Scott Sumner on Monetary Policy Confusion in Our Current Policy Debates

Scott Sumner is the Ralph G. Hawtrey Chair Emeritus of Monetary Policy and the founder of the Monetary Policy Program at Mercatus. Scott returns to the show, to discuss his life post Mercatus, nominal GDP counterfactuals of the pandemic and the Great Financial Crisis, the role of QE in inflation, the fears about Fed independence, and much more. 

Subscribe to David's new Substack: Macroeconomic Policy Nexus

Read the full episode transcript:

This episode was recorded on January 15th, 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 decided to join us.

Our guest today is Scott Sumner. Scott is the Ralph G. Hawtrey Chair Emeritus of Monetary Policy and my former colleague here at the Mercatus Center. Scott joins us today to catch us up to speed on what he has been doing, as well as his thoughts on US monetary policy over the past few years. Scott, welcome back to the show.

Scott Sumner: Thanks for inviting me, David. Good to be here.

Beckworth: Good to have you on. It’s going to be great to catch up with you. It’s been a while. I will remind the listeners that you were the very first guest on Macro Musings, Scott, so you blazed the trail of all these 500-plus episodes that have followed. Also, it’s worth pointing out, and I’ve said this before, but I’ll say it again, Scott founded the Monetary Policy Program at the Mercatus Center. Later, Scott hired me, brought me on, and Scott and the rest of the people at Mercatus took a chance on this podcast. Scott, part of your legacy is this podcast. Thanks for all that you’ve done for the show over the years.

Sumner: Thank you. I think bringing you on board was the best move we made, so glad to be here.

Scott’s Life Post Mercatus

Beckworth: Oh, it’s been great working at Mercatus. I got to spend several years with you, and we try to continue to carry the torch. Now, you’ve been gone a few years, so tell us what you’ve been up to. Now, listeners, Scott does have a Substack. He does still write on monetary policy, but I know you also have been traveling, seeing the world, so tell us about your adventures.

Sumner: Yes, I do some traveling. Usually, every year or two, I go to China and sometimes other East Asian countries like Japan, Taiwan. I’ve been to Europe. I was there during October in France and Spain. Yes, we like to travel. 

I’ve got a new Substack blog that I started about a year and a half ago called The Pursuit of Happiness. I’m still doing a lot of blogging, still focused on monetary macro, although I blog on other topics as well, like movie reviews, and various things like that. I guess on blogging and macro, I was thinking about this just yesterday, how would I frame what I’m trying to do. I think part of it is simplifying a field that I think has gotten overly complex and boiling it down to the essentials.

I’ll just give you three quick examples. One would be really focusing on nominal GDP as the important indicator of macro policy. If you think of macro as three areas—long-run economic growth, long-run inflation, and the business cycle—I think two of those three are basically driven by nominal GDP and especially nominal GDP mistakes. In macro, we often study the mistakes, too slow growth in nominal GDP in the Great Recession, too much in 2022. I think that there’s too much focus on other factors: real shocks, fiscal policy, things like that. A focus on nominal GDP is, I think, the best way to understand why mistakes occur in the macro economy.

The second thing I would simplify is, I would focus mostly on monetary policy as a determinant of nominal GDP growth. Again, I think fiscal policy is overrated in importance. Supply shocks, I think, are overrated, especially in the recent inflation we had around 2021, ’22. If you look at the trends in nominal GDP, you can pretty much explain most of the business cycles and most of the overshoots in inflation. One notable exception is COVID, which was obviously a massive real shock to the economy. I would focus on nominal GDP. I’d focus on monetary policy driving nominal GDP.

Then the third way I’d simplify things is to simplify monetary policy itself. Go back to the pre-2008 system of a small Fed balance sheet. Essentially, the Fed controlled nominal GDP growth at that time just through one major tool, open market operations. They didn’t pay any interest on bank reserves. At that time, the monetary base was like 98% currency. You could think of basically the Fed creating 2% inflation in the long run by just pushing up the monetary base about 2% a year faster than the increase in the real demand for base money. That was almost all currency. Again, it was 98% currency and 2% bank deposits at the Fed.

That framework, I think, was very simple, easy to understand. Moving to this more complex Federal Reserve System, I understand there are some advantages to abundant reserves, but I think the added complexity has contributed to some mistakes in recent years. I think, overall, the earlier, simpler system was superior.

Nominal GDP Targeting

Beckworth: We’ll come back to this operating system question and Fed balance sheets when we get to our discussion later today on Fed independence, because I think they play into each other really well. Going back to this key emphasis on nominal GDP as a way to really gauge what monetary policy is doing, whether it’s too much, it’s too little, it’s refreshing, Scott, to hear someone else say this besides me. I know George also would say this. We fought the good fight. We pushed hard for the framework review. I had no expectation the Fed was going to fully endorse nominal GDP targeting, but maybe use it as a crosscheck, maybe to help it think about ways to proceed going forward.

You mentioned in regards to nominal GDP that there was too little of it. Effectively, monetary policy was too tight in 2008. We see, I think, a hysteresis or a decline of potential real GDP because nominal GDP was too tight during that time. On the flip side, we saw 2021, 2022 nominal GDP surge. I want to try to land the plane in between those two. Insufficient in 2008; too much during the pandemic. The good thing, or the win, I think we can take is we saw that it is possible, though, to return nominal GDP to its trend path. We just overshot it.

I want to look at that. I think too many times it’s easy to be critical that we did overshoot, but at least we know we can return. That should be a data point that tells us macroeconomic policy can quickly return. Now, maybe because this was a real shock, a supply-side recession, maybe that’s part of the story, too. It was a V-shaped recovery. I take solace in seeing that, hey, nominal GDP quickly bounced back up the trend that went too far. I think macroeconomic policy can thread the needle of nominal GDP if their heart’s in it.

Sumner: That’s right. You’re right that the COVID situation was unusual. I think there’s a lot of data going back through history to support the notion that monetary policy really matters in that regard. If you look at the Great Recession, for instance, at the time, especially in the early part of the Great Recession, a lot of people thought it was going to be worse in the United States than in Europe because the subprime crisis had happened in the US. It seemed like we were the epicenter of the Global Financial Crisis. Now, as you know, it actually ended up being far worse in the eurozone. 

It’s hard to explain why the recovery in the United States was much more satisfactory than in Europe—even in the US, it was somewhat weak, but it was way more expansionary than in Europe—other than by pointing to the fact that Federal Reserve monetary policy was more expansionary than ECB policy. I think when you compare the US to Europe during the Great Recession, you really see how monetary policy does make a difference in a side-by-side comparison. I would point to that as an example, but there’s many other examples you can point to of how monetary policy matters, even though it may appear to be relatively ineffective because of, say, the zero lower bound on interest rates. There’s lots of other things the Federal Reserve can do. When it is more aggressive, you get a quicker recovery.

I might add that the recovery that took place in the United States, especially beginning in 2013, occurred against the backdrop of quite a bit of fiscal austerity. If we look at the recent bounce back, that was both fiscal and monetary pushing in the same direction. It’s hard to know exactly how effective monetary policy was. In 2013, fiscal policy got a lot tighter. They cut the budget deficit almost in half from a little over a trillion to about a half a trillion between 2012 and 2013 using calendar data.

That fiscal austerity, which was done to reduce the long-run budget deficit problem, put a lot of headwinds in front of the Fed. In late 2012, the Fed did a lot of aggressive moves, like a more aggressive QE, more aggressive forward guidance, and these moves by the Fed actually caused the recovery to speed up a little in 2013 compared to 2012. We also see, not just in comparison between the US and Europe, but examples of monetary policy working against strong headwinds from fiscal contraction. We can again see where it makes a difference. That’s another example I’d point to.

Beckworth: Yes. I recall from that period there were a number of commentators who were warning us about massive job loss. “Watch out, folks. We’re probably headed to a recession.” Some pretty well-known folks. I’ll mention Paul Krugman. I’m pretty sure he was one of them who was telling us, this is the craziest thing we’re doing, all this fiscal austerity, but lo and behold, it did not manifest in terms of a recession. I think that’s another great example for doing the right counterfactual. There is mixed evidence like QE1, QE2, QE3. How much of a punch does it pack for the real economy?

I think in the case of this experience, what you’re saying is, well, do the right counterfactual. Had the Fed not done the QE during this time, maybe there would have been a recession. Relative to recession, it did pack quite a punch. As opposed to QE4, the pandemic QE, we definitely see a much stronger correlation between QE and inflation and robust real GDP growth.

Quickly going back to the eurozone crisis, Scott, I think you would say this, and correct me if I’m wrong, you would make the case, and I have made this case as well in the past, that the eurozone crisis was triggered by the ECB being excessively tight. They raised interest rates twice in 2011. Now, some folks would say, “Well, it was the debt crisis. It wasn’t the ECB.” I think your reply would be that, “Well, if the ECB hadn’t tightened, in fact, if it had been more accommodative, nominal income would have made those real debt burdens more light, more supportive, more bearable.” Is that how you would view that period?

Sumner: Yes, people are reversing cause and effect with the debt crisis and the tightening. The tightening occurred before the debt crisis mostly. I would add that if you look at mainstream macroeconomics as it existed in the first part of the 21st century, there was a presumption that when you’re not at the zero lower bound, monetary policy is driving nominal aggregates, like inflation and nominal GDP.

Now, there is debate about the zero lower bound, and that’s an important debate, but here’s something I’d like to emphasize. Europe was not at the zero lower bound all through the early years of the Great Recession. It did not cut interest rates to zero as the Fed did. It kept them well above zero. As you pointed out, in 2011, it raised interest rates twice. Soon after, Europe went into a double-dip recession. There’s really no evidence that it was anything other than explicit moves by the European Central Bank to tighten monetary policy.

Now, I emphasize that because I’ve argued in some cases you can have passive tightening where it doesn’t look like the central bank is doing anything in terms of concrete actions. I blame the Fed for doing nothing to interest rates during the middle part of 2008. That was a passive tightening of policy because the natural or equilibrium rate was falling at that time, but it’s harder to see in those cases. Most people look at interest rates, and if you try to convince them that policy is tightening, even though the central bank doesn’t appear to be doing anything, you have a hard sell.

With the European Central Bank, it was taking explicit steps. In late 2008, it actually raised interest rates. I should say maybe mid-year, around July 2008. It wasn’t just refusing to cut them; it raised rates and did so twice in 2011. Both cases, the European economy turned down sharply right after these concrete steps. How you can look at that picture and not see monetary policy as driving a fall in nominal GDP and even creating deflation, it’s hard for me to imagine.

As for why the ECB raised rates in 2011, as I recall, there was some commodity price shocks around that time. China was recovering strongly. Commodity demand globally was increasing, so a lot of commodity prices were rising. You’re getting a little bit of inflationary pressure from the supply side, which you and I both know central banks have to look through supply side inflation. The ECB responded to that in making the same mistake they made in 2008 and 2011. They tightened monetary policy and ended up making the recession much worse, and the recovery in Europe ended up being far slower than in the United States.

Beckworth: You raise a great point. It’s not just where the targeted policy interest rate is currently; it’s where it is relative to the equilibrium or natural or real r-star measure. If the economy is collapsing, then that equilibrium rate is also collapsing. Even if you have a low nominal rate, but it’s not going down—that’s the whole zero lower bound problem. The economy needs a negative real rate to clear the market, and yet it’s maybe stuck at zero.

You mentioned 2008. We’ll provide a link to this. Ramesh Ponnuru and I had a piece in the New York Times, where we made your argument effectively. The response we got is, are you guys crazy? David and Ramesh, what have you guys been smoking? Didn’t you see the Fed cut rates dramatically? It cut all the way down from 5% to 2% in 2008. How can you say the Fed was too tight? It’s the very point you raised. It was tight relative to where the equilibrium rate was.

I would also add, it wasn’t just sitting at 2%. The FOMC was talking up rate hikes. Markets were expecting rate hikes during that second half of the year because of this inflation scare. It’s even worse than just 2%, as I think about it. It’s relative to where markets think the rate’s going to go versus the equilibrium rate. There was a growing gap in 2008 that really put weight and pain on the economy.

Sumner: Yes. I would add a couple of points to that. First of all, people who argue that interest rates are the right indicator of monetary policy have a big problem because when you have highly inflationary monetary policies, like in the ’70s, interest rates are going to be very high. It makes no sense to argue that that was a tight money policy because the inflation was causing the high interest rates. The second point I would make is, the housing decline really reduced the equilibrium interest rate a lot in 2008. Yes, it looked to some people like easy money, but they didn’t cut rates fast enough.

Finally, when the Fed got around to doing QE, a lot of people who focus on interest rates would suddenly start suggesting that QE showed that monetary policy was expansionary. Well, that’s like moving the goalpost, because if QE is the proper indicator of monetary policy, QE is really nothing more than changing the size of the monetary base. Now they’re arguing that printing money is the right indicator of whether policy is expansionary.

If you use that criteria, in late 2007 and early 2008, the Fed dramatically slowed the rate of increase in the monetary base. If printing money is your correct indicator of concrete steps of monetary policy, by that criterion, monetary policy became contractionary in late 2007 and 2008 and could be argued to have triggered the recession.

Now, my own view is that neither interest rates nor the monetary base are reliable indicators of the stance of monetary policy. You really need to look at outcomes. You look at nominal GDP growth or market forecasts of nominal GDP growth going forward as the best indicator of whether money is too easy or too tight. These other indicators—interest rates, money supply growth, and so on—they’re useful at some times in history and at other times they’re completely unreliable for very good reasons.

Beckworth: I would just add to that, someone like a New Keynesian would say, “Oh, but Scott, if we just look at the gap between the neutral rate and then the federal funds rate,” and then your response would be, and my response would be, “Well, we don’t always directly observe that neutral rate, and it’s changing in real time.” Theoretically, I think that’s fair, but in practice, you’ve got to look at something you can actually observe. 

Same thing with money. Yes, money might be useful, but you’ve got to look at it relative to real money demand, and we don’t observe real money demand either. There’s all these unknowns, so let’s keep it simple. Let’s focus on the nominal outcome.

Sumner: The original idea, I believe, for the natural rate of interest was the interest rate that would keep the macro economy in equilibrium. I think Wicksell talked about a stable price level as an indicator of the economy being at the natural rate. If that’s your criteria for the natural rate of interest, why not just cut out the middleman and ignore the rate of interest entirely and just focus on the variable you’re trying to stabilize? In Wicksell’s case, it was a stable price level. I would argue for stable nominal GDP growth at about 4% a year.

Whatever your goal is, that should be the indicator of whether monetary policy is too expansionary or too contractionary. If the goal variable is measured with a lag, and it certainly is for nominal GDP, then you’d want to look at market forecasts, I think, of where nominal GDP growth is likely to be going forward. I think those market forecasts are probably the best indicator of whether the stance of policy is too easy or too tight. Just looking at interest rates doesn’t answer that question at all because the natural rate moves around, as you say, in a way that’s unobservable in real time.

Quantitative Easing

Beckworth: Let me go to QE4 or the pandemic/COVID QE program. Let’s put aside the March through, say, June version of that. It was truly market functioning QE. The Fed stepped in to keep markets from collapsing. If you go beyond that, let’s say late 2020, ’21, and early ’22, that QE really did seem to pack a punch in terms of inflation, robust real GDP growth. It was much more apparent and obvious in QE1 through QE3. We just talked about, well, maybe we need to do the right counterfactual for those earlier QEs. In general, other than maybe doing the right counterfactual, why do you think QE4 was so much more effective, at least in appearance, it looked like it, than, say, those earlier versions of QE?

Sumner: Good question. Let me start off with an analogy for interest rates because this is a counterintuitive area. I think most listeners understand interest rates better than QE. With interest rates, you have the same problem. Sometimes you’d look at a low interest rate policy and see very expansionary results, so much so that people think of low interest rates intuitively as an easy money policy. You might see rapid growth. Other times you might see low interest rate policy lasting for decades, like in Japan, and very little growth. Again, the question is why does it seem to work in some cases and not in others?

That’s because it’s, I think, wrong to think about interest rates itself as a policy. It’s more like an implication of deeper policies. In some cases, with a low interest rate policy, you’re pushing rates below the equilibrium or natural rate, and that expands the economy. In other cases, the natural equilibrium rate has fallen very sharply, and you’re cutting interest rates to zero as a defensive move, trying to make up for the fact that the equilibrium rate has fallen well below the policy rate. In that case, it may have relatively little effect because even at zero rates, the policy may not be effectively very expansionary if the natural interest rate has gone negative.

The same thing occurs with QE. Some types of QE, most famously hyperinflationary countries, are highly expansionary because you’re putting money into an economy where people don’t want to hold that much more money. That money generates hyperinflation because people try to get rid of these excess cash balances. This is the Venezuelas, the Zimbabwes, and so on, Germany during the hyperinflation. In other cases, when you pump a lot of money into the economy, you’re accommodating a demand for liquidity because interest rates have fallen to zero, and money is a safe asset that can be held at zero rates. There’s no opportunity cost of holding in that situation. There may be risk associated with other assets. The public has a very high demand to hold liquidity, and QE is meeting that demand. 

Now, it gets even worse if the Fed decides to pay interest on bank reserves. That creates even more demand for liquidity. Banks are likely to hold or demand more reserves at a positive interest rate if they’re being paid interest on their reserves. In that kind of environment, you can do a lot of QE and have relatively little expansionary effect.

Whether it be changes in interest rates or QE, in both cases, the effect depends on the macroeconomic background in which the policy occurs. Is it actually aggressively pushing you away from an existing equilibrium, or are you accommodating preexisting changes in equilibrium interest rate or preexisting changes in the demand for liquidity? Those accommodative moves tend to have very little macroeconomic impact.

Beckworth: QE1 to QE3 was more of an accommodative QE.

Sumner: Exactly. The same would be true of the moves in Japan. Now, I’m not going to say that it had no impact at all. If you look closely at the financial market data, the Fed’s decision to do those QEs probably moved the economy somewhat forward relative to a decision not to do a QE, sort of other things equal basis. It didn’t have the dramatic effects we would expect because it was mostly accommodating a demand for liquidity that occurred at the time.

Now, in the Great Depression, when the Fed didn’t really do any QE until about 1932, like three years into the Depression, and there was more demand for liquidity, the Fed’s failure to do QE caused a much bigger drop in the economy than would have occurred if they had done QE. I don’t mean to suggest that it was a mistake for the Fed to do it or it didn’t have any effect at all. I’m just trying to explain why the results would look disappointing to a lot of people because it was mostly accommodating an increased demand for liquidity.

Beckworth: Then it comes down to a policy choice. They could have done more in QE1 to QE3 and done more than just accommodate the increased demand for safe assets for liquidity. Had they done that, then we would have seen higher inflation and more robust real growth.

Sumner: Yes, but I also believe there’s a much more effective technique than just doing more QE or forward guidance on interest rates. I think the most effective technique is a different policy regime, something like level targeting, where you commit to makeup from previous undershoots. I think with some form of level targeting or makeup policy, you create more bullish expectations. When that occurs, any given amount of QE or any given interest rate policy becomes more expansionary in that environment. These more fundamental decisions about the type of monetary regime you have and where you’re setting your targets, I think, are actually the most effective way to achieve the goal.

For many people, this is deeply counterintuitive because our instincts are to look at concrete steps, like, what do you do to interest rates? What do you do to the monetary base with QE? When I suggest, well, the way to get more rapid nominal GDP growth is to commit to more rapid nominal GDP growth, that just seems like something that’s too good to be true, like a Peter Pan story. If I believe I can fly, I can fly. Any commitment of that sort has to have real credibility. In other words, the central bank has to be committed to actually do the steps in the long run that are required to achieve that goal.

I don’t mean to suggest that words alone can be successful. There have to be credible actions by the central bank over time. I do believe that commitment, if it is credible, makes things so much easier that I would go so far as to argue that if we had had nominal GDP level targeting back in 2008, at, say, a 5% trend line, which was the trend back in those days, they probably never would have had to cut interest rates to zero. It would have been like Australia, where they could have kept rates above zero all through that recessionary period because the more bullish expectations about future nominal GDP would have supported a much higher natural interest rate than what actually happened.

Much of the fall in the natural interest rate was due to the market very correctly perceiving that nominal GDP growth was going to plunge and stay low in the 2010s, which is exactly what happened. When that perception developed around October 2008, that’s when the bottom fell out of the financial markets, and that’s when the banking crisis got much worse. Expectations are crucial, not just for stabilizing the macro economy, but for reducing the severity of financial crises.

Beckworth: I love your Peter Pan analogy. That’s wonderful. You believe it will happen.

Sumner: I didn’t invent that. I think someone like Paul Krugman or someone came up with that, but it’s a good story.

Beckworth: It’s a good point, though, and it speaks to the importance of credibility. My question would be, how do you get that credibility in the first place? Do you have to first set up a new regime, and say, “Hey, we’re going to do nominal GDP-level targeting, get used to it. Let’s try it for a few years, and hopefully have it in place before you hit 2008”? Or, could we have switched midstream 2008?

That’s the comment that Ben Bernanke made, I believe, in his book when they were discussing different monetary policy targets in 2011, that nominal GDP targeting did come up, but people were like, “Man, why would you want to change?” We’re implicitly doing inflation targeting. We’re going to something different. It’s hard. It’s different. If you’re given the magic wand to set things anew, how would you set this up in an ideal world so that it would have that credibility so that QE and interest rate policy would be much easier?

Sumner: That’s a good question. There’s a lot that can be said on this. First of all, one criticism is, the public doesn’t understand nominal GDP targeting. That’s probably true, but it’s equally true the public doesn’t understand inflation targeting.

Beckworth: Good point.

Sumner: When Ben Bernanke suggested we needed to raise inflation in the early 2010s, a lot of people were scratching their head because, although the public maybe sees a headline about the Fed’s 2% inflation target, the public typically views that as a commitment to keep inflation no higher than 2%, not to raise it from 1% to 2% if it falls below. The average voter has no idea why the Fed was trying to raise inflation in 2010. What I would do is, I would adopt nominal GDP level targeting but continue to have a, say, 2% inflation target as the headline that’s presented to the public.

In other words, I would do something like this. I’d say, “Look, we would like to keep inflation at about 2% on average. That’s been our goal. That will continue to be our goal. We have other mandates as well, like high employment. To meet our various mandates, we think the best way to do that is to generate growth in nominal GDP at about 4% a year. We believe that will allow for about 2% inflation but will also help us better achieve our high employment goals at the same time than if we just focused only on 2% inflation.”

In other words, 2% inflation would still be the headline ultimate goal, but nominal GDP targeting would be the regime that is being given to the financial markets to better understand what the Fed is doing on a month-to-month and year-to-year basis. They’d be telling the financial markets, “Look, we’re trying to achieve 2% inflation over time, but we also want to stabilize the business cycle. We think the best way to do that is nominal GDP level targeting at about a 4% growth rate.”

That’s the actual way that policy gets implemented. The public would probably tune out the nominal GDP stuff because most voters don’t even know what nominal GDP is or how it differs from real GDP, and would continue to focus on the Fed’s 2% inflation target.

That would be one way of doing it. There’s probably others as well. Politically, I think it could be handled, especially because there is a dual mandate. If the Fed had been given a single mandate for 2% inflation or price stability, let’s say, it would be hard for the Fed to move to nominal GDP targeting because it would seem to violate that mandate. Given they have the dual mandate, I think nominal GDP targeting is a natural way of balancing those two objectives of high employment and stable prices.

Beckworth: One more question about QE, and we’ll move on. Let’s go back to the pandemic QE programs. Again, we had really high inflation, historically high inflation, the highest it’s been since the early ’80s. There’s been many explanations for it. One is, oh, it was largely supply chain disruptions. There’s choke points in the global supply chains. The Fed did as good as it could. It’s all about supply side. Then others would point the finger at the Fed. The Fed just did too much. Others point the finger at the helicopter drops in the federal government, the stimulus checks, the generous unemployment insurance. Where do you land in terms of what caused the inflation that hit almost 9% in 2021, ’22?

Sumner: I have an unusual perspective on this. I don’t think many people are where I am, but I’ll try to explain it this way. I think the fiscal policy was excessive. I agree with people like Larry Summers that criticized some of the stimulus checks and so on. Obviously, we needed to run larger budget deficits in the COVID period for various reasons. There was high unemployment. The federal government did need to step up, and that would have resulted in a substantially larger deficit. We went well beyond what was required for meeting the needs of people that were unemployed. That would be the first point. Fiscal stimulus was excessive. A lot of people believe that. 

Where I’m more of an outlier, I would say, is that I don’t believe the excessive fiscal stimulus caused the high inflation. That’s obviously a minority view. Let me explain it this way. The monetary authorities can always offset fiscal actions that are either too expansionary, as in 2021, or too contractionary, or just say contractionary, as in 2013. If they fail to do so, in my view, we should put the responsibility for the inappropriate movement in nominal GDP on the Federal Reserve. It’s their job to produce roughly 4% growth in nominal GDP each year.

Even if fiscal policy is irresponsible, they can adjust monetary policy to offset that and produce the appropriate amount of nominal GDP growth. Fed officials will actually sometimes say this when prompted by reporters about what’s going on with fiscal policy. I’ve heard Fed officials say, “Well, our job is to take fiscal policy as a given and do what we think is appropriate to hit our macroeconomic objectives.” They almost have a mandate, in a sense, to offset the effects of fiscal policy that are inappropriate. Not to sabotage fiscal policy, but to offset effects that would move the Fed away from its policy goal of 2% inflation and high employment.

If they fail to do so, then I think it’s appropriate to put 100% blame for the miss on nominal GDP growth, and there was a huge overshoot, on central banks. Causation is tricky. You can always do counterfactuals of what if this person had done this, what if this person had done that, or this institution. To me, the most useful counterfactuals are the ones that are useful for public policy purposes. If you believe, as I do, that the central bank should be given the task of producing stable growth in nominal GDP, then the appropriate counterfactual would be what if the central bank had behaved differently and set policy at a level that produced the appropriate level of nominal GDP growth?

They should have been aiming for nominal GDP to quickly return to the trend line, which it did in late 2021, and then stay on the trend line. Instead, they produced a policy that led to nominal GDP quickly returning to the trend line, but then dramatically overshooting the trend line because they didn’t push back against fiscal enough. Now, some people tell me, “Oh, it would have been very unpopular to do a tight money policy when the fiscal authorities were expansionary in 2021,” but I’m not even asking for a highly contractionary policy all through 2021. The interest rate was at zero. The Fed was doing QE all through that year.

In other words, if they had set interest rates at zero but not done QE, that alone would have been a considerably more contractionary policy, and interest rates still would have been at zero. For people that say it would have been too controversial for the Fed to do a tight money policy, I’m not suggesting they should have raised interest rates to 5% or something like that in 2021. At a minimum, they should not have been doing QE at a time when the economy was overheating. They could have certainly nudged interest rates a little bit above zero without a lot of political controversy, I think.

Although I should say that that was also an interesting period where President Biden was considering who he was going to appoint as the next Fed chair. You can speculate as to whether that long, drawn-out process made things more difficult for the Fed. I don’t know. That’s a political judgment. In a technical sense, what I would say is we would have wanted the Fed to be less expansionary, enough so that they at least offset the overshooting part of fiscal policy.

Fed Framework Review

Beckworth: Now, again, Scott, you’re a big champion of nominal GDP level targeting, so you’re a supporter of having the right framework or target. We just went through a framework review at the Federal Reserve. We went from the flexible average inflation targeting, which can be viewed as a version of a temporary price-level target. We went from that back to a flexible inflation target. Now, the new flexible inflation target has some language in it, the new consensus statement that it allows the Fed to be creative and to probably bring out balance sheet policy, even maybe bring out makeup policy, again, should they need it, but they’ve minimized those tools. They’re stressing that they’re back to FIT, flexible inflation targeting, because then a framework gets robust in many environments. What are your thoughts? We went from FIT to FAIT, back to FIT.

Sumner: Okay. Again, this one’s a little bit tricky. I think the announced policy in 2020, flexible average inflation targeting, was an excellent policy. As you know, that’s not the policy that was implemented. When the policy was announced—you can correct me if I’m wrong—didn’t Jim Bullard have a comment that it was like nominal GDP level targeting?

Beckworth: Yes.

Sumner: Okay. As you know, what was implemented was nothing like nominal GDP level targeting. Nominal GDP overshot, I think, by more than 10 percentage points, the previous trend line. Second, the Dallas Fed put out a statement to the public explaining flexible average inflation targeting and said the policy is symmetrical. You make up both inflation overshoots and undershoots. As you know, that’s not the policy that was implemented. The Fed later on suggested they were only going to make up for inflation undershoots, like in 2020, but not make up for inflation overshoots like 2021.

It ended up being asymmetrical, and in my view, that was a very bad idea. I have to admit that I was wrong. I thought it was a symmetrical policy, so I applauded it when it was announced. What was actually implemented was, I won’t say a complete disaster, but I think largely explains the high inflation of the period 2021, ’22, ’23, and so on. This is the difficulty. If you’ve announced a very good policy and implemented something else, and it works out poorly, you’re going to be under pressure to change the policy, even though the announced policy was actually a very good one.

It’s very hard politically to say our new policy is flexible average inflation targeting, but this time we really mean it. Politically, you can’t sell that. They had to go back to the FIT, the simple flexible inflation targeting. I understand that, but it’s very unfortunate that a very good policy was announced, and then because it was not implemented as announced, it became discredited in most people’s minds. We’re back to an inferior policy, the one that led to the very slow recovery from the 2008 recession.

Now, you said they might still do a little bit of makeup in the future. Maybe privately, they know what went wrong and have learned some lessons and just don’t feel they can credibly announce it without looking bad. I won’t even speculate on that, but we just have to hope that each time the Fed makes mistakes, they learn some lessons from that and at least avoid that specific mistake next time around. There’s always the tendency of the Fed to be like the general fighting the last war, and their mistakes are often in the opposite direction of the previous mistake in the previous business cycle, which is certainly what we saw in the early 2020s, vis-à-vis 2008 and ’09.

Fed Independence

Beckworth: Okay. Let’s transition to more current developments. In that regard, is the Fed’s independence. Does it have independence? Will it continue to have independence? Will certain personalities undermine it, but will also just the growing debt burden undermine the Fed’s ability to do what it wants to do? We talked about earlier the Fed can’t always offset fiscal policy, but is it possible we get to the place where it can’t just because it has to keep the government solvent? It has to step in like in World War II and do yield curve control, things like that? What are your thoughts?

Sumner: Yes, I think it’s possible, but I would say that at this point in time, there’s no technical reason why the Fed should be constrained by fiscal policy. When the Fed makes mistakes, as they made around the early 2020s, I don’t believe it was driven by fiscal considerations. I believe the mistakes were driven by concern over unemployment and a desire to bring unemployment down quickly, which is a laudable goal, but they obviously overshot on that objective. The fiscal situation in the United States is, in a sense, unsustainable, unless there’s some sort of AI miracle, perhaps. Let’s say by conventional measures, it’s unsustainable and has been since the late 2010s.

It’s also something that one could imagine being fixed. It’s not like the US has a particularly high tax rate compared to most developed countries. I mentioned in 2013 they did a lot of fiscal austerity. They brought the deficit down in half in one year, from a little over a trillion to a little over half a trillion. There’s no technical reason why a combination of spending restraint and tax increases couldn’t push the deficit back down to 3% of GDP or some sustainable figure. If Congress doesn’t do that and Congress just decides to be irresponsible and accept that the US is going to become a banana republic, obviously, that could happen.

The thing that I would point out here is that even though fiscal austerity is somewhat unpopular with voters, high inflation is even more unpopular. The inflation of 2021/’22 was far more unpopular with voters than the fiscal austerity of 2013. Sometimes I see discussions of fiscal dominance that imply that, well, we’re going to be stuck eventually with high inflation because Congress won’t do the unpopular thing of fiscal restraint. In fact, the high inflation is the unpopular thing. There’s no magic way to monetize the debt without creating a lot of inflation.

Sometimes I see discussion of what fiscal dominance would look like. The discussion implies that it would look like a world of very low interest rates. You referenced the World War II period. What I would emphasize is that, actually, true fiscal dominance is a world of very high nominal interest rates because it would produce high inflation, and through the Fisher effect, you’d get high nominal interest rates, the sort of thing we saw from the late ’60s to the early ’80s. That’s the world we would have if we actually had fiscal dominance.

Are we there yet? I’m a little bit worried, but I would point to the 30-year tip spread, which is the financial market forecast of inflation. It’s about two and a quarter, and that’s for the CPI. That’s about 2% for the PCE inflation that’s being targeted. I would say at least in the tip spreads, there’s no current indication that people expect fiscal dominance over the next 30 years. Now, I know the tip spreads aren’t perfect, and there can be some biases in there, but I don’t think we should just throw up our hands at this point and say, “Well, there’s nothing monetary policy can do because fiscal policy is irresponsible.”

If the Fed continues to stick to its mandated targets, it’s going to put pressure on Congress to do something at some point about fiscal, and if Congress just refuses to act responsibly for year after year, yes, eventually in the long run, we’d end up in a situation some Latin American countries got into where you had to finance the debt by printing money. I don’t think we’re there yet, so I’m not willing at this point to just give up on monetary policy for that reason.

I do concede the current trends in recent years are worrisome in that regard. We can’t continue down this road forever. We do need to fix the fiscal situation at some point. Again, absent some AI miracle, which people increasingly talk about, but I’m reluctant to forecast either way on that.

Beckworth: I think it’s a fair point for those who maybe are more worried about fiscal dominance impairing the Fed that even if it were the case, you still don’t want the Fed, like you said, to throw its hands up and say, “I can’t do anything,” because that would add extra fear to the market, and that would further unanchor inflation expectations. Even if it’s the case, that we’re close to the precipice of fiscal dominance, the Fed still needs to do its job, still needs to communicate confidence like it’s in control, because that itself is worth something.

I also wonder, and I was thinking about this as you were talking, I wonder if it’s as easy to do something like yield curve control today. Our MMT friends would say, “Hey, all you got to do is, like the 1940s, just peg the rates and impose price controls. That solves your problems.” Today, we have a globalized market. How much financial repression could the US government do? We also have crypto assets emerging.

I’m wondering if, in fact, maybe the reason Treasury markets are so confident is because they know the hands of the government are tied at some point. They really will have to rein in, make some tough choices, because they can’t do the kind of financial repression, fiscal dominance tactics they used in the past.

Sumner: Right. I think even if you look closely at the late ’40s—I apologize, I’m relying on memory here—I believe that after World War II, there was initially expectation of deflation because in previous wars, there’d been a lot of deflation and depression after the war ended. That increases the demand for liquidity, and it helps the central bank keep rates low in that environment. Now, we actually know that the period after World War II turned out very differently from previous wars. We did not have deflation. We had a lot of inflation.

I also believe that by maybe around 1948, already some of the yield curve control was breaking down, I think, at the long end of the Treasury market. By ’51, it was apparent we were going to have high inflation if that program was not abandoned. I think it worked a little bit under some very special circumstances of an environment where the market wrongly expected a post-war depression of the sort we’d had after other major wars. The market didn’t understand that we were moving to a new world, a more fiat money world, where you didn’t necessarily have to bring prices back down after wartime like in previous experiences.

Once the market recognized we were in a new world and inflationary expectations started to build up, the Fed realized they had to abandon that policy or they would let inflation get out of control. In 1951, I believe, it was finally abandoned. I believe it was already breaking down around 1948, at least for the longer-term interest rates. I think I wouldn’t put too much weight on that as an approach that could be used going forward in this environment.

Beckworth: You’re saying what began to take place in 1948 would be even more so today because it’s a globalized economy. It would break down quickly. It would be tough to do. You can’t really bank on that hope. Either there’s going to be high inflation or the government’s going to have to get their act together. What you said earlier as well, the public will not tolerate the high inflation, so therefore the government will get its act together.

Sumner: Right. I’ve actually been surprised at how angry the public is about inflation. Before listeners jump all over me for being out of touch, let me point out I’m old enough to have lived through the inflation of the ’60s through the early ’80s. That inflation was dramatically higher than the recent one, and longer. It lasted for a much longer period. When it started to come down in the ’80s, say from 1982 to the end of the ’80s, inflation was brought down to about 4% a year. That was viewed as a big success after the double-digit inflation of the late ’70s and early ’80s. Today, 4% inflation would be viewed very negatively by the public.

You might recall Ronald Reagan was reelected with 49 states in the 1984 election, partly because he got inflation down to 4%. I have to admit, I’m a little out of touch with voters. Yes, people are very upset about inflation rates that, from my perspective, don’t seem that bad. I’m also pleased to find out the public is more worried about inflation than I thought they would be, because I believe that that worry, even if it’s, in some sense, maybe slightly excessive or the public maybe doesn’t understand that inflation affects both prices and wages, so it’s not quite as bad as they might think, that hostility the public has toward inflation helps prevent fiscal dominance.

It puts pressure on Congress not to allow the situation to become so bad that we become like a so-called banana republic and just inflating our debts away. For what it’s worth, I’m actually pleased that people are reacting that strongly to inflation, even if, in some sense, they might be slightly overreacting, not understanding it also affects their wages.

Beckworth: Well, I will remind our listeners that the fact that Paul Volcker was able to do what he did in the early ’80s was because people did dislike inflation so much. One of the early podcast shows I had was with the late Robert Samuelson, the journalist from The Washington Post. He just recently passed away. He had a great book that came out in 2008. It was terrible timing. He had a great book come out on the great inflation in the 1970s. Something I learned from him and I’ve tracked ever since is the Gallup poll on what is the top concern or the top worry.

It’s striking. If you go back into the mid to late ’70s, inflation was the number one worry. All the other things going on in the ’70s you could think of, Watergate, coming out of Vietnam, all these things were second, third, fourth-order importance. Inflation was number one. It seemed like a very different world because when I talked to him, it was probably 2012, ’13, ’14, sometime in then, and we’re worried about disinflation, even deflation. We’re worried about the other problem. Yet, it was a strange world to process. People were that worried that they tolerated Paul Volcker’s treatments.

Then we come to ’21, ’22, lo and behold, we awaken that beast. That inflation-hating beast in people. I’ve collected data on the Gallup question. If you look at it even today, inflation concerns are still higher than unemployment concerns. If you look also at Google searches for inflation, you can use that trend tool, you can look up different terms, you see a clear structural break in ’21. It’s level, and then it spikes. It comes back down, but it’s now at a permanently higher level, which means people are now more actively searching for inflation than they did before 2020.

People are now more price-sensitive. I think the saying, “Once bitten and twice shy,” would resonate well with the inflation psyche in America right now. I think policymakers do need to be mindful. That is something that probably will affect President Trump. It may affect the midterm elections this year. I agree with you. I’m surprised. If you go back in history, maybe we shouldn’t be surprised.

Sumner: I’m really going to date myself here. If you go back to maybe the ’60s and ’70s, there was a sense in the air that inflation was just a part of the modern world. As time goes by, things get more expensive. I remember when I started teaching in the early ’80s, I would show my students graphs of the US price level from the Revolutionary War to today. For the first 150 years of the country, there was no inflation on average. There were ups and downs, periods of inflation, periods of deflation. The price level in the early 1930s was about the same as at the beginning of the country, 150 years earlier. People thought in terms of high and low prices.

Then we started to have continual inflation year after year. Then in the ’60s and ’70s, it was accelerating inflation. People began to think about it as just a part of the modern world. I think they also thought about it, in some sense, as being maybe a side effect of prosperity. At that time, there was a lot more focus on unions pushing wages up. Unions were a lot stronger in the 1960s than today. There were big strikes to get higher wages. There was a perception that inflation was also affecting wages in a big way, and that it was perhaps part of prosperity.

In the Great Depression, we’d had falling prices. Older people in the ’60s could remember in the 1930s how bad things were with falling prices. Today, we don’t have any historical memory of that sort. We have a generation or even two generations that experienced long periods of roughly 2% a year inflation and never experienced any deflation or didn’t experience something like the Great Depression. Their baseline assumption was that prices should be fairly stable, just creeping up a little bit over time.

Then we had the big spike in inflation in 2021 and ’22. There was a sense, well, prices needed to come back down. Whereas by the 1970s, nobody expected prices to be coming back down because they’d been rising for quite a bit for many, many years. It was just seen as being a part of the modern world. Like, we have prosperity pretty much compared to the Great Depression, but one of the side effects of prosperity is there’s so much money here that it’s creating inflation and workers are demanding much higher wages, et cetera.

There was a different mindset about what the process was. Certainly, inflation was unpopular. I think what Paul Volcker did of just slowing the rate of increase to 4%, to many people, that felt very good after the double-digit inflation of previous years. Today, 4% inflation would not be considered very satisfactory by the public. We’re at about 3%, supposedly trying to get it down to 2%, but the public is still very frustrated with prices, and they’re still frustrated over the fact that the previous price spike didn’t unwind, as it would have with level targeting, by the way. In fact, I would go further. If we had had nominal GDP level targeting, the price spike itself would have been much smaller in 2022 under that regime.

Beckworth: It would have minimized the overshoot to begin with.

Sumner: Right. You would have had markets pushing interest rates dramatically higher in 2021 as they saw the overshoot beginning to occur and were anticipating already the Fed’s makeup policy. That market anticipation of future tighter monetary policy would have restrained aggregate demand in the short run. What a lot of people don’t understand about level targeting is it’s not done because you have this obsessive fixation on getting back to a trend line on a graph. The whole point of level targeting is to make the overshoots and undershoots themselves smaller by having stabilizing expectations about the future path of policy. That’s really the justification for level targeting, not just getting back to the trend line because you like trend lines.

Beckworth: You can adjust that targeted trend path, too, right? If real GDP changes, your nominal GDP level target can gradually update to reality. Again, just to stress this point, people really dislike inflation. If we talk about the German hyperinflation in the early ’20s, that still resonates with a lot of Germans today. It resonates with some of the leaders of the ECB. In fact, its 2008 mistakes that it made, and in 2011. Yet they had the Great Depression. A lot of people say the Great Depression played an important role in Hitler coming to power. They don’t remember that as much as they do the inflation.

I’ve talked to people on the show, and they say inflation is fundamentally different. Everyone gets affected by inflation. It’s unfair, especially if it’s rapid and unexpected. It really does something to the human psyche. If it’s a sudden spike in inflation, no one can really escape it. I do think politicians, words of wisdom to our leaders today, need to be very mindful and careful with this.

Sumner: Yes. We’ve had a pretty long period of, I guess, a healthy job market other than COVID, obviously. I think the public views COVID as an exception and a special case. If you take out the COVID period of high unemployment, which was fairly brief, much briefer actually than a lot of people, including myself, expected, we’ve had fairly low unemployment for quite a while. I think people do take the job market for granted now and focus more on inflation. 

The German case is interesting. As you suggest, it was actually the Great Depression that put the Nazis in power. The hyperinflation occurred well before then. I’ve talked to German people, and they’ve told me that in schools it’s taught the other way, that it was the hyperinflation was the big problem, that led to the rise of the Nazis. Even five years after the hyperinflation ended, the Nazi Party was still very small and unsuccessful in electoral politics. Then the Great Depression hit, and immediately the vote share of the Nazi Party started shooting up dramatically. There’s no question in my mind it was the Great Depression. 

When we actually do have a very bad job market, then the public does care a lot about employment. As I say, other than the special case of COVID, which people tend to discount, we’ve had a pretty good job market for a number of years now. I can understand why people would focus on inflation. I also, as an aside, think some of the poll numbers are a little bit misleading, that the electorate has become very polarized. You can see this in splits between the two parties. That polarization, I think, does make people more negative about things. If you poll people on “How are you doing personally in your financial situation?” or “How is your state doing economically?” the poll numbers look better than if you ask them about the national economy.

I’m not suggesting people aren’t unhappy about the inflation. I think they are. I also think the more polarized political environment perhaps ratchets up some of the negativity that you see in how the economy is doing. As you probably know, there’s a little bit of a disconnect between the economics profession that sees 4.5% unemployment and 3% inflation as being a reasonably good economy. The public is reporting it’s one of the worst economies in their lifetime, right?

Beckworth: Sure.

Sumner: A lot of the survey results I’ve seen have the economy at levels that are maybe the worst of the 2008 Financial Crisis, which I find just shocking that people would view the current economy as being that bad. I don’t know what to make of that. I’m not sure my comment on political polarization is explaining it either. There’s certainly something odd going on in terms of disconnect between the normal macro data we look at and the very negative view of the public on the situation.

Beckworth: Okay. With that, our time is up. Our guest today has been Scott Sumner. Scott, thank you so much for coming back on the podcast.

Sumner: Thank you.

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.