Skanda Amarnath on the Future of the Federal Reserve and its Framework

Skanda Amarnath is the executive director of Employ America. Skanda returns to the show to discuss the standing of Humphrey’s Executor, the prospects for the Fed’s Framework Review, the case for NGDP Targeting, and much more. 

Subscribe to David's new Substack: Macroeconomic Policy Nexus

Check out our new AI chatbot: the Macro Musebot!

Read the full episode transcript:

This episode was recorded on April 16th, 2025

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, and I’m glad you’ve decided to join us.

Our guest today is Skanda Amarnath. Skanda is the executive director of Employ America and a close follower of the Federal Reserve System. He joins us today to discuss the ongoing Fed framework review. Skanda, welcome back to the podcast.

Skanda Amarnath: Thanks for having me again, David.

Beckworth: It’s great to have you here in the studio. You have two new articles out, really great, very timely. The first one is titled, “The Supply Problem in the Fed’s Framework - Part 1: What Should the Fed Do About Tariff Inflation?” Great hook, great angle, but important issue that both of us don’t think the Fed wrestles with enough in this framework review, and we’ll come back to that. Then, today, as we are recording this, April 16, you have a second article out. This one is titled, “The Supply Problem in the Fed’s Framework - Part 2: Did the Fed Miss a Productivity Boom? The Future Warrants More Attention & Agility.”

Again, both of these speak to the challenges of an inflation-targeting central bank when there are shocks or developments on the supply side of the economy. It’s tricky, it’s challenging, and you have addressed it in these pieces. Skanda, before we get into that conversation, it’s an important one, there’s another important conversation happening right now. I’m going to motivate it by reading an article that came out today in The Wall Street Journal from Greg Ip. It’s titled, “How a Supreme Court Ruling Could Weaken Fed Independence and Shake Markets.” Subtitle is, “Justices are revisiting ‘Humphrey’s Executor,’ a decision barring presidents from firing certain federal officials for purely political reasons.” Why should we as Fed watchers care about this case?

Humphrey’s Executor

Amarnath: We should care because the Fed, when we talk about it, we talk about it as an independent institution. Central bank independence is important. It is important for it to be able to make independent decisions from the political cycle, be operationally independent so that it can do what’s in the best interest of both the short and long run of the economy. That’s somewhat precious, and obviously, you can see in other domains of economic policymaking, when that does interfere with politics, you don’t always get the most rational result. You don’t always get the decisions that are in the long-run interest.

This case gets to one dimension of Fed independence, which is this amorphous blob of ideas, but it gets at one part of it, which is presidential removal. Now, there are court cases that have enabled, under the assumption of constitutional power that is vested with the president, the ability to remove, at will, officers of the United States. There is an inherent constitutional power to remove persons in the government and not be protected by statute. There are statutes on the books that say the Fed chair cannot be removed at will by the president, must be for cause. The question is whether that’s constitutional or not.

Obviously, the Fed is not currently in a place where they are directly involved in lawsuits, but there are court cases that are pending, basically, that do touch on these questions of does the president always have the ability to remove someone, and is there an at-will removal power that is inherent in the Constitution. There’s one particular case that protects against that, Humphrey’s Executor. If you say that ruling should be overturned, which is something that the Trump administration said the courts should be willing to do—they said it very explicitly.

Sarah Harris, who’s acting solicitor general now, she, in a memo to certain senators of the Judiciary Committee, communicated that point, that while we don’t think that we need to overrule Humphrey’s Executor to do all the removals that we’d like to do, if you need to do it, or if you feel like you need to do it, you should. That seems to be the case that is probably most relevant, if current President Trump can fire current Fed Chair, Jay Powell, for no specific cause, no for cause threshold, but just because there’s some sort of political disagreement.

Beckworth: This opens up the door for President Trump, or any future president, to stack the deck with governors and Fed chairs that would follow their desires. We can go back to the 1970s, there’s allegations that Arthur Burns succumbed to political pressure. That’s the concern, that this case could open the door for Trump and future presidents to politicize the Fed.

Amarnath: I think that is certainly among the top concerns here, which is that, for people who are very steeped in the discussions around central bank independence, there’s a sense of, in the ’70s, the Fed may have faced political pressure to take more accommodative action than they otherwise would have, that supported higher inflation, and maybe was too focused on lowering unemployment. There are various takes on this particular part of the story, but it’s one that definitely informs why we keep some insulation around these institutions.

I think maybe more directly, if you compare different parts of the government, and which parts of the government have more state capacity, more ability to engage in evidence-based reasoned decision-making that is reasonably apolitical, both in terms of how it makes decisions, but also the culture around it, so the bureaucracy around it, the political principles—I think if you take, Jay Powell is a Biden appointee, he’s also a Trump appointee, he is also an Obama appointee. He’s also a George H.W. Bush appointee.

Beckworth: He has a long history.

Amarnath: He has a long history there. There are people who have come through the Fed who see things because they’re trying to evaluate the economy. I think that’s true of everyone who’s on the current board of governors. There are seven seats. If you were to say that you had the ability to remove at will all the Fed governors, that obviously could lead to a very slanted outcome, where you say, if I wanted to get rid of the Democrats but keep the Republicans, you could quickly just end up with a much more lopsided partisan slate, or intentionally turn the Fed into a more partisan institution.

I think these are some of the questions that are probably going to vex the Supreme Court because the most recent set of cases has really harped on the fact that where they did find the president had some inherent constitutional power to fire agency heads, they specifically harped on single-member agency heads. In the case of the CFPB, in the case of the FHFA, there were two major cases there in which they found the fact that there was a single member leading it, as opposed to the Federal Reserve Board, the Federal Open Market Committee, which are obviously deliberative multimember bodies.

In the case of FHFA and CFPB, they found that because there was a single member, and there was some substantial executive power, I believe was the phrase used, and they found that obviously those agencies engage in regulation, because of those two facts, the president could remove them at will. The Fed board does engage in regulation too, but it’s multimember. The same thing applies to a couple other agencies where the lawsuit is involved. The FTC is another place where this is relevant. There’s the Merit Board. There is the NLRB, National Labor Relations Board. These are all multimember boards that are involved in some regulation. Because of that, there’s high-stakes questions about how much do the recent rulings apply to multimember bodies. If that’s true, then does it apply to the Fed as well?

Beckworth: Yes. This could be a very consequential case, maybe even more so than what we’re going to talk about next, the framework review. I noticed on X, both you and I responded to a tweet from someone who was like, “Meh, what’s the big deal? Worst-case scenario, the Fed is split in two, you take off. This might lead to the bank regulatory side of the Fed being siphoned off with this push to maybe consolidate federal banking, and then monetary policy does its own little thing.” You said, “Not so fast. Even if you take that benign view, or maybe best-case view, there are still reasons to worry, even then.” Right?

Amarnath: Yes, I am more fearful of the next several years. If you put a gun to my head and you tell me, “Will President Trump be allowed to fire Jay Powell and get the support of the Supreme Court?” I’d say no. But we are seeing a number of opinions, and if you think about the younger justices—so Justin Walker delivered an opinion for the DC Appeals Court, which was then revised within the DC Appeals Court later. He wrote a 2-1 opinion basically saying Humphrey’s Executor only applies to Humphrey. Maybe I’m being a little bit uncharitable there, but he was basically saying we should not weigh this opinion very much.

Beckworth: Oh, wow.

Amarnath: He did not make any carve-outs for the Fed. That’s, I think, the most important thing. When you think about where the puck might be going, and we think about where we’re trending potentially, if it is the case that future justices may not put as much of an emphasis on the importance of Fed independence. We’ve seen some evidence that people like John Roberts, Samuel Alito, Brett Kavanaugh are pretty sensitive to the Fed independence questions, such that I think that they will find a way to carve it out. If it’s carved out in an ad hoc manner, it’s not obvious to me that every single future justice is going to see things the same way, and especially when things are ad hoc, there is more of a propensity to try and revise the law subsequently.

I think that that is a very dicey question. All matters of executive, or any matter where an agency is involved, involves both some level of execution, but it also involves Congress delegating its power. You can say, “Wait, these powers Congress can delegate in this super special way, but these other powers Congress cannot delegate without the president having the ability to remove.” I think bigger picture is that, yes, there’s a lot of precious capacity, and we also want to have stable regulations, stable policy institutions. I think there are good conservative critiques of how the administrative state became more politicized, and the political cycle gets imbued into every single regulation. So when one president is in power, they say, “The regulation means X,” and then the next president comes in power and says, “No, it actually means Y.”

That’s not helpful. It’s good to have these questions be put on a much more stable plane. I think that’s what’s put at risk if we’re going down a more politicized route, or if we think that all institutions can only have democratic accountability if it flows through the president’s ability to fire you, as opposed to the fact that Congress can delegate power in somewhat more innovative ways that, again, hopefully, protect a certain apolitical culture. We would like to have thoughtful central bankers and people who are making decisions for the best interest of the country, and not necessarily because they’re worried about being fired.

Beckworth: Yes. Just one last thought before we go back to the framework. In the Constitution, the monetary authority is embedded with Congress. It’s not with the executive branch. This should be something that Congress has jurisdiction over, not the president, at least the monetary policy part. I think you can make an argument on some of the regulatory stuff, but that’s something that we need to be mindful of.

Again, I would look at this history and norms around the world. There’s a reason that most countries have ended up in an institutional setup where the central bank is somewhat independent because we don’t want to politicize monetary policy. Yes, there needs to be some sense of democratic accountability, but we don’t want every politician who gets elected to use it, because we see that outcome as well in developing markets and stuff. This is a setup that makes sense. There’s ways to fine-tune it. That’s what we’re going to talk about now, the framework itself.

The Fed’s Framework Review

That is something that’s under consideration now. Let’s segue back to that. Again, your first article to kick things off is titled, “The Supply Problem in the Fed’s Framework - Part 1: What Should the Fed Do About Tariff Inflation?” Now, Skanda, just to talk about where we are, we are in the midst of a Fed’s framework review. Most listeners will know what this is because I’ve talked about it probably too much, I’m sure some would say. Let’s go back to the Fed’s consensus statement, because they have this little paragraph at the very end that says this:

“The committee intends to review these principles and to make adjustments as appropriate at its annual organizational meeting each January, and to undertake roughly every five years a thorough public review of its monetary policy strategy, tools, and communication practices.” This consensus statement is like the Fed’s constitution, outlines what it’s to do, talks about the dual mandate. At the very bottom, it says, “Every five years, we’re going to revisit the strategy, the tools, the communication.” We’re in the midst of our second review.

The first one was 2020. Listeners will know that’s when FAIT came into existence. So far, we’ve had two FOMC meetings discussing it, January and March 2025. There’s a big conference coming up in May 15, May 16. This episode will air before then so people will be better informed walking into that conference because of our conversation today. Finally, we think that the final decision will be announced at Jackson Hole in August, probably like it was last time. Here’s where we are. I would also note there’s very little media coverage, or in my mind, relatively little media coverage compared to the importance of it.

Now, I understand there’s bigger fish to fry right now out there with the trade war, but that’s my sense of where we are. I think if I could read the tea leaves, it looks like to me they’re going to go from FAIT, flexible average inflation targeting, to something more like FIT, flexible inflation targeting. I say that because Jay Powell, last November, in an interview with Catherine Rampell, said his baseline case scenario is a simple reaction function that doesn’t make up for past misses. To me, that was huge news that wasn’t reported on, but that was huge.

Then the past two FOMC minutes have both said that members, participants want to relook and should reconsider what was adopted in 2020. Again, if I’m reading the tea leaves, it sure looks like they’re ready to abandon ship on FAIT, which I understand, they want a more robust framework, but I also worry they’re throwing out the makeup policy baby with the FAIT bathwater because there’s something to be said for that. With all that table setting, I want to turn it back over to you now. You have motivated your part of this conversation by looking at supply shocks, by tariffs in particular. Tell us about this, the way you’re thinking about the framework review.

Amarnath: I think that the Fed wants to use this sort of review to look back at the last five years, but ideally, with an eye toward what may be coming somewhat soon. Maybe it’ll take some time, but we’re dealing with what is still a lot of supply-side turbulence, actually. Right?

Beckworth: Yes.

Amarnath: I think we may have thought in 2024, we’re coming out of this period of turbulence. I think it’s safe to say in 2025. There’s at least a lot of uncertainty and a high risk of turbulence. We had that over the last five years, and it led to a variety of different phenomena coinciding. Among them, we started with the transitory debacle of the Fed characterizing inflation as transitory. That’s probably a term that people ran with from the media as a way to say, “Look how the short-term, people, inflation was.” I think it has a deeper meaning, but nevertheless, not great communications.

The Fed even acknowledged as much. “We threw away the transitory and responded decisively in 2022,” they will say. Nevertheless, we had these very confusing supply shocks through that period. Even there was obviously a huge demand component to inflation as well. We’re coming out of a depressed period of activity, suppressed services activity. Fiscal impulses were obviously very present. As a result, we can always debate how much was it demand, how much was it supply, how much was it one-time effects, how much do you need monetary policy to respond.

We did have big supply shocks at the Fed, at first thought, maybe this is actually all demand. Maybe this is all the kind of inflation where only a recessionary unemployment rate increase would solve it. Then they subsequently came around to the fact that actually we could get a soft landing. If you compare what Jay Powell was saying in September 2022, when he had a very somber tone and what he was saying in September 2023, when he was certainly more optimistic, a lot of what drove that change in thinking was, oh, actually, there were supply shocks that were muddying what the Fed was seeing.

It was probably a, in some ways, more fortuitous time for supply shocks that they were also happening in robust demand conditions. You had this correlation of the labor market looked pretty strong, and inflation was elevated as well or rising. That doesn’t really make a lot of difference how much supply, how much demand, and our preferred measures, nominal income growth, nominal consumer spending—these measures were also reasonably firm. It wasn’t that costly how much supply, how much demand, and they had some ability to correct. They have reduced interest rates in Q4 of last year or last three meetings.

The tricky thing comes in moments when there is more divergence. Let’s say demand isn’t quite so robust, but we do see supply shocks. We may see conditions in which inflation rises, but also weak activity. What matters more? How do you navigate them? Because those are actually the most important moments in a lot of ways. I don’t think it’s that surprising. People think of, “Oh, we had historic inflation now, so this is just so unique.” Actually, you go back, 2008, they were focused on inflation for much of that period until Lehman failed. Even after Lehman failed, they didn’t cut rates right away.

Beckworth: They still didn’t cut, yes.

Amarnath: They waited a couple of weeks to cut. Inflation was the big phenomenon. At the same time, nominal income growth was capitulating in ways that they could have seen in real time. They chose to ignore it because they were so worried about inflation, even though that inflation was really about supply. I think if people say, “We had the inflation we had in the last few years because it was demand,” there’s obviously a part of that story that’s true, but I’ll just note, you can have pretty firm inflation outcomes even as the demand deteriorates and the demand response might be more persistent than the supply.

If you go back to the year 2000, oil prices were surging at that time, too. That also turned out to be the eve of a more recessionary outcome. If we think back to Gulf War oil price shock as well, that also tends to be preceding/coinciding with bigger demand effects. I do think that that’s not just pure coincidence. It’s also about how it vexes central bankers into balancing these tradeoffs and communicating them to markets, to the public. These are, every time there’s some kind of balancing act, if you’re doing it purely discretionarily, it’s, one, really hard to get right, and, two, really hard to communicate. Markets may see inflation is high, why would you cut rates? 

On a different note, maybe we see real output is actually weak, but the labor market might be, largely speaking, in place, and inflation might be firmer. People are like, “The GDP is negative, so we obviously should cut.” It’s like, not necessarily, but that not necessarily, that’s where the communication strengths are really important. I think there’s probably a lot of folks in the building, at the Fed, who feel burned or some sense of buyer’s remorse for some of the changes that were made during the framework review. I actually think trying to fixate on taking this, what kind of inflation targeting, can really miss a big point that’s of high relevance right now, because we are dealing with potential supply shocks. They will create some weird constellations of data.

Beckworth: Yes. Just to be clear, the tariffs are a form of negative supply shocks, you can think of them that roughly, approximately, where inflation and the economy go in opposite direction. A negative supply shock means your real GDP is going to go down, unemployment may increase, but inflation is also going to go up. We don’t like those worlds, those situations. More importantly to the conversation today, it’s really hard to navigate them if you’re a central banker. 

I guess the point you’re making, which is such an excellent one, is once again, the Fed might be fighting the last war. The FAIT framework was looking back at the zero lower-bound world of the 2010s. It made sense for that battle. Then we had something else happen. Now, your concern, probably mine too, is they’re going to look back at FAIT and say, “Oh my goodness, why did we do this? We clearly need to think about all environments, including robust demand environments,” when in fact, what we might be experiencing this year is going to be very different than even that framework or that setting. Maybe more 2008. Maybe, hopefully not as bad as 2008, but the point being 2008, nominal income was going down.

The market signals like break-evens were going down. Everything was signaling weak real activity, but inflation was high because of oil prices. You missed the forest for the trees. That’s your concern, is, once again, if we go back to FIT, flexible inflation targeting, it’s this eternal struggle of how do you divine, how do you read the incoming inflation data? It could be demand-driven, which you would worry about that, but it’s supply-driven, so no, you don’t. It seems like they may be putting themselves back in a situation where they’re going to once again be struggling, right?

Amarnath: I think that’s a very high risk, and this is also what a framework is supposed to be for. It’s supposed to be for robustness. It’s about being robust to a variety of scenarios, like the problems of the last five years will not be identical to the next five years. Maybe it’ll be inflationary. Maybe it’ll be a different kind of inflation. Maybe we will get more recessionary forces, maybe neither. I do think it’s all of these risks need to be accounted for. I think this is the danger of just weighted toward one-risk scenario in this moment when I think Chair Powell has said it himself, that, one, we have underrated the supply side in both directions in the last few years.

Also, that we should be quite humble and feel a sense of deep uncertainty about the currents that are proposed/enacted policies—I’m talking about tariffs here—what kind of impact they’ll have on the economy. Maybe they will have a bigger impact on investment and financial markets, and maybe that’ll end up having more initially demand-side consequences. It could also have very obvious cost-push implications. We’ve seen how supply chains can break; break because of maybe strong demand, but also break because there are just blind spots. Oh, we didn’t realize we needed this input, and this input may come from a specific country. It turns out this may actually become a lot more problematic.

Beckworth: Yes, very consequential.

Amarnath: There’s a complexity to this stuff and a set of consequences that are just unforeseeable. Because of it, it’s important to think through all of these scenarios. I agree with you, there is a rational purpose for the average part of FAIT, the flexible average inflation targeting. Maybe it can be anchored to particular conditions. Maybe when you’re at the zero lower bound, maybe when you’re in a period of depressed activity and employment, when it’s very identifiable, like right after a recession, that there’s a reason to aim for some makeup or get back to trend on some set of benchmarks.

The set of benchmarks you use could vary. It’s as much art as science there, but there’s a purpose for it that is really important. Saying average versus not average can also just miss, hey, when we’re evaluating inflation, even when we say inflation targeting, there’s a sense of short run versus long run. I think what most Fed economists will tell you is, “We’re still being flexible. Inflation targeting will give us space to respond to higher slack. It’ll give us space to respond if there’s a recession, we’ll be flexible about inflation in that environment.”

How flexible? The Fed did cut rates quite dramatically in the year 2001. In 2002, we went from a 6% Fed funds rate down to 1%, and yet it was all a little, too little too late. That was the real risk, the issue there. While we may be in a place where interest rates are higher now than they were in 2019, when the Fed was doing its previous framework review, the notion that we may end up in a zero lower-bound environment in the future, the risk that we might end up where Fed policy may still be vulnerable to a lot of the mistakes of the past.

It’s just important to think about what are the cleanest ways to both execute policy, and also ultimately most of what monetary policy does is markets are trying to front-run it. Markets have to have a good sense of what’s the reaction function, what variables are of most importance to the Fed. Especially in a moment of supply shock, the most important data point may not be inflation. Even if you’re trying to target long-run price stability, just, there are things that can push up inflation. We saw a surge in commodity prices in 2008 that the Fed was itself distracted by, I would argue.

Beckworth: Very distracted by.

Amarnath: They missed the forest for the trees in that moment. Markets follow the Fed. Markets priced in.

Beckworth: No, absolutely.

Amarnath: A very different reality.

Beckworth: That is the concern and the practical side of this is you can say, “We target median-run inflation at 2%. Yes, we’re flexible,” but in reality, when push comes to shove, most central bankers will err on the side of what’s happening to inflation right now. Here and now, even if they’re not sure what’s driving it. That, to me, 2008 is a good example of this. They were worried about inflation. As late as August, I know the minutes said the balance of risk is toward more inflation. We’re going to destabilize inflation expectations. They were worried. You mentioned September 2008. They still didn’t cut rates.

Bernanke in his book says that was one of his biggest regrets. It was this fixation on inflation as opposed to what’s happening to nominal income in the economy. We live in a dollar-denominated economy. It was collapsing, and we were hung up on temporary spike in inflation. I fear that’s going to happen again. We go back to FIT, flexible inflation targeting. It seems easy. It seems natural. “That’s what people worry about, David.” I have people who follow the show, they’re like, “I love of your podcast, David, but you talk way too much about nominal income targeting, nominal GDP targeting.” I keep saying to them, “Look, we’re going to end up in the same place that we’ve been before, where on a practical level, it’s just easy for central bankers to get caught up in the moment with what’s happening this month on the CPI print, as opposed to, let’s look over the medium to longer term.”

Yes, Fed officials will say, “Oh, we look at long-term inflation expectations,” but in the near term, we have all this evidence. You mentioned the Fed. Fed Fund futures, they spiked in 2008 because they were reading what the Fed was saying. ECB actually did tighten. 2011, ECB, even better story. I think two times.

Amarnath: 2008 and 2011, both for oil price/supply-shock reasons.

Beckworth: I argue they were a key part of the Eurozone crisis. Yes, there was debt and other issues, but they were lighting the match that triggered this flame in Europe. I still don’t think there’s a lot of soul-searching that’s been done over that incident. The fact that we’re potentially going back to a flexible inflation target, it assumes central bankers are going to be omniscient. Oh, yes, we’ll divine, this is really a supply disturbance and we’ll be smart about it. In practice, time after time, they get caught up in here and now. I think, to me, that’s one of the appealing arguments for a nominal income target or nominal GDP target is you acknowledge your ignorance. We don’t know. We step back. Neither of us think this is going to happen. Yes, but at a minimum, and we’ll get to this later, let’s at least cross-check what we’re doing here with the Taylor rule, with these aggregates.

Amarnath: I think it can actually serve as a valuable—we’re not going to change to a nominal income targeting world—it can be a very valuable communication device. I think that you can’t look at the past four years as anything other than a communications debacle. And I have a lot of empathy for the fact that this is a very tricky moment to navigate; the Fed has navigated pretty admirably. I’m not here to bash the Fed, but they started with transitory. There was a lot of good reasons to. The word is, again, loaded in central bank parlance versus maybe more public understandings of the word.

Nevertheless, they obviously scrapped the word. “We usually look at core inflation, but actually, these supply chains are really messed up. Maybe there are some good services, consumption switching. Let’s focus on core services. Core services will give better guide to, because the good side is really messed up, the services side is elevated. That’s a reason to respond.” Then it turned out, wait, hang on a sec. There’s one part of services that’s really high weight and it lags a lot. We know that, say, rent inflation is poised to slow. We see it in the market rent data.

It turns out this inflation data that we use for CPI and PCE lags by about 18 months, maybe two years in some sense. We actually need to look at nonhousing services, nonenergy, nonhousing services. Okay. What we actually find out in the PCE mix is that these are also prices where there are a lot of weird bits. There are things that just basically follow the stock market. When the stock market is doing well, that’s going to push up some financial services prices. There are some parts that are basically input cost indices for nonprofit services that are provided.

These are a function of some wages here and there. It’s not exactly market prices. Then Jay Powell has said very recently, I’d say a few meetings ago, we’re actually focused on market-based, core, services, ex-housing prices. You’re talking about maybe food services, airfares, healthcare. Are we talking about a representative basket here? Does this sound like clear communication to you?

Beckworth: No.

Amarnath: I understand why they’re doing it.

Beckworth: Sure.

Amarnath: I do think there is value to building it, seeing how the sausage is made with inflation that has a certain informative truth to it. If anything, it tells you you’re going to need more information. We’re going to need a bigger boat. We’re going to need a more robust suite of information to really evaluate inflationary pressures than just trying to do a lot of the slicing and dicing. I’ll just say the trade policy effects that we may be dealing with here, contrary to the belief that maybe the good side, tradeable side will be the part that gets affected, but the nontradable side won’t be and so we can look at that, a lot of this is a very instructive example. 

We learned there was obviously both maybe some firmer demand dynamics. Also, really, we saw automobile production went into a two-year depression. In that time, yes, used car prices spiked. New car prices rose. Parts rose, but so did the cost of a lot of services that are tied to it. Think about rental, leasing, repair, insurance. These are all tied to the value of automobiles, auto parts. One supply chain bottleneck can also have services impacts. We could be dealing with those kinds of shocks again now.

We may, we may not. That itself is instructive about do you want to fixate on spot inflation data. If not, you have to give some guidance for what that other thing is. For us, I’d say, even just as a guidance tool, a communications tool, nominal consumer spending data is very useful because it tells you, are people really spending for this stuff, or are we just lifting the prices because the costs have gone up, there’s a shortage? That is useful information. Knowing how much consumers have available to spend, typically speaking, most of consumption is funded through income. Some of it’s wealth, some of it’s some specific transfers. By and large, cyclical consumer spending is funded through cyclical income growth, which is really labor income growth. Right?

Beckworth: Right.

Amarnath: We should evaluate, what’s the trends in job growth and wage growth. These are things that we can evaluate in real time that will give us a pretty good guide for something that’s more proximate to the demand side of the economy, more proximate to what the Fed’s policy tools are affecting. I think you and I would both settle for something that was a nominal consumer spending target, or a nominal labor income target.

Beckworth: Absolutely.

Amarnath: Even if it’s not quite the aggregate that you and I might suggest, I think those are actually quite close to what is relevant to inflation, which is how much are consumers spending on these consumption goods and services?

Beckworth: We get monthly measures of those, right?

Amarnath: Yes.

Beckworth: We get the aggregate payroll index. There’s other ones. That one, for example, it’s a monthly measure of aggregate labor income. Then we get the monthly PCE. We can even do a market-based PCE if Powell wants that, right?

Amarnath: Yes. Look, there are probably a variety of measures to use for consumer spending. You can use retail sales, because it comes out earlier, plus complement it with some services measures. You can look at labor income. There’s a variety of different ways you can put this together. It’s just a good way to be robust about what’s going on, as opposed to telling people, “Actually, don’t look at this component. Look at this component.” I think that’s a pretty fragile place to be, precisely when you think about market psychology, where if you talk to a lot of market economists in 2021, and I’ll include myself in this sense, we knew there was going to be an inflation bulge.

We also said, “This is inflation bulge. Okay, it’ll lift the inflation forecast a bit. We’ll have to just see through it.” What happens when inflation exceeds your forecast? I’m sure in early 2008, the Fed was still baking in some elevated inflation because commodity prices had risen even in the prior years. Inflation had run a little bit above target for a persistent period of time. At that time, it wasn’t a formal target, but there was an understanding it was 2% PC inflation, roughly. The Fed probably said, “I expect some commodity price volatility, but not this much. If it’s this much, then that must mean we are way behind the curve.” 

That is typically a thing that happens, where if you say, “This is my inflation target, I’m going to be flexible about it,” if you didn’t bake in enough flexibility ex-ante, and no one has perfect omniscience, there very quickly becomes a lot of internal pressure and external pressure to do basically an overcorrection. To say, “Oh, actually, I totally missed how oil prices could get to 120, 140. Therefore, I need to really swing the other way. I need to now be showing that I’m more credible, more forceful on inflation.” Is that actually the right way to be setting policy? Markets will fixate on the spot data point that the Fed gives them, right?

Beckworth: Yes.

Amarnath: If the Fed says, “We’re targeting 2% PCE inflation with no real communication devices,” markets are going to look at every single PCE inflation print. I look at every single PCE inflation print.

Beckworth: Right. If you’re going to make money, that’s where your living is, you got to go there.

Amarnath: Is that really the best way to make good long-run policy?

Beckworth: No, absolutely not.

Amarnath: I think that’s where it gets tricky.

Beckworth: Yes, and that’s what I like about your article is it’s not just about cross-checking yourself, being more robust; it also requires good communication. You’ve got to do both. If you have this communication where you’re making it up as you go along, actually, it’s this component. Next month, it’s going to be a different component. What do you do? I think you’re absolutely right. Look at something that acknowledges our ignorance, that measures what really is driving spending and nominal incomes.

Again, just at a minimum, do a cross-check. It’s not that hard to do. A previous guest of the show, a friend of our program here, Evan Koenig from the Dallas Fed, when he was still at the Dallas Fed, he did a little exercise with some of his colleagues. I believe it was in early 2022, and it was published on the Dallas Fed website, where they just did a simple forecast of nominal income based on the consensus forecast, and it blew up. Little cross-checks like that, okay, maybe we need to look at a number of indicators, and it didn’t happen.

Fed’s Communication

Communication, let’s go there. Communicate, communicate, communicate. This is supposed to be a part of their framework review, but it’s going to be very tricky, I think, if you go back to a flexible inflation target. Let me throw out an example I’ve mentioned many times in this podcast. I believe it was around QE2, Bernanke was before Congress, and he said one of the objectives is to raise the inflation rate. The GOP at that House, probably House Financial Services Committee, they blew up, like, “What are you talking about? Why would you want to raise inflation when the economy is so depressed? It’s just going to further erode purchasing power, people are going to be ticked off.”

That was a fair, I think, response to a poorly communicated goal. What Bernanke really meant is we want to restore nominal demand, we want to restore nominal incomes to where they should be, not at the depressed level. If he had said that, I think, would have resonated much better than this price story.

Amarnath: I think, especially if you’re using both an income and a spending metric. I think that this notion of how could you possibly communicate when the economy is too hot, that you want to curb income growth? Especially if you’re anchoring in terms of spending. I think there are enough people who believe, say, too much government spending could be inflationary. I think that’s a majoritarian view, and that comes with a certain political valence to it, but it’s clearly legible to a lot of people. To be able to say that the volume of nominal consumer spending is something relevant for the inflation risk profile is a very straightforward translation.

I think that is not something the Fed should fear when saying we are valuing nominal consumer spending trends to understand is this really supply, is this really demand? If this is something where it’s running so hot that we do need to worry about inflationary consequences, is this something running actually cooler even as inflation’s running strong and we actually need to start being more attentive to? Because the tragedy of that, one, macroeconomics, business cycle macro is about a lot of sparse events. It’s these big episodes that really count.

These big episodes where we don’t necessarily have super high sample size, but they’re high leverage. In 2008, we had this conflict of both demand-side shocks from the housing financial mess and also the supply-side issues on commodity prices, and there had to be a way of navigating that moment well. We didn’t navigate that moment well. Objectively, that proved to be macroeconomically tragic, one, because obviously it’s a supply shock that the Fed got juked out from. They got bamboozled by it, and it had supply-side consequence too.

I think ultimately, there is a good reason to think that if we were able to avoid such a precipitous fall and maybe encourage a more robust recovery in nominal consumer spending back to its pre-crisis trend, it would have been not just that maybe we would’ve got a mildly higher inflation, but there was also real output losses that could have been foregone. We could have actually preserved more capacity, maybe had better labor market outcomes, better supply-side outcomes in the process. Because ultimately, we’re talking about hysteresis here. That recessions can do damage to the supply side.

It’s probably why a lot of what looks like recessions has both a demand- and a supply-side component to it. Demand-side shocks to the financial sector can ultimately restrain credit creation to both preserve and expand capacity. These are all things that are interconnected, and that’s why it counts. It really counts to get these moments right. We talked a lot about 2008, but I’m glad you brought up 2011. 2011 was a moment when the Fed, I think, was trying to signal in a more dovish direction. They were trying to be more imaginative about the balance sheet, but we were going through a period wherein commodity prices were rising.

There was commodity price inflation, and if you looked at what was priced into Fed Funds futures, you see a similar story. You see, one, the Fed’s got to hike soon because inflation’s printing well above the Fed’s 2% target, maybe for reasons that are really about headline inflation more than core inflation, but that’s untenable. How can the Fed possibly keep rates at zero in this environment? It makes sense if you think the nominal incomes are still very depressed versus its not-so-distant trend, and that’s why a catch-up policy is valuable.

It’s also why it’s important to have good communication about this. You have to guide the market and the public toward what you want them to pay attention to, or else they will fixate on every single monthly inflation print to a point of almost religious conviction that if three inflation prints come in hot, that must mean the Fed’s got to hike. If they come in cold, boom, we’ve got rate cuts. There’s macroeconomies more than just three PCE inflation prints, as someone who fixates on PCE inflation prints for a living partly.

Beckworth: I listen to you every morning that time of the month when inflation comes out, you do a little Spaces event on X and you talk about this stuff, but it does say something about us that I have to tune in and listen to Skanda whenever the new CPI print comes out. It’d be nice if the markets wouldn’t get so turbulent based on that print, one print. If, in fact, we are doing flexible inflation targeting, it’s supposed to stabilize inflation over the medium run. They would all tell you that’s textbook, but we get so caught up in the moment.

Again, I think it’s just human nature to get caught up in growth rates, recency bias. It’s there in front of us, and we’ve got to react. The Fed, again, the target, their framework itself feeds into this, the communication. It’s such a big deal. I want to read something from Chair Powell that speaks to the issues we’ve been touching on. This is from a speech he gave in November 2023, and this was at a monetary policy panel at the IMF in Washington, DC. He gave, I think, the opening remarks. He gives a lessons learned, a tentative lessons learned at this point.

He addresses several questions, and the second question is the one that we’ve touched on. He goes, “Turning to my second question, for many years it has been generally thought that monetary policy should limit its response or ‘look through’ supply shocks to the extent that they are temporary and idiosyncratic.” He justifies why that’s the case. Then he goes on to say this a few lines down, “Our experience since 2020, highlights some limits of that thinking. To begin with, it can be challenging to disentangle supply shocks from demand shocks in real time.”

That there is the problem. You’re trying to disentangle in real time, because you’re focused on inflation, was that demand-driven or was that supply-driven? I think that’s an interesting question maybe in retrospect, but in real time you need to be looking at what’s happening to incomes, what’s happening to spendings. That would be the way to maybe not have to play God and divine the source of the inflation movement.

Amarnath: Yes, and I think what Chair Powell said and what you quoted is the case for nominal aggregates being your target, and specifically around income and spending. Especially as it pertains to your nominal consumer spending measures, your nominal labor income growth measures, maybe personal income. There are a few different ways to go about it, and I’m sure the Fed would triangulate if they did formalize this. I think it’s important to formalize it, especially because we came through a period now where we were fortunate, but also you can think of it also as margin for error, which is nominal income growth was robust.

Therefore, the risk of when you’re raising rates to curb demand causing real balance sheet impairment, really tightening financial conditions in a nonlinear manner. One thing that’s actually very impressive is the 2022, 2023 period when the Fed was tightening, I think there’s a lot of evidence of restrictive impact, but it did not cause a big credit crisis. We see that. We saw credit spreads blow out at various points, but it wasn’t of the kind that we would associate with, hey, the financial market plumbing can’t withstand this. There’s obviously Silicon Valley Bank, and again, the Fed was able to come up with some targeted solutions there. I’m not going to paper over that.

I think that’s an important part of this experience as well, but it is telling. That there is more room, so you make less big policy errors, and you can navigate those moments, and you can get to something that feels approximately a soft landing. I think obviously we’re now in a new phase, but I think that’s telling, and it helps give the Fed some credit for the fact that they were able to be decisive.

If they formalize it now, one, it helps you make sense of the last few years where the Fed ultimately did move decisively in 2022, even if it took a little bit of extra time relative to what some Fed economists would have wanted, but also going forward now, which is these might diverge. You want to make sure that’s legible to the broader audience of market participants, the public, Congress. I think now is really the right moment to do so. I don’t think that has to be inherently political.

I would also flag, as part of the piece I put out today regarding the productivity performance we’ve seen over the last five years, and really six years inclusive of some revisions in 2019, is much better than what the Fed had forecasted back in December 2019. At that time, we also had a trade conflict. It was, I’d say, a smaller scale and scope, but I think it’s worth remembering that just because good surprises and bad surprises happen—we don’t know when, we don’t know how much—it’s important to be prepared for both of those when thinking about what’s coming.

As it currently stands, I don’t get a sense when we’re tinkering between saying we’ve got to take out the average from FAIT, one, I think there are some real risks associated with that in terms of what that means for how you’re going to guide on both how you’re achieving your inflation target and also what policy rates are supposed to do in response to it. Also, there’s a bigger set of issues out here. There’s a whole set of issues that I think gets to the communications challenges the Fed has had to face every single year since the pandemic, basically.

Productivity

Beckworth: Yes. You raise a great point. I’m glad you brought this up. It’s tied to your second article on productivity. We want a framework or an approach to monetary policy that’s robust to both positive and negative supply shocks. We don’t want to be, “What do we do? Inflation’s dropping too low. Is it good? Is it bad? Inflation’s too high. Is it good? Is it bad?” These questions come up on both sides. If you look at the FOMC minutes that they discuss their framework, the participants say this. They say, “Look, we want a framework that’s robust to all environments, inflationary, deflationary.”

Great. Guess what? Looking at nominal labor income or PCE expenditures, that is a way to address that question. In the near term, you may not know if it’s a sustainable productivity or if it’s a permanent one. Don’t try to play God. Keep it simple. Accept humility. I want to go back to that Powell speech briefly here, because in addition to his concern about not knowing how to know in real time the demand for supply shocks, he brings up a second concern of looking through supply shocks. He says, “Supply shocks that drive inflation high enough for long enough can affect longer-term inflation expectations of households and businesses. Monetary policy must forthrightly address any risk of a potential de-anchoring of inflation expectations as well as anchored expectations to help bring about inflation to target.”

His second concern is that, okay, we look through supply shocks, but what happens if there’s a bunch of supply shocks? Oh man, we might de-anchor the economy. My reply to him and to anyone who might ask is, “Well, that’s what looking and anchoring on the, say, labor income does.” If you stabilize labor income growth, you have a nominal anchor. You can look through the supply shocks. It solves both problems.

I’ll plug an article I recently published with my colleague, Pat Horan. This is in the Journal of Policy Modeling. The title is, “A Two-for-One Deal: Targeting Nominal GDP to Create a Supply-Shock Robust Inflation Target.” You literally get a nominal anchor. You get an inflation target that can deal with supply shocks over the medium run. You can have your cake and eat it too. Nominal anchor, and you’re able to see through supply shocks.

Amarnath: I know we keep talking about 2008. Maybe I’ll talk about 2004 to 2008 is a very good example, because we did see persistent commodity price inflation. If you look back at every commodity price of relevance—nickel, zinc, a bunch of ag prices, oil, natural gas—they made record highs during this period. There was a lot of persistent inflation because there was a China boom going on. The China boom was also fueling a global growth boom. There was just a lot of demand for commodities overseas, and that pushed up prices. That’s a form of domestic supply shocks, because commodities aren’t as available.

If you just look at inflation expectations based on that, you’re going to miss a lot. I grant that, obviously, the Fed has to take inflation expectations seriously, but it’s inflation expectations and it’s income expectations together. Unfortunately, we have some surveys of inflation expectations and don’t have as many surveys of income expectations. We have a break-evens inflation market. If you want to use nominal income, you probably look at this. We have 100 to maybe be a proxy for that, but I think these are what really matter. Because if you are not growing your incomes, but inflation is rising, you are imposing real income declines, right?

Beckworth: Absolutely.

Amarnath: How sustainable is that? How much can that feed on itself as an inflationary spiral? It’s always got to have some connection back for income to keep recycling through the economy. If you’re charging me more money, and then I’m just spending the same amount, I’m going to get less stuff. I might have to spend more for that stuff. If that’s the case, then my capacity to keep spending is going to be diminished. That connection is really important. Again, that’s why I said I would love to, in their consensus statement itself, there are ways they could incorporate some nod to nominal income and nominal spending measures. Nominal consumer spending and probably nominal household income measures. Probably a variety you can put together, but that gets that income spending nexus. 

That’s the part where if you’re not getting to that nexus, just saying there’s an inflation spiral risk because—I don’t know, right now we have the University of Michigan survey where surveyed inflation expectations are spiking. You know what’s also going down? Their expectations for income, so it’s a hard thing to balance. It tells you that there’s a diverging data set emerging.

We are likely to face, again, very weird constellations of data over the next five years. I think it’s really important that the Fed think carefully about this and not just treat it as, “We should just be a little more tough on inflation.” I think that is the cheap answer to the last five years, not thinking through what happens in a variety of different data constellations and macroeconomic environments we’re talking about.

Beckworth: Great points. The inflation is ultimately tied to the nominal size of the economy, nominal income growth, nominal spending. Again, if you anchor that one part, you’ve done your job and you allow inflation to temporarily go up, go down. It will, over the medium run, be anchored. In fact, the idea for the paper that we wrote, the “Two-for-One Deal,” I actually got it from Michael Woodford, of all people. He made this very point in an earlier paper that if you target something like a nominal GDP-level target or something like that, you are effectively targeting medium-run inflation.

You just adjust your target gradually over time as potential real GDP or whatever measure of capacity you have, and it will be good enough. Now, I want to go to a point you mentioned, what could the Fed look at for expectations for nominal income? There are, you’re right, fewer resources, but I bet if that were something the Fed did, they would find a way to create a new resource or they would bring other data together. The data is somewhat endogenous to your target.

Amarnath: I think that’s right. There is an endogeneity. It’s not as if we are poor in terms of data here. America still has, knock on wood, leading data collection capabilities. Every Friday, we can get a pretty good read on how our assessments of nominal labor income growth has evolved with the jobs report. Today we got retail sales. That gives you a pretty good guide for where consumer spending is trending. We can compare the revisions. There’s some mess to it, but it’s not something that’s impossible for us to figure out this is how our assessment’s evolving. We can do the same thing with the broader consumer spending release.

There are other ways to measure nominal labor income growth as well. There’s some other wage data we might want to incorporate. There may be some other types of job growth data. We have it, actually. This is already pretty good. If we want to develop a version of break-evens but for nominal income, that might be harder, but it’s not impossible. There are certainly asset prices that are basically a function of nominal income growth expectations. It’s not that hard to replicate. We won’t be flying blind. In a lot of ways, I think we’re going to be addressing the big blind spot.

I think it’s somewhat on easy mode for monetary policy. I’m not saying it’s easy. It actually achieved a lot of what the Fed has managed to achieve in the last couple of years, but it is a little easier when there isn’t really an obvious tension in the data set, where there isn’t a tension between dual mandate objectives. Employment’s strong, inflation’s high, you can probably lean a little more on the restrictive side. Inflation’s low, employment’s falling, easy enough. We should probably be using policy. Those are all things that you can fix. The implied nominal income dynamic there is pretty straightforward.

What do you do when these things are in tension and you have to do some balancing? The ability to combine that in a coherent way is what nominal aggregates represent. Maybe use nominal consumer spending, nominal labor income, or something roughly equivalent. Again, all these nominal aggregates are, roughly speaking, correlated. They should reflect the same thing we’re talking about, but these are all ways in which you can really get at that answer in the tough moments. It’s those tough moments that tend to be business-cycle relevant. They’re the ones that are highest leverage, because the risk of a policy mistake is high. It’s just hard. It’s really hard to interpret data.

The more you can clean that up, the more you can communicate that in a clean manner, I’d say there’s less risk. I don’t think that the Fed really wants to get into this business of this component, that component of inflation. “Actually, we’re stripping out this, because we realized it included portfolio management services, which is tracking the S&P 500. We were not trying to do that. What we really wanted was something labor sensitive, so we ended up indexing on these other things that track deposit rates that banks charge.” We’ll go in circles all day if we do that.

Beckworth: Absolutely. There was definitely a sense there was a moving target of what exactly inflation was during that period when it was rising. 

Skanda, this has been great covering these issues. To circle back to our first part of the conversation on Fed independence, I would suggest that the framework itself can affect independence. If it looks like the Fed is making it up as it goes along or if it’s struggling, that too can undermine that independence.

Amarnath: I think that the framework is valuable for making policy decisions more legible. That’s really important in a moment that feels politically fragile and polarized. There will be people who will complain no matter what the Fed does. That is somewhat inevitable. You can try and mitigate that by being transparent. If the Fed has to cut in the next six months because there’s actually a bigger demand-side shock, but there’s also higher inflation, there’s going to be a sense of, maybe the Fed’s being politically manipulated. There’ll be those calls.

The more you do the work up front through the framework review, the more you can make those issues more legible. I’d say, we’ve talked a lot about Fed independence, but I think it’s important to say the Fed’s supposed to be accountable to Congress. That’s the mechanism of democratic accountability. I do think that Chair Powell takes that mechanism very seriously in terms of understanding how both Democrats and Republicans can understand what the Fed’s doing. It doesn’t think it’s in any way foul play or being overly political. They have objectives to balance and it’s really delicate.

For that reason, there’s ways to do that work up front. Maybe things related to tariffs may prove to be more cost-push, but not necessarily demand negative. Maybe it may be the case that there are other positive forces in the economy beyond tariffs. We’ve talked a lot about it. I think we both fear it, but there may be positive things that the administration does or positive things that are just going on with technological forces that may lead to better outcomes that either allow the Fed to keep policy rates where they are, or maybe they’ll have to cut, maybe they’ll have to hike.

That legibility is really important for political legitimacy reasons. I think that’s a very precious thing, that, again, the framework reviews should be seen as an opportunity. Obviously, they still need to be restrained to their mandated objectives, what Congress gave them, but I think there’s a way to really take a richer view and do the hard yards up front so they don’t get caught in a bigger political pickle later.

Beckworth: Yes. Everything we’ve talked about today is for the Fed’s own benefit. On that high note, our time is up. Our guest today has been Skanda Amarnath. Check out his two articles. We’ll have them in the transcript and the show notes. Skanda, thank you for coming on the program once again.

Amarnath: Thanks for having me, David.

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.