in the system where it's now been able to both adjust interest rates while also having a Fed balance sheet that has trillions of dollars of long term assets on it, lots of quantitative easing. Part of this also changes the system of from those that are familiar corridors and floors, and this kind of gets back to the whole reverse repo IOE up versus IBeer rates being part of, say an abundant reserve system.

Can you explain us, what the difference between corridors and floors are and can you explain sort of what's going on now and that with Fed or sort of with various central banks winding down or trying to wind down their balance sheets now that we're sort of in better times now and have been trying to escape inflation. Can you explain what the shift that's kind of going on now where central banks are really trying to get away from the abundant reserve supply driven system to some degree, and maybe with the exception of the Fed trying to incrementally get back to a demand driven system and maybe not fully a quarter system. 

But can you explain what all this means and where things are at and where things you think are going?

>> David Beckworth: Yeah. So you've covered a lot of ground there. Prior to 2008 in the US and in many places around the world for advanced economies, most central banks were operating on some version of a corridor system.

So it's also called scarce reserve system. Bill Nelson calls it a necessary reserve system. But very few reserves were actually in this system, at least compared to today. Very few reserves in the system also compared to the currency. So in terms of central bank liabilities is mostly currency, a little bit of reserves.

And the interest rate that was set on those reserves was determined in some interbank market. Banks would lend their excess reserves to each other. And in the case of the US it was the federal funds market and all that the Fed would do, the Fed would adjust the size of its balance sheet, inject or pull out reserves liabilities, and that would move the supply up and down a downward sloping demand curve for reserves.

So if the Fed wanted to target 5% interest rates, it would adjust supply and then you would be on that. And it required very few reserves, it was scarce. As you said, at the other extreme, where we are today, you have a floor system where you increase the supply of reserves so much you get to a part of the demand curve for reserves that's flat.

And when you get to that point, no matter how many reserves you increase, even decrease some, it's gonna be at that level. And typically that kicks in during a zero lower bound environment. But when you add administrative rates, you mentioned interest on reserves, you also mentioned overnight reverse repo.

That's kind of like the leaky floor system. Sometimes market rates don't stay at that interest in reserve. So the Fed had introduced another way to maybe clog that floor overnight reserves. So the Fed can prevent risk from going down. Now on the top, it also has interest rates from it going too high. It has a discount window rates, it has the also the standard repo facility. So the Fed has these facilities that try to keep rates within it. And given that their administrative rates, it's in a very narrow window. And if in a perfect world we didn't have market frictions, it would be like a straight line. But it's not quite that way. There are arguments for both systems. So the scarce reserve system, you have a very small balance sheet. So the financial footprint of the Federal Reserve is much less. You also have a close link between the size of the balance sheet and monetary policy itself.

So if you wanted the Fed to again change policy, you would have to increase or decrease reserve, change its balance sheet. Today that's not the case, you noted this. The Fed can simply change an administrative rate without changing the size of its balance sheet. Now for advocates of the floor system, that's a big deal because then the Fed can inject liquidity into say, a strained banking system without affecting the stance of monetary policy via interest rates.

For advocates of a corridor system, they'd say that might be a bad thing because if you have a system like that, it becomes a temptation to politicians. So you're telling me, Fed, that you can keep interest rates pegged and increase your balance sheet, but why don't you buy up a bunch of bonds to build a wall or to fund a green New Deal?

And so the balance sheet, if it looks like a free lunch, it's really not. But as you know, is a good economist, but to many politicians it becomes a free lunch. So there's arguments for and against. There's also questions of what's easy. Advocates of the floor system said this will be super easy because we're using administrative rates.

Hasn't turned out so easy as many expected. Also, when you have a corridor scarce reserve system, it relies much more on banks to fund each other. And we have largely lost that in the US. The federal funds market is a shell of itself. And this is a nice transition to the second part of your question.

Other central banks around the world are actually gradually moving back. They're a long ways from a corridor system, but they're gradually moving away from supply driven kind of ample, abundant reserve system. And their motivation has been twofold. Number one, there's been a lot of losses on central bank balance sheets and so the fiscal authorities, parliaments, they're worried about that.

But secondly, and particularly in the case of ECB, because I talked to Isabel Schnabel on the podcast about this, their motivation for going back was to resurrect the overnight unsecured interbank market where banks would lend to each other unsecured. Well, if you're going to lend unsecured, you better know your counterparty really well.

So there's price discovery, there's, there's learning about new risk that are, that currently is missing. You don't find this in any other market and they, they want to resurrect that. Now again, we're not probably going to go to balance sheets the size they were before 2008 but in my mind that's a great step.

We're going back, we're getting the markets more involved. Banks rely on each other less. On the central bank for liquidity in the US it hasn't happened, I think in part because of some of the financial regulations. I think there's just some complacency. I think you get the right people at the Fed though, things could change.

And I think ultimately you alluded to earlier, we were slow to adopt inflation targeting. The rest of the world was on that bandwagon. I think eventually we'll, we'll have to come around as well just because the cost of maintaining large balance sheets are just going to continue to grow over time.

And I think doing more moving in direction of something between a floor and a scarce reserve system, a corridor system might be useful. The central bank of Norway, the Norges Bank, it has something called a tiered reserve system. Loretta Mester has advocated something like this where some of the reserves are remunerated.

They do get something at the bank, but many of them do not. And so it's kind of a compromise. I think that's where the world is going, and I suspect with enough time, the Fed will be there, too.

>> Jon Hartley: Interesting. I mean, it's fascinating. We've been in this period for almost 20 years of era of big balance sheets, big central bank balance sheets, lots of quantitative easing, abundant reserves.

It's interesting to see now central banks really sort of trying to think about some sort of an exit strategy, the extent that it's possible. And I still don't think we're necessarily getting back to where we were prior to the global financial crisis anytime soon. But it's very interesting to hear about these sort of alternatives that other central banks around the world are doing.

A real honor to have you on, David, and hear about your career and ideas. I know there's so much we could talk about and I would love to have you on again at some point. I really want to thank you for joining us and really for the great service that I think you've done for the economics and macroeconomics and finance community with all the wonderful podcasts and writing that you've been doing over the years.

>> David Beckworth: Well, thank you for having me on your program.

>> Jon Hartley: This is The Capitalism and Freedom in the 21st Century Podcast, an official podcast for the Hoover Economic Policy Working Group, where we talk about economics, markets, and public policy. I'm Jon Hartley, your host. Thanks so much for joining us.