Bryan Cutsinger, Peter Ireland, and Will Luther on Lessons Learned from the Fed Framework Review

Bryan Cutsinger is an assistant professor of economics at the College of Business at Florida Atlantic University. Peter Ireland is a professor of Economics at Boston College. Will Luther is an associate professor of economics at the College of Business at Florida Atlantic University and is the director of the American Institute for Economic Research’s Sound Money Project. Bryan, Peter, and Will return to the show to discuss the big takeaways from the 2025 Fed framework review, the flip flopping of FIT to FAIT back to FIT, the biggest lessons from the 2020 Fed framework review, the case for NGDP targeting at the Fed, hope for future reviews, and much more. 

Subscribe to David's Substack: Macroeconomic Policy Nexus

Read the full episode transcript:

This episode was recorded on May 6th, 2026

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: Welcome to Macro Musings, where each week we pull back the curtain and take a closer look at the most important macroeconomic issues of the past, present, and future. I am your host, David Beckworth, a senior research fellow with the Mercatus Center at George Mason University. I’m glad you decided to join us.

Our guests today are three veterans of the podcast: Bryan Cutsinger, Peter Ireland, and Will Luther. They join us today to discuss a recent paper they have produced on the Federal Reserve’s framework review. Gentlemen, welcome to the podcast.

Will Luther: Thanks for having us.

Bryan Cutsinger: Good to be here.

Beckworth: It’s great to have you back. Again, you’re veterans of the show. Will, you weren’t on too long ago. Maybe last year, I believe.

Luther: Yes, I think that’s right.

Origins of Bryan, Will, and Peter's Paper

Beckworth: We talked about Bitcoin, stablecoins, the future of money. Today, we’re going to talk about the future of the Fed, and lessons learned or not learned from its framework review. Tell us, how did the three of you get together and decide this was something that needed to be done?

Luther: Well, I think it began with the work that we were doing prior to the framework review. Bryan and I had both organized a special issue of the Southern Economic Journal. You participated. Peter participated in that special issue. We put together all of this work, hoping to inform this process. 

Then, in January of 2025, the Fed kicked off its official framework review. In that first press conference, Chair Powell laid out basically what was going to happen, but the review hadn’t really started yet. That raised some eyebrows that it looked like this was worked out in advance. That’s when I started paying close attention to the actual review process and then spoke to Bryan and Peter about reviewing the review.

Beckworth: Okay. Anything you want to add to that? Any interest or perspective?

Cutsinger: I think for me, one of the things that I found frustrating is that we have 40-year high inflation. It seems like that would be the appropriate time to maybe radically rethink the framework. Think about something like nominal GDP targeting or price-level targeting or whatever. To just basically admit we’re going to go back to a not great version of the status quo seems like a large missed opportunity to me. I think that was the other factor driving wanting to write this paper.

Peter Ireland: Right. It’s an exaggeration to say that things are as bad as or worse than they were in the 1970s. As somebody who studied Fed Reserve policy and the history of Fed Reserve policy, it’s painful to see elements of the mistakes of the 1970s play out again. It’s an opportunity to highlight the similarities, the problems that arise again and again, over the course of monetary history, in hopes of improving things in the future.

Beckworth: Yes. The review process itself, was it not delayed a bit? We were told at one time they got pushed back because of other issues. The timing itself was a little unique, wasn’t it?

Luther: Yes. Very unclear about when it was going to start, when it was going to conclude. In hindsight, that’s a bit strange because they had an outline of what they were going to discuss. They knew how many meetings that was going to take. Not clear why they didn’t just say in advance, we’re going to conclude and reach a decision by August. Yes, a lot of uncertainty about the process up front.

Big Takeaways

Beckworth: What are the big takeaways from your review? We’ll get into the details, break these down further as we get into the conversation, but what are the big takeaways that you want the listeners to think about as we move forward in the conversation?

Cutsinger: I think one aspect of the paper is just reviewing the actual timeline. In the paper, we have a review of the Summary of Economic Projections beginning in 2021 and then through 2022. It’s clear that the FOMC was looking at this problem as a supply shock, and they continue to look at this as a supply shock despite the evidence turning against this much earlier than Chair Powell admits the evidence turning against this.

Obviously, with your show being a fan of nominal GDP targeting, had they been looking at the level of nominal GDP implied by the 2019 Summary of Economic Projections, nominal GDP was back up to trend by April of 2021. Certainly, by July of 2021, it would have been very clear. One of the main takeaways here is that there was a serious misdiagnosis of the problem at the time, and it’s unclear from the framework review process that they actually internalized the mistake they made in misdiagnosing the problem.

Ireland: Right. Then going back to the theme of historical problems repeating themselves, if you think about the 2020 review and the statement that followed, there was something to like about the 2020 review. What it seemed to do on the surface was to adopt a level targeting strategy, which we may have a chance to talk about in more detail later on, which many of us thought is exactly the right thing to do when policymakers are confronted regularly with the difficulties associated with the effective lower bound.

The trouble came through a myriad of details in implementing that levels targeting scheme. On the surface—and we hear this being described this way by colleagues and by the media and by policymakers—it seems like the 2025 statement is a return to the sounder principles originally outlined in 2012. In a way, it is, and maybe that’s a good thing. On the other hand, when you look at the details that distinguish 2012 from 2025, those details—and again, we can talk about this more—still raise concerns about biases in policy, incompleteness in the policy strategy going forward.

Beckworth: Missed opportunities is what I’m hearing, and many ways this could have gone differently. We want to talk about that. Maybe let’s begin with the 2020 framework review. Peter, you just talked about that. What were the big changes it introduced? You mentioned level targeting, but maybe spell out the details and how it was at least publicized.

The Fed’s 2020 Framework Review

Ireland: It was very clear when the Fed Reserve conducted the 2019 strategic review what the big issue was. It had to do with the effective lower bound or the zero lower interest rate bound. If, as seemed likely at the time, the ZLB was going to be a recurrent constraint preventing the FOMC from delivering sufficient accommodation during deflationary recessions, how could the committee prevent inflation from falling modestly, but systematically below 2% for an extended period of time, giving rise to larger and larger discrepancies between the actual price level and the level of prices that households, consumers, and firms expected when they enter into longer-term contracts?

At the time, 2019, the academic literature had provided a pretty clear answer to that question. When the ZLB is a problem, you switch to level targeting. The biggest advantage is just the mechanical one. It means that if you do have a series of one-sided deviations, you work harder to bring the level of prices back to where everybody expected them to be. The trouble came in entirely through, again, what turned out to be a myriad of details.

First, there was an asymmetry in the level targeting scheme where the strategy statement itself indicated that there would be a makeup following undershoots of inflation, but did not explicitly indicate that there would be a similar makeup when there was an overshoot of inflation. It was unclear whether that omission was intentional—just to emphasize that the strategy really was asymmetric—or whether it was meant to be symmetric, but they were just describing, for communication purposes, the most relevant example.

In some of your work, you rely on excerpts from speeches FOMC members gave that strongly suggest the asymmetry was intended. Then, very briefly, the 2019 statement in a variety of ways also appeared to elevate the employment mandate over the inflation mandate by specifying a policy response to shortfalls as opposed to deviations of employment from its maximum sustainable level. Then, most puzzlingly, of all, the 2020 statement dropped a reference to a balanced approach to policymaking in the aftermath of aggregate supply shocks.

I think the problem, in a nutshell, could be summarized with reference to that balanced approach. If monetary policymaking is not balanced, then what is it? It’s imbalanced, and it’s heavily biased toward the creation of more inflation. When circumstances suddenly changed in 2020, the Fed was caught off guard, and high inflation is what we ended up with.

Beckworth: That point about the maximum employment emphasis taking center stage, redefining it was broad-based, not just regular maximum employment. I think that mattered as much as the fact it was more of an emphasis. I think a lot of people, a lot of commentators look back and say that was probably a big mistake. People you wouldn’t normally expect, the Romers, Gauti Eggertsson, Donald Kohn, they all said, “They pushed too hard here.” I think there’s probably broad consensus that was a mistake. Let’s go to the area where there’s some support, even we would support some version of the makeup policy part.

You mentioned this asymmetry, which I do think was intentional. Let me tell you why I think they meant it that way. I want you guys to respond to this. In my view, it seems they are trying to implement a version of Ben Bernanke’s temporary price-level targeting. The way he viewed the world—and where I’m going to get with this, is it would have been a whole lot easier just to do nominal GDP level targeting, cut to the chase. What he’s doing is, he’s saying, look, the real demand problem that we will probably face and have faced is demand shortfalls.

Therefore, the real problem is from below. We make up from below. Then from above, it’s just regular inflation targeting because we can handle above inflation. What’s the big deal? We can deal with that. It’s a way to deal with what they thought would be the problem. Bernanke also mentioned in his talks that he wanted to avoid the problem of responding to negative supply shocks. If the price level went up because of negative supply shock, you don’t want to generate a recession on top of a weakened economy just to bring it back down.

To me, of course, this is all screaming, well, just cut to the chase. Just stabilize aggregate demand itself. I guess in my view is they were groping in the dark toward what we would think would be optimal, a nominal GDP level target in an imperfect way, and they ended up with FAIT. How do you respond to that?

Luther: I would say there are, broadly speaking, there are two justifications for level targeting. The first justification, the justification that folks like Bernanke have offered is this effective lower bound problem. The other justification, though, is that people are trying to make plans. In order to make plans, they have to have some sense of what the dollar is going to be worth over the course of the contracts that they’re entering into. If those dollars are worth more or less, then they’re going to make some mistakes along the way.

If you’re level targeting, you can ensure that those dollars are worth what they were expected to be worth when people entered into those contracts, more or less along the duration of those contracts. The difference between those two views is that if you’re only concerned about the effective lower bound, you really only need to worry about errors in the one direction. If you’re concerned about what those dollars are going to be worth over the course of a contract and how that will affect people’s plans, then you need to take a more symmetric approach.

I think one of the things we try to emphasize in the paper is that most people were concerned about their plans being frustrated, but Fed officials were primarily concerned about that effective lower bound.

Beckworth: Yes. The premise guided the outcome. A more robust symmetric view would have helped a lot. Okay. Well, let’s talk about then what they learned from their review, from the process of looking back at 2020, and then also what they went through. What were the lessons they drew, and were they the ones that should have been drawn?

Lessons Learned from 2020 Review

Cutsinger: My view is that when you look at how they describe what happened in 2020, 2021, 2022, they still don’t see it as primarily a surge in aggregate demand. They seem to look at these problems as mainly a problem of supply shocks. If you continue to misdiagnose the problem, as they seem to have done, then it’s unclear that you’re going to get a very effective framework review because you misunderstood the nature of the problem.

If you go back again, as I mentioned earlier, and you look at the Summary of Economic Projections, by September of 2021, the implied inflation rate for the remainder of the year, for that projection to be accurate, would have to be, I think, it’s 1.9%, actually below-target inflation with no change in the median FOMC Fed funds rate projection. Again, when you look at how they went back and looked at their own performance here, it’s very clear that they saw this as a supply-side problem, even though the evidence was very clear that it was a demand-side problem. The lesson they seemed to take away from this was that it was not a demand-side problem, which is just very bizarre to me.

It seems to me also that they’re continuing to make the same mistake right now, which is that we continue to have inflation that’s above target. Every FOMC meeting, we’re getting another excuse for, well, we’ve got this supply shock, we’ve got that supply shock. Again, if they were just to look at the level of nominal spending, either use your measure of the nominal spending gap or Will’s measure of the nominal spending gap, it’s been shooting up. Supply shocks don’t cause that problem.

I look at this and say, again, you missed a big opportunity here to think a little bit more radically about a framework, whether a price level target or a nominal GDP target. Instead, we’re left with perhaps an even worse framework than what we had in 2020.

Nominal GDP Targeting and Productivity Shocks

Beckworth: Now, I should mention to the viewers and listeners of the podcast that Bryan also was recently on the podcast discussing a paper where he looks at positive supply shocks as well as negative ones. The focus in the literature and these discussions at central banks is largely let’s look through negative supply shocks. In fact, we could have AI-driven productivity surges that would force us, if we’re being consistent, to look through the positive ones. That’s hard, right? That’s really hard.

If your frame is inflation, inflation, inflation, it’s hard to see anything other than a fall as being problematic, right? Where a temporary fall might actually be optimal, an optimal response. Could you speak to that, Peter? Why is nominal GDP targeting, in your view, optimal when you have these productivity shocks both above and below?

Ireland: In a nutshell, nominal GDP targeting gives you a better answer because it’s a summary of what aggregate demand is doing to the economy. It’s really aggregate demand that the Fed can influence. If we see, as we did post-2008, sluggish growth in income and below-target inflation, that’s a pretty clear signal that monetary policy, in fact, is not doing all it could to support the economy. It underscores the importance of bringing inflation back to target when it’s running chronically below.

On the other hand, if you see oil prices or tariffs or a pandemic pushing inflation and growth in opposite directions, that’s the hallmark of an aggregate supply shock. That’s the price-level movement that the Fed Reserve ideally would look through. We talk about this in the paper. Flexible average inflation targeting or flexible price-level targeting don’t preclude this looking-through approach, where the chair and FOMC members in speeches can explain, “Well, we’re tolerating slightly higher inflation because of nonmonetary influences that monetary policy can’t do anything about.”

The advantage of nominal GDP targeting is that it enforces that looking-through behavior. If output growth and inflation are moving in opposite directions, nominal GDP remains on track. It’s a signal to central bankers, “Okay, monetary policy is doing all it can too.”

Beckworth: One of the beauties of nominal GDP is it provides a cross-check. You don’t have to be an official targeter. You can just look at it. Are we growing faster than some benchmark measure of where nominal GDP would be going? You’ve mentioned this, Peter, a lot in your work—use it as a cross-check.

Ireland: A cross-check or another set of words I’ve heard used to describe it is the two-for-one deal. Okay, you can think of nominal GDP targeting as a substitute for inflation or price-level targeting, entirely new framework. Maybe that’s what all of us would ideally like, but the two-for-one deal emphasizes that it doesn’t have to be a replacement for either inflation targeting or price-level targeting. It can be, as you say, a cross-check where you indicate at times you want to look through observed movements in the inflation rate because they’re driven by disturbances from the supply side.

Luther: The ability to use this as a cross-check really makes it amazing that Fed officials don’t seem to have learned from the events of the last five years. If you go back to September of 2021, what do you see? You see that real output had almost entirely returned to trend, and then the inflation started rising more rapidly. Then, if you move forward to December of 2021, they’re making projections for 2021, inflation in 2021 in that Summary of Economic Projections. They have data through October.

They lived through November and part of December, and yet actual inflation for 2021 was above the range of projections from FOMC members. That combination of both being able to use this as a cross-check and surely seeing that you were really, really wrong there and remained wrong over the months that followed, you would expect that to prompt some introspection.

Beckworth: Some deep soul searching.

Luther: Yes, and yet that doesn’t seem to have happened.

Beckworth: I will put a plug here for Rich Clarida. He was on the podcast recently, and he goes, yes, we should have looked at nominal GDP targeting in the very time you just mentioned, Will. He’s a big believer. It’s something to cross-check yourself with. To his credit, he actually invited me to the board. It was just him and his aides, but he really was interested. I do think there’s an opening. I think these conversations, your paper, is exactly what we need to have this conversation. We all beat that drum really hard before the framework review. Look at nominal GDP targeting. Now, I was under no illusion they were going to adopt it, but at least again, a cross-check, a two-for-one deal, do something with it. 

Let me give one other example, a concrete example why this is an important tool. The Great Recession, 2008—we had negative supply shocks. We had commodity shocks. Inflation was rising, particularly the second half of 2008. We already have a weakened economy from financial system, from other problems, and the Fed is leaning into tightening.

It’s signaling rate hikes in the second half of 2008. If you’re looking at a neutral rate or some equilibrium rate, it’s probably falling to the floor, and they’re signaling rate hikes. Had they been looking at forecast of nominal GDP, they would have seen this was a foolish decision to make. I think the issue is, it’s just hard for central bankers whose minds are wired on inflation targeting to see beyond that. They see high inflation, and their natural first instinct is to panic. We need to get in the habit of, okay, it’s high inflation. “Let’s look to nominal GDP and get a reality check.”

Cutsinger: I recently presented the paper that you alluded to. There was an economist in the audience who was very skeptical of the idea of letting the price level adjust in either direction, either lower or higher in response to supply shocks, because he said, “Well, in the long run, I agree with you that there’ll be a labor market reallocation if we’re dealing with, say, less sticky nominal wages. In the short run, this could be quite bad.”

It dawned on me as he was making this comment that actually the reverse is true. If you’re an inflation-targeting central bank and nominal wages are sticky, the adjustment process is actually going to take longer in that regard, rather than if you were targeting nominal GDP. I think the problem that you point to runs not just deeply in terms of how central bankers think about monetary policy, but even other academic economists think about this.

Again, if your concern is about the labor market adjustment in response to an aggregate supply shock, either negative or positive, if you think wages are sticky, and I do, then actually the case for nominal GDP targeting over some version of inflation targeting is clearly going to result in a much faster labor market reallocation. Even if, by the way, the effects are heterogeneous. Even if you have two labor markets where one is affected by the supply shock and the other one is not, the adjustment process of moving labor from one market to the other is going to happen faster if you let the price level adjust to the change in productivity rather than trying to wait for nominal wages to adjust because you’re inflation targeting.

Beckworth: I love this conversation because it’s turned from a discussion of your paper into nominal GDP targeting. I often get listeners say, “David, you talk too much about nominal GDP targeting.” There’s been a drought of nominal GDP targeting talks. We’re making up for it in the show, so feel free to continue. One last question about nominal GDP targeting, and we’ll get back to the paper and the review process.

In your view, what are the best arguments for it? You mentioned, Peter, it’s a nice way. It’s a two-for-one deal. You cross-check yourself. It’s a combination of the real and the nominal. There’s been other stories about its forward guidance. Jim Bullard has, in a world of high leverage and fixed nominal debt contracts, it’s a way to stabilize. Debt can effectively become like equity when you’re providing the right offsets. I guess what friction and therefore what theory or motivation for nominal GDP targeting is your favorite?

Luther: I think most macroeconomists accept that expectations are important. If we look at what the average person expects, they expect the central bank to do something like nominal income targeting. That is, when prices are rising, they expect output will be falling. When prices are falling, they expect output to be rising. They have some view of the world that seems to presuppose a stable path for nominal spending.

I think central bankers have a choice to make. They can work really hard to communicate some other way that the public should be thinking, or they can adopt a monetary policy strategy that takes those expectations as given and then delivers the best policy that they can in light of those expectations. Nominal GDP targeting, I think, is a good solution there if you’re taking that latter approach.

Beckworth: The challenge is, of course, the name nominal GDP targeting. It’s a mouthful. It sounds strange. It sounds foreign. I think you guys have all talked about before, you package it as you’re stabilizing dollar incomes. People, as you said, they make plans for their mortgages, their car loans based on an expected path of their dollar incomes, and you’re just stabilizing that. If you’re talking to a trade group, we’re stabilizing sales for industry in the aggregate. You can market this to the average firm or person, right?

Luther: Yes, I think so. The key is that you just need to do as good or a better job as you would do with inflation targeting. It’s really difficult to explain inflation targeting to the public. Try to explain to them that you want prices to rise faster. You’re going to get a puzzled expression in those cases, right?

Beckworth: Ben Bernanke had that experience, I believe, in 2010 with QE2. “Why are you trying to increase inflation? Peter, you’ve done a lot of work empirically on this issue. One of the critiques of nominal GDP targeting is, well, what about data revisions? What about GDP? Sure, I buy your argument in theory, but, man, you’ve got these big, imprecise estimates, initial, then second, then the third. What do we do with that as champions of nominal GDP targeting?

Ireland: Well, the challenge in monetary policymaking is always that by trying to do too much, you might end up doing too little or even doing bad, making things worse instead of better. Data revisions, noise in data is always going to be a problem, but I would argue along those lines. It’s an advantage, for instance, that nominal GDP comes out on a quarterly basis. It means it smooths out the month-to-month noise. It’s an analog under inflation targeting based on the idea that you use unemployment as a leading indicator of inflation when making policies.

You routinely have these situations where noise in the monthly employment statistics or in the monthly inflation statistics cause swings in expectations, swings in how the FOMC needs to communicate with the public that could be avoided by taking a slightly longer horizon. As a matter of fact, I would go further along those lines. In implementing nominal GDP targeting, maybe look at something like year-over-year growth in nominal GDP as an indicator of total growth in spending and incomes. Again, as a way of smoothing things out, keeping the focus on intermediate-term trends, taking the pressure off policymakers to have to respond to every single monthly data release that comes out.

Reviewing the Fed’s 2025 Framework Review

Beckworth: Let’s switch back to your paper. I know this is in your paper, nominal GDP targeting, but let’s talk about the process of this review. I want to try to be charitable to the FOMC here, so I’m going to take their case, and you can push back, respond to it. They just came out of this really, really high inflation. They got wounds. They’re being attacked politically by the public. We relearned that the public really dislikes inflation. It was the number one issue on Gallup polls and many polls. In fact, I’ve just been collecting data on this. It’s still a very prominent issue. It hasn’t really faded. 

The Fed’s like, man, we need to just go back to the basics. We need to go back to regular, flexible inflation targeting. We tried what these crazy people, the Michael Woodfords, the Peter Irelands of the world advocated, level targeting, and look what it did to us. We got burned. Maybe we should just keep it simple. We’re going to go back to flexible inflation targeting. Am I being too charitable here?

Luther: Well, let’s start with what Fed officials say. For starters, they don’t say what you just said. In fact, Powell says that the framework was not a problem. The framework did not preclude them from doing what needed to be done. At least in terms of their official statements, the problem with the 2020 framework was not the framework, but that the framework was difficult to communicate. We’re already at odds with your story just on the premise. In your story, you have to start with there being a problem, which they don’t concede.

Cutsinger: I’ll actually say it’s a little bit worse than this. In fact, the same remarks that Will was alluding to with Chair Powell, he says that the inflation was exogenous. Now, it’s unclear what he means by exogenous. He never goes on to clarify. Presumably, he means outside of the Fed’s control, which is another way of saying not aggregate demand-driven, but instead being aggregate supply-driven. It’s unclear based on that statement that if we take Chair Powell’s view as being the view of other members of the Fed, which, of course, it’s probably not, but the official statement there is that the problem was outside of the Fed’s control, and the framework, therefore, was irrelevant. To the extent that it was relevant, it worked as intended were basically his remarks.

Luther: It’s also not just statements from Chair Powell. If you go and look at the staff papers from Fed economists that were prepared for this framework review, what you’ll see across those papers is a reiteration of there being a lack of consensus on what caused inflation in 2021, 2022. It’s not surprising then that Powell would say that this was primarily a series of supply shocks because that’s what’s coming out of those staff papers that were prepared for the framework review.

Beckworth: What about the argument that fiscal policy played a large role? You mentioned exogenous. Maybe he’s referring to fiscal policy. Again, let me go to play the devil’s advocate here. Whenever we have huge wars or huge emergencies, the Fed has to step in. I don’t want to use the term fiscal dominance, but play second fiddle to what Treasury and Congress is doing. Massive expenditures.

David Andolfatto has this paper. He says, yes, it was excessive. We didn’t need that much fiscal stimulus. In real time, in the fog of war, the Fed had to step in and basically monetize. It had to step in and do what it had to do. The high inflation is just the price we paid to get through the COVID pandemic. The pandemic was just another world war, a public health war. We couldn’t have done much better. How would you respond to that?

Cutsinger: Perhaps that’s true, and let’s take that all as given. If President Biden and Congress at the time had gone to the voters and said, this is what we’re going to do—of course, President Trump in the first term had gone to the voters and said, this is how we’re going to respond to the pandemic. It looks like we’re probably going to end up getting quite a bit of inflation. That’s just part of the cost of responding to this. If you want that, we’ll do it, but give the voters a choice.

It’s unclear in the way that this worked that the voters actually got a choice in this process. If the Fed is passively responding to what’s happening with fiscal policy, then in some sense, the voters are not getting a say in what the consequences are of the policies. That would be my response.

Ireland: Right. I also think it’s worth thinking about how events played out in real time. By mid-2021, it was abundantly clear that while the economic recovery was uneven, it was robust and sustainable. If you think about the range of outcomes that we would have had to worry about in mid-2020, the entire lower tail had been cut off from the range of possibilities. Yet, we had two additional huge Keynesian stimulus packages in late 2021, enabled by a monetary policy that was obviously overly accommodative. It seems hard in retrospect to justify any of that with an economic stabilization argument.

Luther: If we zoom out here as well, it’s a little odd to talk about fiscal policy shocks from the perspective of a monetary policymaker. That fiscal policy doesn’t change that quickly. There’s no reason why Fed officials should be surprised by this massive fiscal expansion. You take this time to draft a bill. You argue over the bill in Congress. Then it passes. Then the expenditures get made. This takes a lot of time. There’s no reason why Fed officials should be surprised.

If they’re committed to stabilizing spending, then they should be conducting monetary policy with that fiscal policy in mind. David Andolfatto says this is the best we can do, and maybe he’s right. Maybe we needed an inflation tax on the American people to carry through this fiscal policy. Let’s not take that to mean that monetary policy was conducted well from the perspective of monetary policy. It was a policy decision. They weren’t surprised by the fiscal policy. It resulted in very high inflation. That’s not supposed to be what they’re doing. Either we have to accept that the Fed made a mistake, or we have to accept that it was doing something other than what its dual mandate says it should be doing.

Beckworth: Okay. What I’m hearing from all of you here, I think, is that by late 2020, early 2021, it was very clear we were getting back to full employment economy. In terms of nominal GDP, we’re getting back to where it would have been had there been no pandemic. Clearly, the signs were there for a tightening to start taking place. When did they first start tightening? Was it in March?

Cutsinger: March ’22.

Beckworth: March ’22. It was a whole year later. I think even more egregious would be the balance sheet purchases that continued to go on for some time. I think you’re telling a story, and maybe there’s more nuance here you would add to this, that early on, and you can understand why they were aggressive, but at some point, it became clear there was more than sufficient aggregate demand. In fact, there was too much aggregate demand, and they should have stepped in sooner and done their thing. That’s where you guys would land. Is that a fair assessment?

Cutsinger: To be fair, Will and I, with a couple other co-authors, actually have a paper in the Southern Economic Journal that we wrote in 2020, and I think it ended up getting published in 2021, where we actually gave the Fed credit and said, actually, yes, if you look, PCE inflation actually fell in May of 2020. There was both an aggregate supply and an aggregate demand shock that hit the economy with the onset of the pandemic. We praise the Fed’s original response to acting aggressively to make sure that you stabilize aggregate demand.

When you go back to your earlier remarks about the Fed starting to tighten in March of 2022, it’s important, I think, to get the timeline correct here. In December of ’21, Powell’s saying, “Hey, we’re retiring the whole transitory thing. Demand is an issue here.” Yes, they raised rates in March of ’22, but it’s only 25 basis points. Real rates were still negative. In fact, real rates didn’t become positive again until they raised rates at the July ’22 meeting, which is a full year after nominal GDP had got back to the implied path from the 2019 Summary of Economic Projections. The idea that the Fed responded aggressively and appropriately once the data turned against the supply shock view, it’s just hard to square with the timeline here.

Beckworth: Okay. Let me try to salvage a win from this takedown you guys have provided today of monetary policy. The one thing I do find encouraging from this experience is that we did demonstrate that you can return nominal GDP or aggregate demand quickly back up to trend, right? If you do it aggressively enough, quickly enough, now we need to thread the needle better, is the big point here. They overshot that path. Coming out of 2008, it wasn’t clear to many people that it was possible, right? 2008, persistently below the path. One could argue maybe there was some hysteresis, some long-term effect potentially, and potential real GDP. There was a lot of like, “Hey, man, the Fed can’t do anything.” Then suddenly we get to the opposite. 

Luther: There were all those talks about the Fed being out of ammo. Do you remember that?

Beckworth: Yes.

Luther: They had ammo.

Beckworth: They would say fiscal policy played an important role, too, in that story. To me, I want to take the win here in that it is possible to do something like nominal GDP level targeting. That, to me, is one takeaway. Yes, we did more than that. We overcompensated. Am I claiming too much there, or is that fair?

Luther: I think that’s fair. The risk when you do too much is that it makes it harder to do the right thing the next time around. If you go back to 2008, 2009, Chair Bernanke was sitting before Congress, and he was getting peppered with basically different versions of the same question, which is, you’re not going to make this recession worse by throwing some inflation on top of it, right?

My concern is that the next time around, when we actually have a reduction in aggregate demand, and Fed officials want to boost spending to shore up the economy, that they’re going to be peppered with those questions again. It’s going to be harder to restore spending because they did too much this time around. They’re going to say, “Don’t you remember what happened in ’21 and ’22? Are you trying to engineer another 40-year high inflation?”

Beckworth: Once bitten, twice shy of Congress, there’s no thank you. We’ve already tried that medicine, and it wasn’t very pleasant. Let’s go to the new framework. We’re dancing around it, but let’s go back to it. You guys touched on it earlier. On the surface, it’s a return to flexible inflation targeting, but there were other things added to it or changed in it that suggest some openings, some possibilities. If push comes to shove, maybe we’ll pull out some of the old tools. Tell us your concerns with the new framework.

Ireland: Well, maybe I could say, going back to another point you made, one way of finessing the whole thing, going back to something more sensible without publicly admitting mistakes, would have simply been to say that we’re now in an environment where the ZLB doesn’t appear to be as big of a threat as it appeared to be post-2008. We’re just going to go back to 2012. I think a case could be made that would have been a good outcome, maybe not just a good outcome, but the best possible outcome where you just admit level targeting, it’s too fancy, it requires the ability to do too much fine-tuning. We’re just going to go back to the tried-and-true inflation-targeting practices.

That would have been fine, but we all know from studying the Fed Reserve that the details really matter. If they wanted to just bring back 2012, they could have just brought it back verbatim. Instead, again, when you look at the details, number one, in 2012, in the original strategy statement, the 2% inflation target was discussed first, and then they went on to discuss the objectives for employment. This time around, the order is reversed. The employment objectives are discussed first, suggesting that there is continued elevation of the employment mandate over the inflation mandate.

There is also language that is carried over from 2020. Again, it’s not quite the broad-based and inclusive language, but if you compare the way that employment is described as a social objective versus inflation, there still seems to be additional weight attached to employment. Then finally, the balanced approach language, thankfully—I think this is a good thing—was brought back. But it’s supplemented at the very end of the new statement with puzzling language that we try to parse out in our paper, which suggests that they’re going to try to implement a balanced approach, but then again, if the outlook for employment deteriorates, they’re going to be leaning toward the employment objective.

Maybe that’s just a clumsy way of saying, “Well, even in the absence of level targeting, we can use some makeup strategy implicitly to help.” In the paper, we allow for that generous interpretation. The question is, if that’s what it is, why not say that? Instead, is it just leaving the door open for a discretionary case-by-case response of exactly the kind that’s caused trouble in the past?

Luther: I would say when I parsed the words of the new framework statement, the thing that jumped out to me right away is that they removed this shortfalls language, but they didn’t reinsert the deviations language. Now, is that just because they thought that was unnecessary? I think if you just read the strategy statement, then that’s a fair interpretation, much as it would have been a fair interpretation with the 2020 framework to think that the framework was symmetric, that they were doing a symmetric makeup policy.

Just as you have gone back and looked at statements that were made in that 2020 framework process, if you go back and look at the Fed staff papers that were prepared for this more recent review, you will not see them say that this shortfalls approach was a problem. To the contrary, they will say, if we return to the effective lower bound, this shortfalls approach can be a useful strategy.

You get the impression that it’s not something that they’re quite ready to throw out, that they want to preserve that optionality. What do you do if you believe that? You get rid of the shortfalls language, but you don’t reinsert the deviations language.

Beckworth: Interesting.

Luther: You allow for that optionality should the need, in your view, arise. There’s the strategy statement, which you can read, but you can also read that strategy statement in light of all of the work that was prepared for this review. I think when you do the latter, it raises some questions about whether they’ve actually abandoned that shortfalls approach.

Beckworth: Well, what about the absence of any explicit balance sheet discussion in the statement? Because that seems to be a permanent fixture of monetary policy. It’s no longer an unconventional tool.

Ireland: For sure, that was a missed opportunity. It’s disappointing as well because many outsiders specifically called for this strategic review to address that question. I think, for example, of the paper that Bill English and Brian Sack presented at Brookings in a conference that was intended to be a pre-strategic review, ideas clearinghouse. There was another Brookings conference where Anna Cieslak, I believe, and her co-authors looked at the effects of quantitative easing, large-scale asset purchases, and suggested that there were difficulties communicating with the public what the strategy was.

Thinking hard about whether it is possible to articulate a systematic rule-like framework for QE in the same way that, say—the Fed isn’t going to commit to an interest rate rule, but at least in the monetary policy report to Congress, they present a suite of interest rate rules that can be used as benchmarks against which to evaluate the current setting for the funds rate. It would be really helpful to think about a rule for QE, and yet that was completely absent.

Beckworth: Going back to the process, we were all part of a conference that you organized, Will, I believe, with your organization. It was great. We got to get up there. Many of the same points we’re making here today, we made then, because we were doing our own unofficial listens event, I guess. We were hoping the Fed was listening to us. There was an official listen event. The Fed was soliciting papers, as you mentioned, from staff. There were also, Peter, as you mentioned, these presentations at Brookings. Brookings seems to be the unofficial think tank for the Fed outside the Fed. 

You mentioned Bill English, Brian Sacks, the Romers presented papers there. Even a Fed staffer for Michael Kiley had some pretty critical papers, which was really interesting, refreshing to see that happen. What did you think of the internal process itself? We have these outside events, your event, the Brookings events, but what about the actual main conference, the framework review conference, and the internal debates that the FOMC had?

Ireland: Going back to the issue of the staff papers, I think the staff papers were actually pretty good for what they were and all they could be. Each of them deals with an issue of obvious relevance to the strategic review, presents the arguments in a fair and balanced way without really leading the committee in one direction or another, so providing them with the background they need. What is missing, really, from those staff reports is a more critical and clear-eyed review of what went wrong with flexible average inflation targeting and, in particular, what went wrong post-2020.

We all know from our work in large bureaucratic institutions that it’s very hard for insiders to come to a meeting and just let loose with blistering criticisms. You do that and you get labeled as unhelpful, and then you don’t get invited to the meetings anymore, and you don’t get raises, and you don’t get promotions. It’s not that anybody is retaliating. That’s just the way things work in big bureaucratic institutions, but that underscores the need to bring outsiders into the picture.

If you take a look at the papers presented at the Thomas Laubach Conference, which was part of the review process, many of them were really quite good. Carl Walsh’s paper talked about the mistakes made post-2020 and how, in a troubling way, those same mistakes echo things that went wrong during the 1970s. There was a nice paper by Coibion and Gorodnichenko that looked at the behavior of expectations and basically made the point, you think that inflationary expectations remain anchored, but if you actually look at survey data, we’re not so sure.

The nice thing about having that conference is that it allows the staff papers to respond and say, look, I’m not saying that you responded. Carl Walsh and others have said that mistakes were made in a way that eerily resembles what happened during the 1970s. We must be cognizant of this. The difficulty was that the description of the strategic review came out in the first four FOMC meetings of 2025, whereas the Laubach Conference was not held until the end of May. Again, decisions had already been made by the committee, and the staff papers they couldn’t cite the Laubach Conference papers, they were written beforehand. 

Luther: There’s actually one exception to that. They cite Bernanke’s piece in the conversation over communications, which, by the way, was the only discussion that occurred after the Laubach meeting. Had all of those discussions taken place after the Laubach meeting, it’s reasonable to expect that those Fed staff papers would have engaged with those arguments as Peter is saying they should have done, but they couldn’t do that because it hadn’t happened yet.

Ireland: This is an issue of process, which you emphasized in your question, because the key elements are all there. The committee is discussing the key issues, the staff, the board staff, and staff from the Fed Reserve Banks are contributing usefully by summarizing the issues. The outside evaluators, so to speak, are being brought in as well, but in the wrong order, really. That’s what process is all about, I think.

Beckworth: Again, going back to the point I made earlier, as I recall, the review process was pushed back. It was supposed to start sooner. I wonder if it had started on time, had the conference been maybe scheduled at a different date, then we would have had the staff respond to all the points that had been raised.

Ireland: Maybe so. Maybe there were scheduling problems. That said, it still compromised the process.

Beckworth: At the end of the day, it was a choice they made to have the staff papers before the conference, and therefore not have the robust discussion in these papers and at the FOMC when these decisions were being made. 

Speaking to this process, this order, I remember back in November 2024, Jerome Powell sat down with Catherine Rampell, then at the Washington Post, and was at an event. She asked him about the review process because I believe it was supposed to have started by then, but it kept getting pushed off. He said his baseline—these were his words—his baseline is to have a framework that does not have makeup policy.

That was huge. I think I was one of the few people getting excited about it on Twitter at the time. There wasn’t a whole lot of discussion, but that was a clear signal of where they were going. Now, again, it could have been a return to 2012 version, which would have been okay, but it was a clear signal that there had already been some predetermined decisions made ahead of, I think, this Laubach conference.

Luther: Brian and I are both at a business school where folks in other departments outside of economics stress process and how important that is. It’s a bit of an oversight in economics. In economics, we solve the model for the equilibrium, and then that’s the right choice. That is the optimal outcome. What’s to talk about? What we miss in doing that is that you learned through this process. There were things that you didn’t consider and others did. Maybe they’re using a slightly different model than you were using. That’s why that process is important.

Whereas here, I think the Fed, which is very keen to give forward guidance, they wanted to give forward guidance about their framework. You have Powell in November saying, “We’re likely to get rid of this makeup policy.” In January, when they’ve had just the first meeting, the meeting where they were supposed to do an overview of what they were going to consider. In the press conference, he basically lays out what the revisions are going to be before any of these discussions about the deep components of the strategy statement were even had, before those Fed Listens events, before the research conference.

Literally, the very first meeting when they start talking about the framework, it looks like the conclusions are already there. There’s not really any process there. It looks like a performative transparency. It’s not real process where you’re going to generate ideas and discover what you’re going to do. Rather, it’s just going to allow you to say that you had a process even though that process didn’t really have any meat on the bones.

Beckworth: Just to go back and again make the charitable case for Jerome Powell and the Fed here. Then you guys can again push back and we’ll wrap this up. The Fed again feels scorned by this 40-year high inflation. They feel burned by FAIT. They feel like, maybe we made a mistake with the 2020 framework review. Let’s just keep it simple. Let’s just go back to flexible inflation targeting. What you guys are suggesting is, okay, that’s fair. Make that your default, but also make it an open conversation. Even if you have your preferences, still be open to other ideas, such as level targeting, nominal GDP targeting, and other things. Is that your critique?

Luther: That’s right. In the paper, there are some bright spots of this framework review as well. The work that the Cleveland Fed did with its Fed Listens event, we say in the paper, was very good. They structured it in a way where they were likely to get a lot of input from a lot of different constituencies. They had a mechanism for bringing that information out, compiling it, and sharing it.

All of those Fed Listens events should have done things like that. There are some bright spots in this process. We don’t want to overstate the case by any means. The problem is that those bright spots occurred in a much bigger process that wasn’t really, or at least doesn’t seem to be interested in generating ideas, actually reflecting on what went wrong. That’s a problem.

Cutsinger: I think, to put myself in the shoes of Fed officials thinking about what just happened, the one thing I might ask myself is, would this have happened if we had a symmetrical flexible average inflation target instead of an asymmetrical flexible average inflation target? I think that would have been a much better approach. We, as in the Fed, just spent the last five years trying to communicate to the public how flexible average inflation targeting works. Why go back to this other system that’s in some sense very different from what we’ve spent the last five years doing and just saying, hey, let’s take what we’re already doing and let’s make it symmetrical? That seems to me a relatively straightforward and introspective conclusion to draw from the experience from 2020 to 2025.

Beckworth: Yes. When I heard Jerome Powell say this in November ’24, and as it became clear they were going to go back to some version of a standard flexible inflation target, one concern I had was credibility. It’s like, okay, we’re going to go bold and big and brave. We’re going to do FAIT. Then five years later, oopsies, we’re going to go back to doing FIT. I was a little concerned that might undermine their credibility, but I’m not sure that it has, or is it that big of a deal? What do you think? Does it undermine their long-run inflation expectations? Does it say anything about them as an institution or not?

Ireland: Well, I think the bigger picture is the Fed Reserve, like a lot of prestigious or formerly prestigious American institutions, really took a hit to its credibility based on everything that we’ve seen since 2020. There’s nothing to be done about that. The question is how best to recover. I think you can’t expect someone to come out and say against a politically charged background, “Oh, it’s all our fault.” You have to find a way of finessing things.

Beckworth: Gotcha.

Ireland: One way of finessing the switch back to FIT would just be to say flexible average inflation targeting was designed with an environment in mind where the ZLB was going to be a serious threat to effective monetary policymaking. At present, that threat seems to be diminished. Then, as Brian said, the move forward would have been simple, just dust off the 2012 statement, bring it back, and say this was what we designed for normal times, and now we’re back to normal times.

The troubling thing is that what we got instead was a variant of 2012 that still seems to have a lot of the asymmetries and ambiguities that plagued the 2020 statement. It’s not just quibbling over process. In the paper, we try and tie shortcomings, weaknesses in the process, to recurrent problems that continue to appear in the strategy statement.

Hopes for the Future

Beckworth: Final word. What do you want to see happen in the future? Next review presumably is five years, four years from now. What do you want to see happen?

Ireland: My hope would be the Fed Reserve’s great strength historically has been its decentralized structure, which has given rise to an institution with multiple power centers, multiple thought centers. Mike Bordo and Ned Prescott have a series of papers—I think they may be writing a book on this—that suggests that, for example, the reserve banks have been generators of new ideas in monetary policymaking and financial regulatory policymaking as well. What I’d really like to see going forward is a return to a Fed that is truly diverse in exactly the sense of having multiple power centers, multiple sources of ideas, not shying away from debate, bringing outsiders in, and arguing over the basic issues.

Cutsinger: I would like to see them adopt an approach that makes aggregate demand shocks less likely. The reason I say that is not just because I think avoiding negative aggregate demand shocks or positive ones is good in and of itself, but there’s some research that suggests that after negative aggregate demand shocks, we get these populist backlashes where the voters end up supporting all these policies that end up leading to these negative aggregate supply shocks. Casey Mulligan’s book, The Redistribution Recession, would be a great example of all the labor market policies that were adopted by the Obama administration after the initial hit of the aggregate demand shock.

I think thinking about the broader political economy of your monetary policy framework matters. It matters because it can perhaps avoid populist backlashes against these shocks. It matters because it can actually take some of the pressure off of the Fed. I think one of the downsides with the current approach right now is that because they say they’re trying to do a balance, but it’s unclear how that balance actually works, that opens up the scope for political pressure from both Congress and the president to maybe put the thumb on the scale just a little bit as opposed to adopting a framework where the Fed basically says, given our framework, we can’t. We can’t put different weights on these because this is our actual framework. I think that that would deal with a lot of the institutional credibility that Peter alluded to, but would also deal with some of the broader political economy aspects of aggregate demand shortfalls in particular.

Luther: I would say my short-term hope is that Fed officials will be a little less constrained by their communication concerns. I’m a nerdy monetary economist. I spend a lot of time talking to other nerdy monetary economists, but my bellwether on the world is conversations with my aunt, who is clearly not a monetary economist. She lives in the real world. She does not know the particulars of the flexible average inflation target or a flexible inflation target. She couldn’t tell you how they differ. She can’t explain why the Fed should be targeting inflation.

The idea that there are just some things that the Fed cannot consider because the public won’t understand, I think we’ve got to scrap that idea because the public doesn’t understand what the Fed’s doing already. They’re not going to understand for the most part what the Fed decides to do in the future. What they’re going to know is whether that works well or not. Don’t be so constrained about that.

In the longer term, my hope is that we pay more attention, that Fed officials pay more attention to trend inflation and what that right trend inflation rate is. There is this consensus that 2% is ideal, and you can’t really question it. If you actually go to the literature, there’s not a lot of support for that view. Across the lion’s share of the literature is that zero inflation is optimal. Maybe you want to adjust that based on real-world data and have something more like 1% or 1.5%. That’s not even a conversation that is taking place at all.

The only exception to that I would point to is Larry White’s piece in the Southern Economic Journal on whether the Fed should raise its inflation target. He reviews the evidence there and says definitively no, but also suggests that maybe a lower rate would be better. I don’t know whether we would be better off with a 1% inflation rate or a 1.5% inflation rate, but I think that’s something we should be thinking a bit more about rather than just taking this 2% as given.

Beckworth: The next review, everything should be on the table then, including the actual number for the inflation target, not just the notion of an inflation target or some other version of a target.

Luther: Yes.

Beckworth: Absolutely.

Luther: Let’s have an open conversation with a solid process and consider those alternatives. Even if those alternatives are found wanting, they will inform the ultimate decisions. If a nominal spending target is on the table, even if it doesn’t get selected, it forces you to think a little more seriously about how what you do select responds to supply disruptions. Yes, let’s have those conversations.

Beckworth: Well, this has been a great conversation. The paper of these gentlemen is titled “Reviewing the Federal Reserve’s 2025 Monetary Policy Framework Review.” Peter, Brian, Will, thank you so much for coming on the program to discuss this work and to keep us abreast of what’s happening at the Fed.

Luther: Thanks for having us.

Cutsinger: Thanks for having us.

Beckworth: Macro Musings is produced by the Mercatus Center at George Mason University. Dive deeper into our research at mercatus.org/monetarypolicy. You can subscribe to the show on Apple Podcasts, Spotify, or your favorite podcast app. If you like this podcast, please consider giving us a rating and leaving a review. This helps other thoughtful people like you find the show. Find me on Twitter @DavidBeckworth, and follow the show @Macro_Musings.