Peter Ireland on the Fed’s Pandemic Performance and the Path Forward for Monetary Policy

Peter Ireland is a professor of economics at Boston College, a research associate at the National Bureau of Economic Research, and a member of the Shadow Open Market Committee. Peter has also been a visiting scholar at numerous Federal Reserve Banks and is a returning guest to the podcast. He rejoins Macro Musings to talk about U.S. monetary policy during the pandemic and what the path forward looks like for the Fed and the policy landscape. David and Peter also discuss the current state of macroeconomics, including the most influential and popular business cycle theories, the present direction of policy macro, and whether or not the Fed’s current framework should shoulder blame for its pandemic policy missteps.

Read the full episode transcript:

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

David Beckworth: Peter, welcome back to the show.

Peter Ireland: Thanks, David. Thanks a lot for having me back.

Beckworth: Well, it's great to have you on. I was checking, it's been about five years, so back in 2017. So you were on the show, just in the early stages of the podcast, so you helped make it, so thanks, Peter. But also, looking back, man, a lot has changed since then. 2017, what were the issues we worried about? Well, the Fed was beginning to tighten… normalization, we were still wrestling with low inflation, and here we are, five years later, we have the opposite problem, we have high inflation, uncomfortably high inflation. So what has changed since then?

Ireland: I know. Well, when you were describing this situation, I think we have to file it under B for be careful what you wish for. The old days don't seem quite so bad anymore, the problems don't seem quite as bad anymore. Yeah, here we are back in an environment that for many of us is reminiscent of the 1970s. Inflation moving higher, the Fed behind the curve, the painful trade off. Do you tighten more quickly to bring inflation back down with the risk of sending the economy into another recession, or do you move more slowly and hope for the best on inflation? I think they're in a real jam right now.

Do you tighten more quickly to bring inflation back down with the risk of sending the economy into another recession, or do you move more slowly and hope for the best on inflation? I think they're in a real jam right now.

Beckworth: Yes. I like to call this the Fed's delicate dance. They're trying to reign in inflation and engineer a soft landing, and that seems incredibly challenging, given all the pressures going on at the time right now. But yeah, we've seen some big changes, the size of the Fed's balance sheet. I know you've done work on this, it's close to $9 trillion. We're recording this the week of the Fed's meeting, and we expect them to announce they will start actually shrinking the Fed's balance sheet. They got $9 trillion there. If you look at the size of the balance sheet relative to GDP, it's gone from about 18% to 35%, a sizeable increase.

Beckworth: Also, I think it's important to look at the liability side closely on the Fed's balance sheet, and there we see about $3.6 trillion in bank deposits or reserves, another $1.8 trillion on the overnight reverse repo. Those are liabilities for which the Fed has to pay interest payments on to banks and money market funds if they're in the overnight facility. As rates go up, those are going to get larger and larger. You can make the argument banks might be holding treasuries in another world, but the fact is it's going to look really bad as the Fed does it. The Fed's balance sheet is something, I think, that is going to be increasingly a polarizing issue, a political issue for the Fed. They have a lot to wrestle with, inflation, their balance sheet, engineering a soft landing. We're going to get into that today with you. I'm glad I have you, because you are a true monetary economist. Not many macro economists can say they're a truly a monetary economist. You take money seriously, as well as all the other parts of macro.

Beckworth: Before we do that, Peter, you're also an academic, you're in the profession, and I've had a lot of requests to sometimes step back and just take a look at macroeconomics. What is the state of it? What's graduate education like these days? What are the models people consider? So let me start with a very basic question, and that is, what is macroeconomics, and how would you separate it from microeconomics?

The Current State of Macroeconomics

Ireland: Yeah, that's an excellent question. I'm not sure I can provide deep insights or controversial statements for you along those lines. I would just start by saying that you could just use what the words tell us, macroeconomics outcomes at the level of the economy as a whole, microeconomics, studying economic behavior and outcomes at the level of, say, the individual decision maker or at the level of the individual market. But I would also say, interestingly, back in my day at Chicago, we didn't call them micro and macro, we had price theory, and we had money. And then you think about topics that are in between, say economic growth. I think most in the profession would say economic growth, file that under macro.

Beckworth: Yep.

Ireland: But Becker at Chicago was making important contributions to the field of economic growth through the theory of human capital accumulation, a very microeconomic approach. At the same time, Romer was there thinking about technical change, market structure, and the macro effects of those phenomena. So I guess my own view might be just to say that maybe we make too much of the distinction, good economics as good economics, and every problem can be viewed on a variety of levels with a variety of models. I'd like to see it done all ways.

Beckworth: Okay, Peter. So you've outlined macroeconomics, and you've brought up a part of macro growth theory, long term growth. Both of us tend to spend most of our time in the business cycle part of macroeconomics, but what do you think is more important, if you had to draw a line or had to cast your lot? Is growth more important? Is business cycles? I once had a prominent economist I won't name tell me that he thought business cycle considerations are boring and irrelevant, and that I need to focus just on the long run issues. But I think one could make the argument that the short run considerations affect the long run growth argument. So how do you approach that question?

Ireland: Yeah. Well, maybe what your other guests had in mind, or I think this is what a lot of us have in our minds when we hear growth versus business cycles, it was those calculations that Robert Lucas did in his book, where he showed that the welfare costs of reducing the rate of economic growth by a fraction of a percentage point, when compounded, looking out into the infinite future, were huge, whereas the welfare costs of volatility of consumption around a growth path were trivial. I think when we step back from those calculations and try to put them in perspective, we can say a number of things.

Ireland: On one level, sure, it must be right. Imagine if your boss gave you the choice, more volatility year to year in your income path, or a higher slope looking forward for the next couple of decades. We can all do the present value calculations and see. It is the growth rate that's most important. On the other hand, all of us also, from personal experience... If you're in the academic world you see how your students do when they graduate and go out and get jobs. There's something missing from the cost of business cycle calculations. When the economy goes into a recession and young people are delayed in their start on a flourishing career, or some people have serious setbacks… there are serious, serious costs to that, that are simply not going to be captured by assuming infinitely live CRRA utility implicitly. There would be complete markets and so on and so forth.

Ireland: When we reflect back as economists and historians at the Great Depression, for sure, how that affected people's lives... Or, again, even you think about what COVID has done to young people. The costs are really significant and enormous. But then finally, I would say, it's a false choice saying, do you want to increase the rate of growth at the cost of having faster business cycles or clamp down on business cycles at the cost of having slower growth? What we know, first and foremost, as economists is that at the level of the society as a whole, you want to put in place a system of property rights, economic organization that allows for the fastest rate of long run growth. Part of doing that is to design fiscal and monetary policies that are predictable, and avoid imposing additional distortions of their own, and to every extent possible, smooth out fluctuation. So it's not a choice between having slower growth and more business cycles and vice versa.

What we know, first and foremost, as economists is that at the level of the society as a whole, you want to put in place a system of property rights, economic organization that allows for the fastest rate of long run growth. Part of doing that is to design fiscal and monetary policies that are predictable, and avoid imposing additional distortions of their own, and to every extent possible, smooth out fluctuation.

Ireland: Beyond that, what was it that shaped my decision to become a monetary economist and focus on business cycles instead of growth? Well, actually, my thesis was on money and growth. So I have done work on growth, and it's not that I think that it's uninteresting or unimportant to the contrary. But you get fascinated by what is it in free market economies that gives rise to business cycle fluctuations? What is it about what the challenges that the Fed is facing now on an economic level that shapes those challenges? And what are the insights that might lead to the most favorable outcomes? I just got fascinated by those questions, and that's what shaped my own decisions about what research topics to pursue.

Beckworth: Yeah. That's a great point. I agree with you, there's an interdependency there that you really can't make a clean separation. Just two examples that come to mind. One is our population growth rate is declining right now. I think over the long horizon, one of the most troubling developments, maybe the productivity slow down is also a big deal, but the decline in population growth rate affects so many things. It also ties into monetary policy. As that population growth rate declines, R star, the neutral rate drops, it affects the ability of the Federal Reserve through its normal instruments, we can debate about whether it's the appropriate ones. But through interest rate targeting, it affects its ability to perform.

Beckworth: So it's hard to disentangle these long-term structural forces that affect economic growth, and some of the short term issues. The other one, going back to what you mentioned earlier about students, think of millennials. I was still teaching in 2008 during the Great Recession, and there's these students who were going out, and you're right, their ability to start their careers impaired. These same millennials, maybe they got a job, but then here comes the pandemic, another big recession hits them. There's some studies from the Great Depression, I think you maybe were referencing them, depression babies, they're forever scarred, they're more risk-averse on the margin.

Beckworth: I got to think millennials are also somehow going to be permanently affected by going through two sharp recessions, which will affect long term outcomes. So it's important, I agree, to look at both of these issues. But both of us have been pulled into the business cycle more than anything else, even though we appreciate the other side. So along those lines, Peter, what do you view as the most influential and popular business cycle theories?

The Most Influential and Popular Business Cycle Theories

Ireland: Yeah, that's a good question. Well, I think if you ask me to just say where I think the mainstream is, I guess I would say it's dynamic stochastic general equilibrium modeling, built on a real business cycle core, but with significant product and or credit market frictions, many of which give the models the Keynesian flavor. The easiest way to say it would just be to say New Keynesian models.

Beckworth: Okay.

Ireland: Now, that's not to say there isn't important work being done using other kinds of models, but with where does the mainstream lie, I think it's with DSGE modeling.

Beckworth: So if I'm a graduate student, and I'm coming into a PhD program, a top 10, top 20 PhD program, and I'm taking a sequence of macro courses, what are the theories and the models I'm going to be introduced to? So you mentioned the New Keynesian model, so it's called the workhorse model, three equation, very basic, but there's also these recent developments in these heterogeneous agent New Keynesian, the HANK models. That seems to be kind of a hot thing. But if I'm a grad student and I'm coming in, is that what I'm seeing, New Keynesian up to HANK models? Am I seeing any other older models, overlapping generations? Any models with money in them? What would I be exposed to as a grad student these days?

Ireland: Yeah, that's an excellent question. Well, if you're talking about from the very start, the fall, when you come in, the way that most first year macro PhD sequences are structured is they start out with the Solow model, and then add the optimizing foundations Cass-Koopmans. So part of this is learning the economics, part of this is determined by learning the tools, the technical tools that you need to analyze and solve macro models. So you do dynamic optimization with Cass and Koopmans, the Ramsey model, I guess, is what people call it, and you learn about infinite horizon, dynamic optimization, but you don't worry about the stochastic part.

Ireland: So long as you're setting up the neoclassical growth model, you can talk about the various extensions to endogenous growth via endogenous technological change or human capital accumulation and so on and so forth, and that gives... Here at BC, I actually don't teach in the macro sequence, I do the math for economists. So while my colleagues are doing growth theory, rushing through dynamic optimization and getting to stochastic dynamic optimization and dynamic programming. So that means either towards the end of the fall or first thing in the spring, you can do this stochastic extension of growth theory, which is real business cycle theory, then you can begin adding frictions. I think adding price and... nominal price and wage frictions, like in the standard New Keynesian model, is probably the most common extension to pursue.

Ireland: The difficulty with the heterogeneous agent models is that that adds another layer of technical sophistication and difficulty to solve them. So whether you could get all the way there in the first year or not, it depends how hard you want to push the students. But presumably, you could do at least a little bit of that. And then I think there's more heterogeneity in graduate training after that, because the field classes are going to depend, in large part, on the personnel who are present and where their interests lie.

Beckworth: Okay. So just to summarize, you come in as a graduate student, and your macro sequence will start with a growth model, build a real business cycle model, and then build onto that the New Keynesian framework. And then depending on where you are, your elective classes, if you continue to go down the macroeconomics path, you might get a HANK model. But maybe some other places, you might also get a monetary search model. So, I guess, if you had to judge, what do most PhD macro economists come out with? What's their framing? What's the models that they have in their head when they graduate?

Ireland: Sure. Well, again, it's going to depend in large part on the personnel who are running the field classes. If you're at Western Ontario and have the chance to study with Steve Williamson, you should, and you will, and you'll learn a lot about micro-founded search models of monetary exchange. I think that's very valuable. Here at BC, with Ryan Chahrour, the expectational models of new shocks and so on, those are the ideas that students come into my office talking about. But again, it's driven by the personnel that will be there, I think, in different departments. But DSGE modeling, fancy, sophisticated... what I would call fancy VAR is elaboration on VARs, time varying parameters. The advances in computational Bayesian econometrics over the past 20 years have been phenomenal. There's so much more you can do with those models today, that you just couldn't do 10 years ago. That's a great area to find research topics. HANK models, search theoretic models, sure.

Beckworth: Those models you're just describing, this interaction of data and theory... So there's TANK models, I know they're very complicated theoretically, but they also fit them with large data sets, micro-based data sets and then Bayesian VARs. So a lot of exciting things happening empirically, theoretically. But let me step back and maybe get my question in a different way. Are we graduating generations and generations of New Keynesians on balance? Is there any other dominant paradigm? I mean people who will end up maybe working for central banks, going to the policy world. Are they getting exposure to other ideas? I mentioned monetary search model, your work on money, how many people out there take quantities seriously? I know you can't answer that question precisely, but what's the general direction, I guess, in terms of policy macro?

The Current Direction of Policy Macro

Ireland: Yeah, that's a good question. Well, let me in turn back up and try and give some rationale for all of this. One aspect of the design of a first year macro PhD class would be to get students who plan to study macro up and running. But what you also have in mind is that if someone's going to have a PhD in economics, but will go on to do labor or econometrics or even micro theory... what do you want them to leave with knowing about macroeconomics at the PhD level? My answer to that question probably would be, you want foundations in neoclassical growth theory and the endogenous growth extensions, and then over on the business cycle side… I think the logic and the method of dynamic stochastic general equilibrium modeling, probably of the New Keynesian sort. Then, when you move on to the second year, partly, you want to introduce students to research topics that they're going to build on with their own dissertations later. But you also want to introduce them to the tools that they can use, no matter which direction they go. And so, to some extent, learning how to solve a HANK model, even if you're not going to do research on HANK models, it's still going to be very valuable, or learning the basics of MCMC Bayesian estimation. Even if you don't run time bearing parameter VARs, it's really hard to pick that kind of stuff up later on your own.

If someone's going to have a PhD in economics, but will go on to do labor or econometrics or even micro theory... what do you want them to leave with knowing about macroeconomics at the PhD level? My answer to that question probably would be, you want foundations in neoclassical growth theory and the endogenous growth extensions, and then over on the business cycle side… I think the logic and the method of dynamic stochastic general equilibrium modeling, probably of the New Keynesian sort.

Ireland: So I really, to this day, think back about my teachers, Paul Romer and Lars Hansen and Heckman and others. I thank God for those guys. I wouldn't be able to do the research I do now without them. Not because I build on Heckman's research directly, or even Lars', but just due to the fact that because of them, with confidence, I can go to Econometrica... I can't do theorems and proofs and econometrics, but just to lead through theorems and proofs and say, okay, okay, I get what's going on and I can see how I can use these methods in my own research.

Ireland: So a big part of graduate training is just about introducing all PhD students to the basics of the field, and then feeding those who want to do macroeconomics, introducing them to the tools that they'll need. But when you talk about then who winds up going to the Fed? Well, we can see that there's diversity there, that there are huge groups of economists in the Fed that do DSGE, New Keynesian modeling. The dream always was to replace the old style Keynesian models with DSGE models, and that's happened, that's really great. But you also have people like David Andolfatto and Steve Williamson... Steve has left, but at the Fed, using search theoretic models to answer interesting questions about trading patterns in markets for liquid assets.

Ireland: You also have people who think about term structure of interest rates, coming at it from a macro finance angle. Think about like Eric Swanson, Refet Gurkaynak. Both of them have left, too, but you still have economists at the Fed doing work in their style, which has been hugely influential and I think hugely productive. So there is still, I think, a lot of diversity in methods and in the overall style of research. You might say, well, maybe there's not enough, because it's all neoclassical. Where are the radical political economists amongst us? You could also say... I do think many students today leave PhD programs with a mindset that finds it very difficult to read a book like Friedman and Schwartz and understand what Milton Friedman and Anna Schwartz were all about. Or, you could even go back to any issue in the history of thought, who knows what. Sure, that's true, but they can always learn on their own, too.

Beckworth: Yeah. I don't want to downplay the diversity point. I agree, there are a lot of people doing different interesting projects at the Fed, but sometimes you get the impression that maybe the Fed officials themselves, not all of them... Rich Clarida is an example. He definitely brought in some serious modeling, and he was very much a rational expectations thinker at the Board of Governors. But you often see... Again, this is just my view, but you often get the sense that they're thinking in terms of simple Phillips curve trade offs. They're not thinking through a DSGE or monetary search model. I guess that's the job of the Fed staffers to help nudge them in that direction. But I guess at the policymaker level, you revert back to very simple models you can do in your head.

Ireland: Well, I would agree with that, but let me both put it in a more positive light, but maybe identify part of what makes, I think, both of us uneasy. I agree. If you were to ask the typical FOMC member for a description of how monetary policy affects the economy, even Clarida, he's not going to go say, well, wait a minute, I must run the thing in Dynare before I can tell you.

Beckworth: Right.

Ireland: What they're going to say is, well, what the Fed does is it manipulates the federal funds rate, and via the expectations hypothesis, pulls market rates up and down with it. Because of the presence of nominal rigidities, those movements in nominal rates impact on real rates by manipulating real interest rates, that's how we influence the time path of spending. And then working through a Phillips curve by changing the rate of resource allocation, we can steer the rate of inflation. That's not even a New Keynesian story, that's more like an old style Keynesian story. But, there are two big lessons, not that you learn by solving things in Dynare, but there are two big takeaway points, in my view, from the New Keynesian approach beyond the Keynesian approach.

Ireland: One of them has been fully internalized, luckily. There has been some slippage here, but luckily not much, and that is the idea that expectations matter. So it is not... When we look ahead to later this week and ask, the Federal Reserve raises the funds rate by 50 basis points. Well, that is a sizable increase, and it probably will have some significant effects on spending and inflation and economic outcomes. But, okay, it's still just one half of 1%. Why is it going to have outsized effects? It's because it's not just that they're raising rates by 50 basis points, they're signaling an aggressiveness in response to heightened inflation that signals to markets that more large rate increases are ahead, and that's why you already see mortgage rates going up, 10-year bond rate going up.

Ireland: So counterfactual, let's suppose instead they only raise rates by 25 basis points. They will all understand that the effects of doing that are not really the effects of just a 25 basis point smaller rate increase. That's going to shift expectations out into the foreseeable future and have an enormous effect because of that. So, that is a message for modern macroeconomic theory, which comes out of New Keynesian economics, but not old Keynesian economics. I'm sure you would get no disagreement from anybody at the Fed on the FOMC about that. But here's another lesson, I think, equally profound and important from the New Keynesian model, that for a while, I think, did have significant traction, and now I'm not so sure. It's the divine coincidence, the notion that there is not, under most circumstances, a trade off between unemployment and inflation, but rather, by stabilizing inflation first, that's the pathway towards creating the stable monetary environment that lets the market provide for maximum employment.

Here's another lesson, I think, equally profound and important from the New Keynesian model, that for a while, I think, did have significant traction, and now I'm not so sure. It's the divine coincidence, the notion that there is not, under most circumstances, a trade off between unemployment and inflation, but rather, by stabilizing inflation first, that's the pathway towards creating the stable monetary environment that lets the market provide for maximum employment.

Ireland: That is an idea that has its cleanest and most profound statement or manifestation in the New Keynesian model. Gali and Blanchard coined the term divine coincidence with specific reference to the New Keynesian model. It was also, though, from a broader perspective, the kind of mindset that underlied the move towards inflation targeting and rules-based monetary policy during the great moderation. More recently, though, I detect a kind of backing off from that mindset. In the aftermath of the 2020 economic shutdowns and the redesign of the Fed's intermediate or long-term monetary policy strategy, there was a lot of talk about the employment mandate, and the notion that we're not even going to think or worry about increases in the federal funds rate until we see signs that the economy has returned to maximum employment. Now, although I agree the Fed has to tighten, and I'm worried about higher inflation, I am also worried by the fact that there is, a likewise, almost single minded obsession with inflation, this kind of lexicographic, stochastic shifting between lexicographic preferences for unemployment versus inflation I thought was a vestige of old style Keynesian, thinking that we had left behind. Now, elements of it are reappearing.

Ireland: I want to be fair, I don't think Chair Powell or anyone else on the FOMC said that this would be an ideal time to bring back the worst aspects of stop-go monetary policy and see how it works this time. But I did detect a kind of shift in mindset that went back to the thinking of the trade off. When you say Phillips curve, I think what you're referring to, and what makes us both uneasy... It's not that we think that there's no Phillips curve to be found in the data, it's the notion that the central bank is trading off employment versus inflation. That seems to be back in the thinking of the FOMC. I think that's a bad innovation. If I was advising a Federal Reserve Bank president or an FOMC member, I would say instead of arguing about 25 versus 50 basis point increases, because that's what everybody else is going to do, think about how to shift the mindset back to, our goal is to focus on stabilizing prices first, not because we're inflation nutters, but because we realize that by doing so, we're recreating the environment of monetary stability that will also allow the labor market to provide Americans with best opportunities with maximum sustainable employment.

The notion that the central bank is trading off employment versus inflation. That seems to be back in the thinking of the FOMC. I think that's a bad innovation.

Beckworth: Well, you've kind of answered my next question already. How do we get to this point, this quagmire that the Fed finds itself in? We talked about it earlier, high inflation, but the Fed also wants to have a soft landing, this delicate dance between soft landings and reigning in the heat in the economy. You've touched on some of the items, but let me step back and maybe ask a more general question about that. Do you think this change in its framework, so going from a flexible inflation target to a flexible average inflation target that looks at shortfalls to maximum employment, not symmetric deviations, do you think the framework itself is a culprit here, or is it just a run of bad luck? Maybe the framework's part of it, but we had a pandemic, we had a strong fiscal tailwind, just a lot of uncertainty that maybe even an ideal framework would've had a hard time handling.

Is the Fed’s Framework to Blame for Policy Missteps?

Ireland: Yeah. Well, part of that has to be true. The economy closed down in spring 2020 and fell into enormous... Well, there's nothing at all that monetary policy... anything more that they could have done to fix that. But here's what I would say. If you go back to 2019, when the reviews started and before the pandemic happened, it was pretty clear to me at least what the big policy problem was. If you accept, and this seems likely, that the zero lower bound is going to be a recurrent constraint that prevents the Fed from delivering sufficient monetary accommodation during recessions via interest rate increases, then how exactly do you prevent inflation from falling below 2% on average over the cycle? Now, the good news is that circa 2019, I had thought that the academic literature had provided a pretty clear answer to that question. You want to move away from a strategy that targets the growth rate of a nominal variable to a strategy that targets the level. So the easiest thing, inflation targeting to price level targeting. You could say, well, I prefer nominal GDP level targeting. I do, probably you do, too. But that's like a second to order thing. You target the price level. Now, okay. There were a whole bunch of papers written, different flavors of that targeting… the average or actually having a price level target, but something to make up for past misses is what you needed, and AIT had that flavor. So what I would've said is by itself, AIT was a sensible policy strategy, a solution to a very real problem, and by itself, something, therefore, that made a lot of sense and had significant support from contemporary state-of-the-art macroeconomic theory.

Ireland: I think the problem had more to do with the way that AIT was described and implemented. What went along with AIT was the reinterpretation of the employment mandate away from maximum sustainable employment, around a natural rate that was unknown to the notion of full and inclusive employment. In the strategy statement, there was language about most of the time, the two goals are complementary, but in instances where they are not, we take a balanced approach. They kept in the thing, I think, where they said, most of the time they're complementary, but they took out the balanced approach. You would say that's a minor thing, except that all of these words, they're so carefully thought out and debated. The whole thing had the flavor of, we're backing away from the idea of price stability first, and we're returning to the mindset of trade offs. The lesson that we learned from the aftermath of the Great Recession is that we should have cared more about unemployment, and we care more about unemployment now. Everybody cares about unemployment, but the thing is that we created all of the problems, the inflationary bias, all of the problems that we suffered through during the 1970s.

Ireland: So I think average inflation targeting by itself was just fine. Remember, this happened even before the pandemic. Remember that Fed Listens event, where they had all of the academics there, and John Taylor was there, and George Selgin was there, and I just got the feeling that there was something else going on, that the connection had been lost. Somehow there was this idea that the Fed really needed to do more fine tuning, that somehow monetary policy could be used better to create jobs, but we know that that's not true. So then we did have bad luck in 2021, in the sense that inflation accelerated far more rapidly than anyone had expected inside or outside the Fed. The new strategy was like, in baseball, a base runner, off first base, meaning so far towards second base, that it was an easy pick off. They set themselves up for this, I feel.

I think the problem had more to do with the way that AIT was described and implemented. What went along with AIT was the reinterpretation of the employment mandate away from maximum sustainable employment, around a natural rate that was unknown to the notion of full and inclusive employment.

Beckworth: Well, I agree with that. I want to paint a more benign interpretation of what happened in the past two years, but I do want to go back to that Fed conference, the Fed Listens, the change in the mindset that you perceive.  I think one can attribute it in part to the previous decade. [During the] previous decade, we had low, low inflation, and the Fed's balance sheet was still really large. So it created the impression of a free lunch. Both of us have expressed concern over this operating framework, the floor system, and George Selgin's been very vocal and a few others, that it separates the stance of monetary policy from the size of the balance sheet, and that creates this impression of a free lunch. Therefore, why not use the Fed? You're not causing an inflationary problem.

Beckworth: So I would attribute some of this thinking or maybe complacency, whatever you want to call it, to the change in the operating framework, which took place in 2008, but was officially sanctioned a few years ago. I think that's a key part of the story, an underappreciated part of the story. But let me step back and go back to the Fed's new framework. I want to give a mea culpa here, Peter. I had a New York Times piece last year, February, 2021. I was invited along with my co-author, Ramesh Ponnuru. So I'm not throwing him under the bus. Really, I put all the blame on myself, but I did not pick the title, but this is the essence of the piece.

Beckworth: But the title was, I hate to say this, *Stop Worrying About Inflation,* February 2021 in the New York Times, paper record. So it's a rather embarrassing moment in my career, it looks horrible in hindsight. But here's how I was thinking about it. I think to some extent, the Fed probably was thinking about it, as well. That was, at the time, we were focused more on the supply shocks being a cause of inflation. In fact, those would eventually work themselves out, we believe in capitalism. At some point, it would sort itself out. What I failed to appreciate was the impact that 2021 ARP package had on aggregate demand at the time. There were some who got it right, Larry Summers, Jason Furman. But at the time, I didn't see it. I was relying on some market signals that didn't turn out to be as accurate as I thought. And so I guess here's my question, could it be that it was just a misreading of the tea leaves, it was just a bad call, as opposed to a shift in a mindset? Is that a possible interpretation?

Ireland: Maybe, and it's always hard. That's why I say, maybe you... Who knows whether you actually learn anything from [inaudible] or not? Maybe we will. In economics, as in a lot of warfare and a lot of other things, as time passes, perspective improves. But I guess what I would come back to is... You've done this since so have I. You just draw a target path for the level of nominal GDP, and you base the target path in the fourth quarter of 2019. What you see is that throughout 2020, and even on into early 2021, the economy was still in a big hole.

Ireland: So you could say, wow, compared to a Taylor rule, which is focused on growth rates, even though we're having inflation that's above 2%, that's just putting us back to where everybody thought they would be when implicit or explicit nominal contracts were signed, when decisions were made before the pandemic. But with that kind of target, the nominal GDP goes back to the target path in the fourth quarter of 2021. In that case, what the strategy tells you is you should be all the way back to neutral. The thing is that... I'm willing to forgive some of this, but at a minimum, a consistent strategy like that would've dictated an earlier start to normalization. By not making reference to the target path all along, they got caught in a very difficult situation, too, because it became clear then later on in the year that inflation was going to be a problem. But the way the Fed operates, you have to prepare markets, and you have to prepare the political system. Actually, I was worried, late last year, that what they were going to do out of fear of the political system is just hope they could get away with not even talking about anything. They had that thing, we're not even talking about talking about this. They were-

Beckworth: Right.

Ireland: ... going to try and do that until the middle of this year. That would've been a total disaster. But the point is that they weren't set up to make the transition they needed to from extraordinary ease to the beginning of normalization fast enough. They would've been able to do it… David, you could have maintained a consistent viewpoint, and I think you did by saying like, look, I'm looking at this target path. My strategy is nominal GDP level targeting, and then you could have said as the year wore on, “well, look, my forecasts were wrong, but my strategy remains in place, and now the strategy dictates an earlier start to lift off. But that's simply because the economy is doing so much better than I expected, coupled with the fact that the price increases seem more persistent and broad-based than I expected.”

Beckworth: Yeah, I agree with that completely. In fact, mentioning Jason Furman, he actually took my rule that I developed in a paper I did on nominal GDP targeting, and he told me in an interview last year about midway through said, "Well, David, your rule would imply the Fed needs to raise rates right now." I was a little reluctant to embrace that implication, but he was right. I think part of it also is... for me at least, is I was doing too much looking in the rear view mirror analysis, like, well, nominal GDP still a little bit below, it's getting close. I think what I suffered from was more... I needed to be more forward looking. What's the forecast of nominal GDP?

Beckworth: I think you could make the case that it was beginning to sound the alarm bells third quarter of last year, if not before. But with that said, this is a nice segue. We haven't touched explicitly on any of your papers, so I'm going to do that now. So what is the path forward? We've had this challenge, we're in a quagmire. How do we avoid repeating these mistakes moving forward? You've had a series of papers where you make the case for nominal GDP targeting. Most recently, you had a paper for the shadow open market committee titled, *The Continuing Case for Nominal GDP Level Targeting,* but you also had several papers last year for other economies. You have a 2021 paper titled, *Targeting Nominal Income Under the Zero Lower Bound: The Case of the Bank of England,* and then you also have a paper where you apply this to the European Central Bank, and that one is titled, *Strengthening the Second Pillar: A Greater Role for Money in the ECB Strategy.* So walk us through why you think this is the path forward… this is maybe the next evolution. We have a review, in fact, coming up here in a few years for the Fed, the ECB will probably have a review again. So help us understand your thinking here.

The Path Forward for Monetary Policy

Ireland: Right. Well, no one has made the case for nominal GDP level targeting in a more comprehensive way than you have, David.

Beckworth: Oh well, thank you.

Ireland: I would refer everybody to that Mercatus paper. It outlines, in great detail, all of the advantages, and also takes care to answer some of the concerns. But what I would say... In the present moment, maybe I would identify four big advantages. The first thing is nominal GDP is a nominal the variable. It's not the case that the Fed can exercise control over nominal GDP quarter by quarter or year by year. But when we use nominal GDP to look back at the post-crisis but pre-COVID era, nominal GDP growth was slower than the Fed had intended, and, I would say, gives the correct signal that monetary policy, despite everything, was insufficiently accommodative over that period.

Ireland: The lesson to be learned then would be to think about ways out ways then to have delivered additional monetary accommodation. What the same framework is telling us now is that regrettably, the stimulus in the aftermath of the shutdowns, it was overdone and now needs to be partly reversed. So, nominal tracking, nominal GDP sends accurate signals about what monetary policy is doing at the level of the economy as a whole. At the same time, this is point number two, because nominal GDP is the product of real GDP and the nominal price level, it does express concern for both sides of the dual mandate, but in a way that I like, in a way that's balanced and continual.  So you look at nominal GDP all the time, you don't switch back and forth between unemployment and inflation. That is the recipe for more, not best monetary instability.

When we use nominal GDP to look back at the post-crisis but pre-COVID era, nominal GDP growth was slower than the Fed had intended, and, I would say, gives the correct signal that monetary policy, despite everything, was insufficiently accommodative over that period. The lesson to be learned then would be to think about ways out ways then to have delivered additional monetary accommodation.

Ireland: Then the third thing that I personally like about nominal GDP targeting, this is a point that Bennett McCallum made, even James Tobin before him. MV equals PY tells you that nominal income is shift-adjusted money. So when you're unsure about the stability of the Phillips curve, or when you're unsure how to aggregate the effects of quantitative easing or tightening with movements in the federal funds rate, looking at nominal income can be a kind of quantity-theoretic cross check against the sort of needing by needing decisions. They really have to be sort of Keynesian short-run in flavor.

Ireland: And then fourth, and finally, what has always been an advantage of nominal GDP targeting is that it handles the problem of supply shocks in a rule-like sensible way. And so we say, even now, it must be true that supply shocks are pushing inflation higher, but also slowing down the rate of real economic growth. But we did see that in the latest quarterly GDP report. Because real GDP growth was negative, nominal GDP growth was still quite high, but nominal GDP growth less rapid than inflation. It tells you that, well, yes, some of this is the result of lingering or even enhanced supply side disturbances. That's nothing that the Fed should be worried about, but it is also true under... the underlying monetary component of inflation remains too high. Again, there were many more arguments you could make, you have made, but those would be the four that I'd emphasize today.

Beckworth: Obviously, I agree with all of those, and would echo those. My hope is that this experience will reinforce this argument in people's minds, it will make the case for it. I wanted to compare the US and Europe real quickly here. In the case of the US, it tells us we now have excess aggregate demand, because nominal GDP is higher than any trend path or any kind of expected path for aggregate demand growth. So yes, there's been supply shocks, but we know, for certain, there also has been excess aggregate demand. So this was a guide, a warning, a metric to tell us to reign things in.

Beckworth: Now, if you go and look at Europe... I learned this reading a Martin Sandbu piece in the Financial Times. He tracked a bunch of countries' nominal GDP and their trend. In the case of the Eurozone, it's still a little bit below its pre-crisis trend path, and that makes a lot of sense. They've had more supply shocks. They've had the pandemic, they've had now this war, energy shocks. So a nominal GDP level target for them would tell them to take it easy, be gradual, and that's in fact what they're doing right now. So in some ways, it seems like some of these central banks are responding in a similar fashion to what a nominal GDP rule would tell them.

Beckworth: So the next step would be to make it official, forward looking. So they avoid some of these mistakes. I think the time is ripe for a discussion of how to embrace it. it's not to say there's no challenges that come with nominal GDP targeting. I think one of the biggest ones is data issue, data revisions and such, and many people are working on it. I know I have a project I'm working on, on this front. But my hope is that if there's a silver lining to the pandemic, to the big fiscal package we had, to the challenges in Europe, that is that there is a framework that could be used and applied in such settings, and that's nominal GDP targeting.

Ireland: Yeah, absolutely, I agree. I would also say that, although it would be great to see nominal GDP level targeting, or even maybe more explicit price-level targeting built into the official policy strategy statements, at least in the United States, one of the great strengths of the Federal Reserve system, is and always has been, it's decentralized structure. So there would be nothing to prevent... Again, if I were a Federal Reserve research director, you always want to give the president a choice of what topics to emphasize. But it would be, I think, a very effective approach to have a Federal Reserve bank president or one of the governors for a period of several years, all of the speeches, they've added advantages to do less work for speech writing. But you update your nominal GDP level graph and you use it to emphasize the consistency in your decision making strategy.

Ireland: You say a year ago, here's where we were, here's where we were now, above target. This means interest rates have to go up not to choke off the recovery, but, in fact, to bring inflation back to target and to promote the continued recovery. You could even much more elegantly and easily handle the pitfall that you and your co-author fell into. Forecasting is difficult work, and you're always going to be wrong. But the thing about a multi period target path, it's designed to-

Beckworth: That's a good point.

I would also say that, although it would be great to see nominal GDP level targeting, or even maybe more explicit price-level targeting built into the official policy strategy statements, at least in the United States, one of the great strengths of the Federal Reserve system, is and always has been, it's decentralized structure.

Ireland: ... to let you recover from the stakes. So you said like, at this time, last year, extraordinary monetary accommodation had to remain in place. Circumstances have changed, monetary policy has to change, as well. I think that if somebody prominent were to do that over a period of years... It's sort of the same way that Broaddus and Goodfriend did at Richmond, or the guys that St. Louis did back in the heyday, or Kansas City with Esther and Tom Hoenig… In Minneapolis, of course. You repeat things often enough and they're shown to work, and then they become... no one argues with you anymore about it, because it seems like common sense.

Beckworth: Yeah. Well, I like your idea, having a multi-year target to avoid the very predicament I was in last year. So have it over the next two, three years. That target can be updated as your potential real GDP values slowly update, something you won't see immediately. I'm working on a paper right now, Peter, with, with a colleague, Pat Horan, and we're building on the idea related to nominal GDP targeting that Michael Woodford made, I believe, in a 2011 paper or sometime around then. The Fed can still keep a medium run inflation target if it targets the level of nominal GDP. Just recently, Robert Hetzel wrote a paper for us at the Mercatus center, making that same argument, that if the Fed were to target something like a nominal GDP level target, it would preserve a medium run inflation target, because you'd have supply shocks that might one year push inflation above or below, but on average, you could hit that target. So I think there's a way to make a smooth transition from the framework we have. You can still call it inflation targeting of some kind, but it's anchored by a nominal GDP level target.

Ireland: Yeah, I agree with that. I see that the communication problem is a difficult one, and you don't want to come off as saying, well, I'm juggling all of these different models, and I pick the one that just happens to support my more or less arbitrary policy position. But I don't really see any inconsistency with... Say, the Fed just keeps what it has now, that FOMC keeps that strategy statement. I don't really see any inconsistency with members of the FOMC making reference to a multiyear target path for nominal GDP, because implicit in its design is the 2% inflation target. But then you explain, the problem is that month to month and quarter to quarter movements and inflation, measured inflation, are reflective, to some extent, of supply shocks, and we don't want to get distracted by that.

Beckworth: Well, let me ask one last question before we go. Our time is nearing an end, and that is, how easy will it be for us to make such a transition? So I think we both agree where we want to end up, the path forward, but it seems to me... Again, this is anecdotal observation, but as humans, we have a hard time thinking in levels versus growth rates. So, for example, we get worked up about inflation being too high, maybe too low, without looking at the price level, as you mentioned. Let me give you a concrete example of this. So if you go back to the Great Depression after that great contraction, 1929 to 1933, the economy, actually, is recovering relatively robustly in '33, slows down. But in '34, '35, the Fed becomes a little bit nervous, because inflation is starting to rear its head, but it's still relatively low. Moreover, the, was it 30% drop in the price level, there's a huge hole there, but all the Fed officials could see was the little inflation spike. They were missing the forest for the trees.  That's my concern with level target, and I think it's a great idea, but are our brains hardwired to think in terms of growth rates versus level, or is this something we can eventually adapt to?

Ireland: Right. Well, I think that's why just showing the graph is more effective than trying to give any kind of de detailed technical explanation. Look at what happens if you just do the thing of plotting actual nominal GDP against the target path with the base that starts in the fourth quarter of 2019 nominal GDP, I don't know, I can't remember, like 12% or something like that below the target path. You say, that's the ideal picture to show, to explain why, even though, as the economy began to recover, inflation was above 2%. That was the time to say, we're not even thinking about thinking about tightening. We're in an enormous hole, we've got to get out, and the picture just shows that.

Ireland: But again, as the gap diminished, and now it's past the neutral point, then you say, look, there was a lot of monetary accommodation in place over the last two years. Our objective isn't to send the economy back into a recession, it's certainly to make sure the price level doesn't rise even farther above target. So I think you can show the pictures, and that works better than anything else, so people can see what's going on.

Beckworth: Yeah, that's a great point. I remember Carola Binder had a paper at the Cato Monetary Policy Conference a few years ago, where she showed that level path up till 2019. It was relatively straight. Her argument [was] had the Fed just said, "Hey, we're on this path," instead of the inflation rate, it would've been a much easier job communicating.

Beckworth: Well, with that, our time is up. Our guest today has been Peter Ireland. Peter, thank you so much for coming back on the show.

Ireland: Oh, great. Thanks for having me.

Take the Macro Musings listener survey here.

Photo by Win McNamee via Getty Images