Scott Sumner on How Central Banks Should Respond to the Coronavirus Threat

Scott Sumner is the Ralph G. Hawtrey Chair of Monetary Policy at the Mercatus Center at George Mason University, and a returning guest to Macro Musings. Scott joins the show today to talk about the recent market turmoil caused by the COVID-19 coronavirus and its implications for monetary policy.  David and Scott also discuss how the Fed should respond to a possible pandemic, why monetary policy is preferable to fiscal policy during a crisis, and how to approach the central bank credibility problem.

David Beckworth: Hey Macro Musings listeners, as you know last week financial markets were rattled by coronavirus fears. The stock market plunged over 10 percent. The benchmark 10 year Treasury yield had an all-time low near 1.1 percent. And oil prices and expected inflation plummeted. This is a special episode on this financial turmoil that was recorded last week in the midst of the chaos. So some of the data discuss is already outdated. The policy discussion and advice however, remains timely. So check out the show.

Beckworth: Welcome to Macro Musings, the podcast series where each week we pull back the curtain and take a closer look at the important macroeconomic issues of the past, present and future. I'm your host Beckworth of the Mercatus Center, we are glad you've decided to join us. Our guest today is Scott Sumner. Scott is my colleague and the Ralph G. Hawtrey Chair of Monetary Policy at the Mercatus Center. Scott joins us today to discuss the recent market turmoil caused by the coronavirus and its implications for monetary policy. Scott, welcome back to the show.

Scott Sumner: Thank you for inviting me David.

Beckworth: Well, today is Tuesday, the 25th. This has been a big day, markets are down six percent over the past few days, market futures are looking at dim as well, the 10 year Treasury, Scott has reached 1.3 percent, the lowest ever on record. And the question is, what do we make of it? What should the Fed be doing now, if anything? And I'm glad that I have you Scott, of all people on the show with this, kind of walk us through this. You were a voice in the wilderness crying out during 2008 when others were saying everything's fine. The Fed's doing a decent job. But in 2008, the Fed kept rates steady between April and October 2008. And you were telling them they needed to do more. And you were right.

Beckworth: And so it's good to have you back on the show and to guide us through what the Fed should be doing. And as I look around, I see two views on what the Fed should be doing. The first view, is the Fed can do anything. The Fed is a cure all, it can solve any problem, supply shocks. I don't think that's quite right. The Fed can't solve a breakdown in global supply chains. It can't cure people from a virus. But the other error is in the other direction, says the Fed can do nothing. It's a supply shock, the Fed should not do anything at all, that it would be just caving in to Wall Street. So between these two extremes of the Fed can do everything and the Fed can do nothing, I think we're missing some nuance and I'd like to hear your take on that, Scott. What should the Fed be doing if anything during this time?

How Should the Fed Respond to Coronavirus Fears?

Sumner: Okay. To explain this, I need to clear up basically two misconceptions. And the first one is the confusion about the nature of the shock that's hitting the economy. It is true that the coronavirus is basically a supply shock. However, what the markets are really worried about right now is a negative demand shock. So that requires some explanation. And to really understand that, we have to sort of figure out what the role of monetary policy is in all of this. So basically, the Fed is targeting interest rates and it doesn't adjust its interest rate target very often. On the other hand, there's sort of an equilibrium interest rate that moves around with market conditions. And this negative supply shock is probably reducing expected growth in the global economy. And that's putting downward pressure on the equilibrium interest rate.

Sumner: So here's what's actually going on. If the Fed does not reduce interest rates, along with the fall in the equilibrium rate, then monetary policy will get unintentionally tighter. And it's actually going to reduce aggregate demand. So to go back to your original question, you posited two extremes, the view that there's nothing the Fed can do, and there's the view that monetary policy can solve the problem of this coronavirus shock. And those extremes are both wrong.

So to go back to your original question, you posited two extremes, the view that there's nothing the Fed can do, and there's the view that monetary policy can solve the problem of this coronavirus shock. And those extremes are both wrong.

Sumner: It's not really correct to say the Fed should just stand pat and not adjust interest rates because it's a supply side problem. And it's not correct to say that monetary policy can solve supply side problems. In my view, the correct view is, it's initially a supply side problem, which could then trigger a change in aggregate demand and monetary policy needs to adjust to prevent that secondary effect on aggregate demand. That's the one thing monetary policy can do, prevent this supply shock from spilling over and reducing aggregate demand in the broader economy.

Beckworth: Alright, so Scott, what you're saying then is that this shock is a disruption to the potential GDP of the world economy. It's a production shock, and that's going to lower the equilibrium rate or what the Fed officials call R-star. And if the Fed does nothing, then it's passively tightening monetary policy, is that a fair summary?

Sumner: Right. So this is kind of similar to the trade war last year that had similar effects on the financial markets. When there's a disruption to manufacturing supply chains, that tends to reduce business investment, puts downward pressure on demand for credit. That will tend to reduce equilibrium interest rates. In addition, with the coronavirus, there's also a lot of uncertainty in the global economy. And when there's uncertainty, there's sort of a rush for safe assets, people buy treasury bonds, that puts downward pressure on interest rates. So you have this downward pressure on global interest rates.

Sumner: Now while this is occurring, if the Fed holds constant its policy rate, it targets the, say fed funds rate at a little over 1.5percent. While the equilibrium rates are falling, then essentially the Fed will be making monetary policy tighter. Many people make the mistake of simply looking at interest rates and thinking that they're seeing monetary policy. They think high interest rates is tight money, low interest rates is easy money. But it's more complicated than that. What really matters is where the Fed sets the interest rate relative to the equilibrium rate. And you can think of the equilibrium rate as sort of the rate that is consistent with a stable economic growth, consistent with the Fed's targets for inflation and employment, and so on. And if the Fed sets its policy rate too high or too low, we tend to go off track.

Many people make the mistake of simply looking at interest rates and thinking that they're seeing monetary policy. They think high interest rates is tight money, low interest rates is easy money. But it's more complicated than that. What really matters is where the Fed sets the interest rate relative to the equilibrium rate.

Sumner: And this is what confuses people, you wouldn't think you'd need a monetary response to a supply shock that has nothing to do at first glance with money. I mean, it's a physical disruption to the global economy. How can you fix that with monetary policy? But I'm not suggesting the Fed would fix the problem, but rather sort of adapt to it. Think of this metaphor, suppose your policy is to drive your car at 60 miles an hour down the highway. If that's your policy, then you have to put more or less pressure on the accelerator pedal, depending on whether you're going uphill or downhill.

Sumner: And what I'm saying is, if the Fed actually wants to maintain a stable monetary policy, they may have to move their policy interest rate up and down with market conditions to keep the effective stance of monetary policy stable. So again, it's not trying to solve the supply side problem, it's trying to prevent it from spilling over and also impacting aggregate demand.

Beckworth: Okay, so by doing nothing, the Fed is still doing something?

Sumner: Exactly. And in fact, there's many ways of doing nothing with monetary policy. And this is one of the things people don't understand. If you hold the money supply constant, that's one way of doing nothing, but in that case, interest rates will move around. If you hold interest rates constant, that's another way of doing nothing, well, then the money supply will move around. So what we really want is a sort of a measure of monetary policy that is useful, not just arbitrary, like interest rates or money supply or exchange rates. And there's several methods that are useful, market monetarists like me look at nominal GDP growth, the sort of our measure of how it's doing. New Keynesians look at the difference between the policy interest rate and the equilibrium interest rate. And since that's the more common way to look at monetary policy, I usually kind of explain this paradox using interest rates, the concept of needing to move your policy rate up and down with market conditions to keep policy effectively stable.

Beckworth: Alright, Scott. So those are great points. And I think they're useful because again, as we mentioned earlier, you have these two extreme views that say, the Fed should do nothing. If it does something, it's bailing out Wall Street. But on the other hand, the other extreme view is, well, the Fed can do everything, which is not the case. And you've taken a middle ground there, that the Fed's trying to avoid any of these spillover effects that can occur if the Fed fails to respond appropriately. And again, by doing nothing, it might be doing something. And again, just to summarize your first point, I believe was, because the supply shock will lower potential real growth, the neutral interest rate will decline. So by not adjusting the stance of policy to that, it would be effectively tightening.

Beckworth: And then the other one is just the spillover effects through uncertainty, expectations. So it might be the case that markets overreact to the coronavirus and the implications that has. And we've seen the markets, again close to six percent, come tanking down. Greg Ip had an article on that last point Scott, that came out today. And title of his article *Fear of Coronavirus, Rather Than Virus Itself, Hits Economies.* And his point is-

Sumner: That's right.

Beckworth: ... the fear itself is what's generating these concerns. And the way I like to think about and you alluded to this is that effectively, is people get nervous, they get antsy. They overreact out of proportion, and what effectively happens is the velocity of money slows down or the amount of money spending is going to decline or people are going to move into more liquid assets, maybe not buy the big purchases. So it's not really tied to fundamentals. It's an overreaction. And that's where the Fed could step in and offset that. And if they fail to, they're passively tightening monetary policy.

Sumner: Yeah. Let me just interject that. What you've described is another way of saying the same effect I was talking about with equilibrium and target interest rates. So if velocity slows down because people become more cautious, then to maintain the same effective monetary policy, you'd have to increase the money supply to offset the effect on declining velocity. And that would keep nominal spending growing at a stable rate, which is what you really need for a healthy labor market.

Sumner: So there's different languages you can use to describe this process of monetary policy, there's the monetarist language involving money supply demand velocity, there's the Keynesian language, describing equilibrium interest rates compared to the Fed's target rate and so on. But they all reach the same conclusion, that monetary policy has to be nimble and adjust its instruments to maintain an effectively stable monetary policy when conditions change in the economy.

But they all reach the same conclusion, that monetary policy has to be nimble and adjust its instruments to maintain an effectively stable monetary policy when conditions change in the economy.

Beckworth: Yeah, I think we both agree that something is happening now along those lines. But what makes this hard, is that we don't directly observe the neutral real interest rate or R-star.

Sumner: Exactly.

Beckworth: We don't directly observe real money demand or velocity, at least in real time. But we do have nominal GDP, which you mentioned earlier, is kind of a catch-all, looks to kind of reflect those different factors. And it would be really nice if we had a forecast a nominal GDP that would reflect, are people expecting a decline in spending overall? So do we have any insights, any type of real type nominal GDP forecasts that are revealing anything to us?

What is the Role of Nominal GDP Forecasts?

Sumner: Not as much as we'd like, but we do have a small prediction market that Mercatus helped set up, which is at Hypermind, a French organization. And there is a nominal GDP forecast there for the next one year. And that's currently about 2.9 percent, which is kind of a little bit on the low side in terms of what the Fed would like. But let me add another point, even though we don't have perfect forecasts, and we don't directly observe equilibrium interest rate, when there are big changes in the financial markets, it's pretty easy to see the general direction of where the equilibrium interest rate is going. For example, interest rates in the futures market have recently dropped a lot. So that's sort of the market's prediction that the Fed will later cut interest rates.

Sumner: Now, by itself, that doesn't prove anything because interest rates might fall because policies are becoming highly expansionary. But when you combine the fall in interest rates with other indicators like inflation forecasts in the market, are low below the Fed's two percent. Stocks, of course, are falling sharply in recent days, when you look at the whole picture of many different financial variables and put it together like a jigsaw puzzle, there's a lot of indirect evidence that the equilibrium interest rate in recent days has fallen significantly. And that's an indication that the Fed probably needs to reduce the target interest rate that they set on short term rates in order to prevent monetary policy from becoming tighter.

Beckworth: Yeah. And one proposal you have that would make this easier to see in real time, and again, you can look at all these asset prices as you've just outlined, but one proposal that would make it even easier is if the Fed did adopt something like a nominal GDP futures market, is that right?

Sumner: That's right. And what I envision is a market that is heavily subsidized so that it would be highly liquid, much more trading than the little Hypermind prediction market. I mean, that one's nice, but it's not really a true market.

Beckworth: Right.

Sumner: The people can only win prizes and so on. But a real market that had a lot of money behind it to provide liquidity, and I think that would be very useful. I've even proposed that the Fed actually stand willing to take a long or short position on nominal GDP futures as sort of guardrails on monetary policy that would tend to ensure that monetary policy stayed in a position where the market expected nominal GDP growth to be within say, three or five percent, if those were the points at which they took positions in the futures market.

Sumner: So I won't get into the details of that now, but there are ways of using market forecasts to guide monetary policy. And if they don't do that, the next best is for the Fed to look at a wide range of market indicators, and at least heavily consider what those indicators are telling them about policy, perhaps drifting off course. It's not a perfect guide by any means, but it does provide useful information, especially when there's a danger of policy drifting far off course, as it did in 2008.

Beckworth: Yeah. One of the things that this crisis has reminded me about is that it's hard to have a truly clean supply shock to the economy. And that is, it's hard to have a run where you just see the supply side be affected and nothing else. So you don't see these spillover effects. And that's because of finance and money, they're so closely linked to the real economy. And this became really apparent last August when we had all these discussions about the dominance of the dollar at Jackson Hole meetings. Mark Carney gave a talk about replacing the dollar with a synthetic currency, but there's been a lot of discussion about the weight that the dollar brings to the global economy. Hélène Rey has talked about the global financial cycle or the global dollar cycle. The dollar is so dominant, that whenever there is a crisis, the demand for it goes up, increases the value of the dollar and all the dollar denominated debt around the world becomes more of a real burden on economies.

Beckworth: And because the global economy is so closely linked through the dollar, it's hard to have a large global supply shock without having these spillover effects that feed into aggregate demand, as you have explained, and it also speaks to the challenge of management. It's an international monetary order that's been great for global growth. Globalization would not have occurred, probably without the dollar as a global medium of exchange. But in times of crisis, it becomes this kind of burden. And the only institution that can really handle the adjustments needed for a really strong dollar is the Federal Reserve.

The Effects of Global Dollar Dominance

Sumner: That actually raises an important point, there's actually two important spillover effects that are involved here. And one of them is something that you've researched in your paper on the US as a monetary superpower, the fact that the dollar plays a very large role in the global economy. So changes in US monetary policy have ripple effects on many other countries. So that's one sense in which there's a spillover effect. The second is one we've already been talking about to some extent, which is that supply shocks can spill over and impact aggregate demand. And I'd like to, on that one, just to drive home how important that is.

Sumner: If you go back to 2006 to 2008, we had a long decline in housing construction where housing construction fell in half over 27 months. And there was almost no effect on the unemployment rate. It just edged up a little bit between early 2006 and early 2008. And then it spilled over into demand. What was originally just a problem in the housing industry started to affect aggregate demand as a whole, because the Fed didn't cut rates fast enough in 2008. And then the unemployment rate skyrocketed. Similarly, I believe the effect of the coronavirus on aggregate demand this time around would be much greater than the supply side effects.

Sumner: So, your listeners may have studied economics and been taught that negative supply shocks are inflationary, but the markets are acting as if this shock will be deflationary. In other words, the markets believe that the spillover effects on aggregate demand going down will be worse than the initial effects of the supply side. And then if the Fed does not respond properly, if they don't cut rates fast enough, monetary policy gets unintentionally tighter. One side effect of tighter money is the dollar appreciates in the foreign exchange market. This also happened in the second half of 2008 during the financial crisis in the US. The dollar appreciated strongly.

Sumner: And for countries where their currency is linked to the dollar, or they borrowed a lot of money in dollars, this can be a very big burden. So it can impart spillover effects on a lot of other countries in the global economy. And you'll notice that if the Fed adopts an unexpectedly expansionary monetary policy, it not only helps the US stock market, but it also helps many foreign markets because of these interrelationships between the dollar and the broader global economy.

Beckworth: Yeah, and this raises the question of what the Fed should do if this becomes a global financial crisis. And I feel like I'm getting a bit ahead of myself, we're not there yet. I mean, the markets are correcting pretty strongly right now, and by Monday when the show comes out, maybe things will have cooled off a bit, but let's say they haven't, and this does become a full blown financial crisis. What should the Fed do?

Beckworth: Well, in 2008, as you brought up that period, just now. In 2008, the Fed extended a number of currency swap lines to a number of central banks across the world because the Fed was the only institution that could extend dollars to the parts of the world that needed them. And I suspect for the reasons you've just mentioned, the Fed would have to do the same again. And I know this is something that the Fed doesn't like to do, and it's uncomfortable, maybe politically charged, but it's something they may have to do, is extend currency swap lines, and it may need to do more of them. There are currency swap lines at major central banks around the world except in China and a few other places. And China's the big gaping hole when it comes to where should there be a currency swap line where their currently is not one.

Tactics and Strategy of Policy Response

Sumner: Right. So the way I think about the policy response is in terms of tactics and broader strategy.

Beckworth: Okay.

Sumner: Most people focus on tactics because those are the easiest things to see. These include cutting interest rates, quantitative easing, the currency swap lines that you mentioned. These are all ways to address the immediate crisis, the lack of liquidity, the need for stimulus and so on. But I would argue that especially in a major crisis, broader strategic changes in monetary policy are actually much more powerful. And maybe we can get to those later. But I think the Fed has to really think about this on two levels. What are their tactics? That is the concrete steps they take today. And what sort of broader monetary policy strategy would help stabilize the economy, restore confidence, create more bullish expectations about long term growth?

But I would argue that especially in a major crisis, broader strategic changes in monetary policy are actually much more powerful.

Beckworth: So you're saying the Fed should bring forward its review, end it-

Sumner: Right.

Beckworth: Right? And say, "Hey, we're going to do level targeting of some kind." And this would-

Sumner: Exactly.

Beckworth: ... help restore some confidence in the markets. Okay. So you've heard that people, let the Fed do it now. And I would add to that, I mean, if we're going to talk about tactics and targets. I mean, the currency swap lines, maybe do the standing repo facility and make it more accessible in times of crisis to other counterparties. But you're saying, at a minimum, bring forward and announce a level target ASAP.

The Case for Level Targeting

Sumner: That's right. I mean, you and I both favor level targeting-

Beckworth: Yep.

Sumner: ... in nominal GDP. But we also both know that the Fed isn't willing to go there at this moment. So the best we can do is some kind of level targeting of prices. The Fed has a two percent inflation target for PCE inflation. And that actually probably would be almost enough if it were credible. And one way to make it credible is to switch from the growth rate targeting which means you just target each year and forget about the past. You let bygones be bygones. And as you know, over the last decade, they've missed their target on the low side almost every year. I think 2018 maybe was the exception, but basically, they've been under shooting on inflation. And under level targeting, it's much harder for them to do that because it requires the Fed to make up past mistakes.

Sumner: So for instance, one thing that would be helpful in this crisis period would be if the Fed would announce that, if the coronavirus problem gets worse, they will create an emergency level target for the next five years. In other words, they'll commit to an average inflation rate of two percent over the next five years. And if it falls to one percent, for the next two years, because of a slump in the economy, they'll make that up with higher inflation when they're able to do so. That commitment would tend to raise the equilibrium interest rate on five year treasury bonds. It would create more bullish expectations in the five year Treasury market.

Sumner: Now some people ask me, they say, "Well, if the Fed has missed its inflation target, why would anyone have any confidence on level targeting?" And the beauty of level targeting is that it's actually much more effective than inflation targeting, even if they miss by the same percentage. So for instance, let's say they've been missing on the low side by a half a percent on inflation targeting. If over the next five years they came in with a price level that was a half a percent below their target, that would only be a 10th of a percent per year, because you'd take that half a percent and spread it out over five years.

Sumner: So inflation in that case would actually end up very close to two percent. So what level targeting does and people I think often miss this, is it forces the Fed to be much more serious about hitting its targets. It can't just wave away mistakes and say, "Oh, we overestimated inflation, we'll try to do better the next time." They actually have to make up for past mistakes. And the Fed most certainly did not miss its inflation target because it was unable to hit it, as you and I know they raised rates nine times-

Beckworth: Right.

Sumner: ... between 2015 and 2018. And to say the Fed was unable to hit their target would be like someone bringing their new car back to the auto dealer and saying, "I'm unable to get it up to 60 miles an hour, no matter how many times I tap on the brake, it won't go up to 60 miles an hour." You'd probably tell the person, "Try tapping on the accelerator not the brake." So the Fed raised interest rates nine times, each interest rate increase was expressly designed to hold inflation down, because their Phillips curve models falsely led them to believe that inflation was about to rise above two percent because of low unemployment. So they made mistakes. It wasn't that they were unable to hit their inflation target. And with level targeting, they have to make up those mistakes and it ensures the long run inflation rate is much closer to what you want than the current system.

And to say the Fed was unable to hit their target would be like someone bringing their new car back to the auto dealer and saying, "I'm unable to get it up to 60 miles an hour, no matter how many times I tap on the brake, it won't go up to 60 miles an hour." You'd probably tell the person, "Try tapping on the accelerator not the brake."

Beckworth: So you're saying a level target has built into it some credibility already. So despite-

Sumner: Exactly.

Beckworth: ... by design, you don't have to worry about having sufficient credibility, because that is the concern right now, that's one of the concerns of average inflation targeting, is you're just tweaking what the Fed's already doing, now why do we think they'll be able to do anything different with this, to tweak to its existing framework. But you're saying a pure level targeting may have more credibility built in. Your second point is that the Fed simply didn't try hard enough.

Beckworth: And a good counterfactual would be to make a bunch of clones of Neel Kashkari, sit them around the table, at the FOMC meetings and see how the world would have been different with them. And I think that's a fair point and one that suggests at least inflation would have been a little bit higher. Now, I do think it's also fair to point out the Fed was facing some strong headwinds, the demand for safe assets and other forces that were at work. But yeah, they could have done things differently. And we don't know how that would have turned out. Both of us think it would have turned out very differently.

Beckworth: And this leads me to another question I have because some people do say there's lack of credibility, interest rates are converging down. In fact, we mentioned earlier the coronavirus, has lowered the 10 year Treasury yield, as of today to 1.3 percent. There are some estimates out there that could go even lower. But some folks are saying the Fed's out of ammunition. It simply can't deliver. And you need more fiscal policy-monetary policy cooperation.

Beckworth: So for example, Greg Ip, who we mentioned earlier, had another article, and this article in the Wall Street Journal was titled, *The Era of Fed Power Is Over. Prepare for a More Perilous Road Ahead.* And the point of his article is, is that central banks simply can't pack the punch they once did. And you need to rely on fiscal policy and unless fiscal policy gets its act together, unless there's more automatic stabilizers, things like that, we really can't probably expect too much from fiscal policy either. So it's a very dire picture he paints. What is your response to him?

Should Fiscal Policy Play a More Robust Role?

Sumner: Yeah. That's a popular view. But I think it's a dangerous misconception. And here's where I think that view comes from. So you have a lot of central banks that have had excessively contractionary monetary policy. They've made mistakes in recent years, even recent decades. And they've ended up with inflation and growth lower than their target. What people miss is, that low inflation and low growth puts downward pressure on the market interest rate. As interest rates fall, people wrongly think that these central banks have really easy monetary policy.

Sumner: So they think this way, they think, "Well, if the Fed's already cutting rates and doing QE, and we're not getting much results, think how much we'd have to do in order to hit the target." But that's actually looking at the picture backwards. These low interest rates are really a reflection of previous tight money. And really the QE is as well, because they do QE when interest rates fall close to zero.

Sumner: So what we really have to do is look at the picture in a different way, we have to ask ourselves two questions. One is, what kind of target do we want for inflation or nominal GDP growth? The higher the target, the higher the interest rate, and the less QE you'll have to do, because people won't want to hold as much cash if inflation is higher. If you want to have a very low inflation rate, like Switzerland, or Japan, then you'll have to do a lot of QE because people will hoard a lot of cash.

Sumner: So there's kind of a tradeoff there, where society has to think about which thing it cares about more, the size of the central bank balance sheet, or the inflation rate. And the other thing we've got to think about is, are we going to tell central banks to do whatever it takes to hit the target? So in a technical sense, central banks can always create any amount of inflation, up to the hyperinflation you've seen in Zimbabwe or Venezuela. They never run out of an ability to print money. The problem central banks run into is essentially political.

So there's kind of a tradeoff there, where society has to think about which thing it cares about more, the size of the central bank balance sheet, or the inflation rate.

Sumner: So there may be political constraints on what central banks are sort of allowed to do. There might be constraints on, for instance, how many assets they can buy, that is how much QE they can do. But if you allow the central bank unlimited ability to create money, they can definitely hit their target. So then, we really face this choice. Would we rather give central banks the tools and the instruction to do whatever it takes to hit the inflation target? Or would we rather give that over to fiscal policy?

Sumner: And I think for a wide variety of reasons, monetary policy is more effective, more likely to be successful than fiscal policy. It's also way less costly because monetary stimulus does not add to our national debt, it does not put a big tax burden on future generations. So for all sorts of reasons, I favor monetary over fiscal stimulus. But I do understand that the intellectual current right now is running in the other direction. I think people will be very disappointed if they rely on fiscal policy to help us in the next recession, just as it was ineffective in the 2009 recession.

Would we rather give central banks the tools and the instruction to do whatever it takes to hit the inflation target? Or would we rather give that over to fiscal policy? And I think for a wide variety of reasons, monetary policy is more effective, more likely to be successful than fiscal policy. It's also way less costly because monetary stimulus does not add to our national debt, it does not put a big tax burden on future generations. So for all sorts of reasons, I favor monetary over fiscal stimulus.

Beckworth: So based on what you're saying, in some ways, you're agreeing with Greg Ip in that a charitable reading of what he said would be, given the legal and political constraints central banks face, they aren't going to be very effective. Well, you're saying, "Well, change them." Right? "Give them-

Sumner: Yeah, yeah. And so there's sort of some nuance here because the question is, what do you want to assume in this kind of exercise? So many of the people that are skeptical about monetary policy at zero interest rates, like Paul Krugman and others, I think they understand that in a technical sense, the central bank could do any amount of inflation if you took all the shackles off. But they're making a political argument that central banks are conservative, they're constrained, they can only do so much in the real world. So that's why they call for fiscal stimulus. But my response would be, I think it's actually more politically feasible for the Fed to go to Congress and get additional tools right now than it would be for Congress, Democrats and Republicans, to agree on an effective fiscal stimulus.

Sumner: So I'm actually talking using the language of practical politics here. And I still believe that monetary policy is more feasible. In addition, if the Fed were to do the level targeting that you and I were just talking about, you don't need nearly as many tools, you don't need to do nearly as much QE under level targeting as under the current growth rate targeting. So there's other ways that the Fed can address this problem other than just by asking for an unlimited ability to buy as many assets as are needed to hit their target.

I think it's actually more politically feasible for the Fed to go to Congress and get additional tools right now than it would be for Congress, Democrats and Republicans, to agree on an effective fiscal stimulus. So I'm actually talking using the language of practical politics here. And I still believe that monetary policy is more feasible.

Beckworth: No, I agree with that. If you're going to do counter cyclical policy by the government, I think it makes a lot of sense to keep it at the Federal Reserve. I also understand the other side's argument, that if the Fed were to get more of these tools that looks more and more like fiscal policy, because maybe, the Fed's encroaching on ground that the fiscal policy in the past would have done. But your point is, let the Fed do it. It's going to be more efficient, more able. And if they do level targeting, they may not need a bunch of additional tools.

Sumner: Yeah. It's way less costly than fiscal stimulus. Because if the Fed buys a lot of assets, first of all, usually it will make a profit. They certainly made a lot of profit on the assets they bought during the Great Recession. But even if they make a modest loss, that loss is going to be a much smaller burden on future taxpayers than fiscal stimulus would be. So I also think this would be an ideal time for the Fed to go to Congress and ask for additional tools, at least as an emergency provision. Maybe Congress could say, "In an emergency, when interest rates fall to zero, you can buy a much wider range of assets."

Sumner: Now, why do I say this is an ideal time? If you go back five or 10 years, quite a few Republicans in Congress would have been skeptical of giving the Fed the power to have a more expansionary monetary policy. But now that the republicans are in charge of the government, they're more sympathetic to giving the Fed enough tools to keep the economy expanding and to avoid recession. And the Democrats have historically favored a more expansionary monetary policy to prevent high unemployment when necessary, and continue to have that general view of things.

Sumner: So I believe right now, you might actually be able to get a bipartisan consensus in Congress for a bill that would do something like give the Fed a provision where if interest rates fell to zero and they couldn't be cut any further, they could widen the range of assets they could buy, wide enough so that they would have clearly enough assets to hit a two percent inflation target.

So I believe right now, you might actually be able to get a bipartisan consensus in Congress for a bill that would do something like give the Fed a provision where if interest rates fell to zero and they couldn't be cut any further, they could widen the range of assets they could buy, wide enough so that they would have clearly enough assets to hit a two percent inflation target.

Buying Assets at the Zero Lower Bound

Beckworth: So what type of assets do you have in mind?

Sumner: So if the Fed were to buy assets, they already have quite a few that they're allowed to buy, including Treasury securities and mortgage backed securities. And the amount outstanding of those securities is something on the order of 30 trillion. I don't know the exact number, but it's a very large amount. To put that in context, I believe they purchased about 4 trillion during all the QE programs combined. So all those QE programs were very small relative to what they're already allowed to purchase.

Sumner: Now, what I'm saying is in addition, in an emergency, they could be allowed to purchase other assets if necessary. Corporate bonds, stocks, and the way I envision that, they would buy index funds so that they would not favor particular companies. But if you allowed the Fed to buy a very wide range of assets, then there's no doubt that if you purchased enough assets, you would create inflation and be able to hit the two percent inflation target. In my view, they wouldn't actually need to buy things like stocks, because there are so many treasuries and mortgage backed securities out there. And all they would really need to do is promise to do whatever it takes to hit their level target, promise to make up previous mistakes. If they do that, I think that they can do that using conventional assets.

Sumner: But giving them that ability, in an emergency, even if they don't use it, the markets would see that and it would create more credibility. I believe you refer to that as like the Chuck Norris effect. When Chuck Norris walks into a room, he's so strong, he intimidates people to flee the room. If people know the Fed has that much power, then when they start on a stimulus, it starts to create inflationary expectations. And that paradoxically means they don't have to do as much as otherwise, if their policy had no credibility.

Beckworth: Yeah, I-

Sumner: It's the countries that have no credibility, like Japan where they have to do much more QE.

Beckworth: Yeah, I think the key there is the level target. If you have a credible level target, then the purchases will be much smaller. And so let me play devil's advocate, and push back on your point about expanding the number of assets the Fed could buy, right? So you mentioned Japan, they have a balance sheet that's just over 100 percent of GDP.

Sumner: Right.

Beckworth: The ECB is not far behind and Gauti Eggertsson and a co-author of a paper where they respond to the point you just made, because Gauti Eggertsson once had Ben Bernanke tell him while he was a grad student, I think at Princeton, Ben Bernanke, he goes, "Look, if it weren't the case, that the central bank could create inflation and the central bank could buy up the entire planet Earth," right?

Sumner: Right.

Beckworth: Buy every asset. And that can't be true at some point, it'd have to be the case that inflation and aggregate demand would kick in. So what he did in his papers, he went and tried to estimate if you have an inflation targeting central bank, you haven't changed your targeting. And he comes up with a number, of about 400 percent asset to GDP ratio in a central bank balance sheet. And his concern is-

Sumner: Yeah, I think is wrong, but I'll explain. Go ahead.

Beckworth: Well, his concern is, by the time you get to that threshold, it's going to be so large, that you will have reached political constraints that won't allow you to buy that many assets or you've reached asset constraints. There aren't that many assets, if all those banks are doing this. And so, how do you reply to that?

Sumner: Right. So I think the mistake there is, as I said earlier, you look at all they've done, and it doesn't seem to have much effect. Imagine how much more they would have to do. But if they actually had a credible policy, they would have to do much less than they've already done. You see, the actual policy in Japan has been very reactive. It's been making up for past mistakes, accommodating a very high demand for base money because interest rates are so low. And so the public and banks, especially banks are just hoarding reserves. But in countries where the growth rate of nominal GDP is higher, like Australia, the demand for base money is actually relatively small as a share of GDP. I think only about four or five percent in Australia, and it used to be low in the United States before the Great Recession.

Sumner: So what you really need to do is set the target for either inflation or nominal GDP high enough so that people don't want to hold a lot of base money at that growth rate. And if you do so, and then you have a credible whatever it takes policy, then you actually don't have to do very much. If the markets see you're serious, then they will back off and inflation expectations will rise. And just take, to use Bernanke's example or to follow up on that. I mean, one of the points I think Bernanke was making was, what if Japan bought up the whole world, all the financial assets? Now, let's say that fails. Well, even if it fails, Japan would become a fabulously rich country.

Beckworth: Right, right.

Sumner: Right? Would own the whole world. So obviously, that's too good to be true. We know that's not going to be the equilibrium that comes out of that attempt. The world isn't going to sell all its assets to the Bank of Japan in exchange for essentially zero interest rate Yen currency notes, that's just not going to happen. So if work backwards from the final equilibrium, and once the markets understood, the Bank of Japan was serious, they wouldn't want to sell all those assets. And those Yen as they pour out into the financial markets would rapidly lose value. So that's one way of thinking about it.

Sumner: And another way of thinking about it is, what if the Bank of Japan set a high enough inflation target, so that people... The interest rates would clearly be well above zero, and you'd be better off holding safe government securities, earning interest over zero interest currency. Well, in that case, again, there wouldn't be much demand for base money. And really, throughout most of history, up until 2008, most developed countries had safe assets that earned more than zero percent interest. And in that world, people didn't want to hold much currency and banks didn't want to hold much, in the way of bank reserves. And I think that the best thing we could do now is go back to that world by setting a target for an inflation or nominal GDP growth that's high enough so that interest rates go above zero.

Sumner: And we don't have this hoarding of currency that Gauti Eggertsson is talking about in that example. But yeah, if you get deeply enough into a liquidity trap and interest rates get low enough, it's possible central bank balance sheets could get very, very large, larger than many of us imagined. And it will look like monetary policy is ineffective, but it will really be the long run response to previous policies that were too contractionary.

Beckworth: I think the key issue here is credibility. How do you create that credibility? You keep talking about a credible commitment to take whatever-

How Does a Central Bank Create Credibility?

Sumner: I think it's really simple. When policies aren't credible, it's because the markets correctly understand that the central bank won't do whatever it takes. So markets aren't dumb. They sense when central banks are serious and when they're not. So I believe that if the central bank actually sets out to do something, markets will quickly become convinced that it's serious, if it in fact is serious about doing whatever it takes. If it's not, if it's just trying to fool the markets, and it's going to back off, as the Bank of Japan did, then markets will sense that, it will lose credibility and the policy will end up being ineffective... Now, it wasn't completely ineffective in Japan. Things did improve after 2012, under Abenomics. But it didn't improve as much as they would have liked. Because there was a lack of credibility.

I think it's really simple. When policies aren't credible, it's because the markets correctly understand that the central bank won't do whatever it takes. So markets aren't dumb. They sense when central banks are serious and when they're not. So I believe that if the central bank actually sets out to do something, markets will quickly become convinced that it's serious, if it in fact is serious about doing whatever it takes.

Beckworth: Well, the thing is around the world, though, there doesn't seem to be much credibility in central banks to hit their inflation targets, right?

Sumner: Right.

Beckworth: If you look around the world, it's not just a US problem, it's a Japan problem, a Europe problem. An advanced economy problem in general, which seems to suggest there's some political or institutional commitment or constraint that prevents central banks from hitting their target. And so, again, the question is why would they be allowed, politically or whatever the constraint is, to run things up a little bit hotter, so they hit inflation or they overshoot their inflation target? Or they hit a level target. What would be-

Sumner: Yeah, good question.

Beckworth: ... different that allows them to do that?

Sumner: So let's look at this philosophically and think about our role. Your and my role in this process. We have a zeitgeist out there, the idea's swirling around the world and what's considered conventional wisdom in governance and central banking and fiscal policy. And you're right, that right now that conventional wisdom is too conservative for central banks to hit their target. The ECB has additional problems of having to coordinate with like 19 countries or whatever. And so there's a lot of political problems and as a result, right, there's no credibility that they will hit their inflation target. But our role as pundits making recommended changes to monetary policy is to try to change that zeitgeist, try to change the general view of what should be done to make it so that people start to see ideas as being more politically acceptable because there is no alternative.

Sumner: So eventually there will be a high level of frustration with the failure to consistently do what's necessary, and when that occurs, people will eventually look for ideas that were once considered improbable. Let me give you a perfect analogy. When I was in grad school, if you had recommended two percent inflation targeting, that would have been laughed at, people would have said, "Well, it's politically impossible. The central banks have to print money because unions are demanding higher wages," and anyone suggesting two percent inflation targeting would have been viewed as naive.

So eventually there will be a high level of frustration with the failure to consistently do what's necessary, and when that occurs, people will eventually look for ideas that were once considered improbable.

Sumner: And it's true that in the late '70s, it was politically impossible for a central bank to implement that. But people eventually became so fed up with inflation that the intellectual climate turned and by the early '90s, central banks did start to at least informally target inflation at around two percent. And then the US inflation's averaged about 1.9 percent since 1990. That didn't happen randomly or through luck, it didn't happen because of fiscal policy. It happened because of a change in attitude at the Federal Reserve.

Sumner: And I believe that eventually there will be a growing frustration with the failures to achieve adequate aggregate demand, people will look for ideas like level targeting or being more aggressive in terms of buying whatever it takes to hit your target. So when that occurs, now these policies will be more credible with the markets if markets see that the political system is backing up the central banks and that they're really serious about changing their ways and doing something that will really work.

Beckworth: Yeah, I wonder if that's going to be a generational thing, that it takes time, I mean-

Sumner: Yeah, it does.

Beckworth: ... part of what we're doing here, as you mentioned is to change thinking, this podcast the publication's, the work we do, and others like us. But I also worry that maybe we'll go too far. I mean, there's a reason populism has emerged now. I mean maybe some of that angst is an unrecognized need for aggregate demand, or the absence of adequate aggregate demand has led to some of the populism, maybe the rise of MMT and other views that really want to push fiscal policy hard, and that you swing too far the other direction. Any thoughts there?

Sumner: Yeah, that's possible. I'm a little bit optimistic. I do think we tend to swing too far one way or another and overreact. But I also think that each swing is a little bit less of an overreaction because we learn from past mistakes. So we learned a lot from the Great Depression. We came out of World War II and overreacted on the expansionary side in the '60s, '70s. We learned a lot from the great inflation where we had too expansionary monetary policy, then we had some pretty good policy for a while, and then we overreacted to the contractionary side, especially in the last decade. But we didn't do as badly in the Great Recession, as in the Great Depression, we still learned some lessons from that earlier one. In fact, we did much better than in the Depression.

Sumner: And I think that if we do overshoot, with excessively expansionary policies, it won't be back to 13 percent inflation, it might be more like three or four percent. And then we'll start to back off again. So I think we do learn things from previous mistakes. I wouldn't say it's necessarily a generational thing because the Fed is already reviewing its process and looks like it might adopt something that's a little bit like average inflation targeting. I mean, that's not really the best solution. But it's moving towards level targeting. And so it's a further improvement. And there's a few other things that I would point to. We're in the longest expansion in American history. And I think that's not an accident.

Sumner: It's because the Fed learned from some past mistakes and especially in 2019, under their old regime, they would not have cut interest rates three times because unemployment was falling below four percent. So their computer models were telling them, their Phillips curve models were saying, "We're going to overheat." But the Fed started looking at markets and saw the markets were actually warning them that policy was too tight. So they cut rates three times, and the trade war did not lead to a recession.

Sumner: And obviously, you and I have been pushing for more of a role of market forecasts in Fed policy and less reliance on Phillips curve models. You can argue that, I mean it wasn't necessarily our advice, but that the Fed is paying more attention to markets. And that the reason we're in the longest economic expansion in history is we didn't have the sort of normal recession we would have when a shock like the trade war hit, because they paid more attention to markets and were very nimble in reversing course from their interest rate increases. So I'm a little more optimistic that they are learning some things although short term I'm pessimistic about the coronavirus situation.

And that the reason we're in the longest economic expansion in history is we didn't have the sort of normal recession we would have when a shock like the trade war hit, because they paid more attention to markets and were very nimble in reversing course from their interest rate increases.

Beckworth: Right, right. Right, it's a very nice image there, the pendulum has swung far out both directions and it's beginning to swing down towards the middle and settle down in a place where we will be much better off with level targeting. So I like that vision very much. We are running low on time, Scott and I wondered if you want to give any parting words of wisdom to the central banks of the world as they deal with the coronavirus?

Sumner: Well, again, I think that it would be useful for them to have some sort of an average inflation targeting, at least as put it out there as a backup. The Fed could say, "If this gets worse than we currently expect, we're going to commit to make up to any short term inflation shortfalls." I think it would make sense for the Fed to cut its interest rate target because I believe the equilibrium interest rate has fallen. But I want to emphasize that cutting interest rates alone isn't necessarily enough to solve the problem, as we saw in previous recessions. And it's really more about the long run strategy that the idea of making up for previous shortfalls creates much more bullish expectations than you get from interest rate cuts alone, even though they probably should reduce interest rates to reflect the slowdown in the global economy, the greater uncertainty and so on.

Well, again, I think that it would be useful for them to have some sort of an average inflation targeting, at least as put it out there as a backup.

Sumner: In Europe and Japan, I think they need a much more dramatic rethink of what they're doing. And again, there I think Level targeting would be useful. But I think they're further off course, the US Fed's actually done a pretty good job despite undershooting its inflation target a little bit. It's produced outcomes in recent years that are overall pretty healthy, and so we're starting from a little bit of a better place than some of the other central banks. And we should take advantage of that, and not blow it by missing an opportunity to get ahead of this crisis.

Beckworth: So cut rates, but also complement it with a level target, which means in the case of the Federal Reserve, they should come out and announce their average inflation target now.

Sumner: Yeah. I mean, I suggest five years, it doesn't have to be that. But commit that the price level will be 10 percent higher five years from now. And if there's a shortfall in the near term, they're going to make it up. That would be very, very helpful. If they had had that in the Great Recession, we would have had a quicker recovery. They wouldn't have had to do those interest rate increases in 2015, '16, '17, '18, that period, because they would have been able to make up for the shortfall, they could have had a more expansionary policy and a quicker recovery.

Sumner: So, yeah, I think that would be very helpful. And there's some other things again, go to Congress asked for a bill that would say, even if they don't plan to do this, that in an emergency when rates fall to zero, they can buy a wider range of assets. That would also be helpful. It would reassure markets, I think.

Go to Congress asked for a bill that would say, even if they don't plan to do this, that in an emergency when rates fall to zero, they can buy a wider range of assets. That would also be helpful. It would reassure markets

Beckworth: Okay, with that our time is up. Our guest today has been Scott Sumner. Scott, thanks so much for coming back on the show.

Sumner: Thank you very much, David.

Beckworth: Macro Musings is produced by the Mercatus Center at George Mason University. If you haven't already, please subscribe via iTunes or your favorite podcast app. And while you're there, please consider rating us and leaving a review. This helps other thoughtful people like you find the podcast. Thanks for listening.

Photo by STR/Getty Images