Nov 22, 2021

Markus Brunnermeier on *The Resilient Society*

The COVID-19 crisis has taught us the best means of preparing for unforeseeable economic shocks in the future is to build policy with resilience in mind.
David Beckworth Senior Research Fellow , Markus Brunnermeier

Hosted by David Beckworth of the Mercatus Center, Macro Musings is a new podcast which pulls back the curtain on the important macroeconomic issues of the past, present, and future.

Markus Brunnermeier is a professor of economics and the director of the Bendheim Center for Finance at Princeton University. Markus is also a nonresident senior fellow at the Peterson Institute for International Economics. Markus joins Macro Musings to discuss his new book, titled “The Resilient Society,” as well as his work on safe assets and their implications for inflation. Specifically, David and Markus discuss the implications of the fiscal theory of the price level for inflation, the role of the Fed in stabilizing money markets, what is meant by “resilience” compared to “robustness” in economies, and much more.

Read the full episode transcript:

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

Beckworth: Markus, welcome to the show.

Brunnermeier: Thanks, David. It's a pleasure to be with you.

Beckworth: Well it's great to get you on. We have exchanged a number of emails, I've watched your program that you've had over the past year or so on the pandemic, a lot of great names you had on there, a lot of interesting discussion. And your work that we'll get to a little bit later, on the fiscal theory of the price level with a bubble has really captivated my imagination. And I'm doing some of my own work, inspired in part by those papers you've written. So we want to talk about those as well. But before we do that, Markus, tell us about your journey into macroeconomics and finance. How did you end up where you are?

Brunnermeier: So I should probably start. I grew up in Germany, and as a boy I was always puzzled why East Germany was doing so much worse than West Germany in economic terms. And when I discovered the informational role of prices through Hayek and Grossman's work, I was really fascinated. I thought I found it. So initially my research was very much focused on the informational role of prices, and that led me to bubbles and other things. And on my PhD thesis, I assumed in all that and drifted a little bit to finance. But I came from micro theory, but also I had an international chapter in my thesis. So I was very broad, but then my initial phase of my research was very much focused on this informational role of prices and bubbles and finance.

Brunnermeier: And then with the financial crisis, I had worked on liquidity problems before. And then I drifted more back into macroeconomic questions after the great financial crisis. And that's essentially what I ended up with, and my philosophy essentially is every 10 years you have to reinvent yourself and open a new chapter. And that's what I did. Luckily I was a little bit ahead of the financial crisis and had developed some concepts on liquidity issues and informational issues.

Beckworth: Well that is great career advice. I'll take that to heart. So when my 10 years are up at Mercatus, I'll have to start reinventing myself here. But yeah, I remember reading your work around the time of the great financial crisis. I remember first becoming aware of what you've done. And even after that, I think you had a paper on negative interest rates. At some point there's some kind of cost for the banking system and stuff. So I've been following your work and of course following your program, the one you talked about earlier, that you started during the pandemic. And then your papers recently on fiscal theory of the price level, so you've come full circle from micro to macro. Of course this is macro with micro foundations. So maybe you could say you're back where you started. So let's talk about those papers. And I think they're really fascinating, because for me at least, they shed some light on the low inflation we had coming into the pandemic so the decade or so before in the US as well as in Europe and other advanced economies, Japan. In fact, you guys really point to Japan in your paper, trying to make sense of the low inflation despite the large run up in public debt.

Beckworth: In fact I was looking today, Markus, in preparation for the show. The run up in public debt as a percentage of GDP actually grew more between 2008 and 2013 than it has over the past two years in the US, at least. And so if we were going to see this inflation, why didn't we see it then? What happened then? And so I want to start this conversation by looking at the fiscal theory of the price level. Maybe you can explain it to us, and then we can add on the bubble component that you've contributed in your work.

The Fiscal Theory of the Price Level

Brunnermeier: So I think what's important is the fiscal theory of the price level takes the fiscal decisions and the government debt level very seriously. It focuses very much on the store of value role of money and as a key equation, the FTPL equation. So it has, where monetarism focus very much on the medium for exchange role of money and also as a key equation, the quantity equation.

Brunnermeier: And the big puzzle for the fiscal theory of the price level, as you alluded to, is that if you look at Japan, Japan should have much higher inflation. So the key equation is like an asset pricing equation, so it's okay. The real value of government debt, so it's a nominal value over the price level, is the expected future parameter surpluses, what you get from holding these bonds, if you got to hold all the bonds or the government bonds. And you get much less than people expect. So the inflation of price level should be much higher. And you didn't see that at all. Japan was doing more and more, issuing more and more debt, and the parameter surpluses were mostly ... I mean out of 60 years, there were 50 years with parameters deficits. And nevertheless, the value of the debt, the real value of the debt was very, very high, is still very high. And this is the big puzzle. It's a big criticism towards the fiscal theory of the price level.

Brunnermeier: And we tried to resolve this criticism by saying, "Okay, in this equation there's something missing." And what's missing is a bubble trauma. That leads me back to my earlier interest in bubbles, of course. And once you put this bubble in, you can actually revive the FTPL equation and can, to some extent, address the Japan critique towards the FTPL.

The big puzzle for the fiscal theory of the price level, as you alluded to, is that if you look at Japan, Japan should have much higher inflation...And you didn't see that at all. Japan was doing more and more, issuing more and more debt...the real value of the debt was very, very high, is still very high. And this is the big puzzle...And we tried to resolve this criticism by saying, "Okay, in this equation there's something missing." And what's missing is a bubble trauma. That leads me back to my earlier interest in bubbles, of course. And once you put this bubble in, you can actually revive the FTPL equation and can, to some extent, address the Japan critique towards the FTPL.

Share this

Beckworth: Yeah, and I think we can also make some sense of what we're seeing now. I know there's some inflationary pressure. There's some debate, whether it's permanent or transitory. But even the amount we have seems small relative to the growth of public debt we've seen. And your work, I think, helps shed some light on that. And I want to come back to this point about bringing the bubble term in in a minute, because I think the bubble term kind of weds ideas from monetary theory with fiscal theory. But before we get to that, just so we get this basic fiscal theory of the price level equation you mentioned, it's an asset pricing equation. So it's looking at the net present value of future primary surpluses. So is this similar to a pricing equation for a stock?

Brunnermeier: Indeed, it's very, very similar. You can imagine perhaps you have investor who pulled all the government bonds. And what you get whenever the government makes a surplus, then the investor gets the money. When the government makes a deficit, the investor is giving money to the government. So the net flow, the net cashflow the investor gets is essentially the parameter surpluses of the government and then you discount it with the appropriate stochastic discount factor. And then you get the real value of the government bond. And the key essentially is nominal value divided by the price level, and that then determines the price level. And if the government goes more into debt or is expected in the future to go more into debt, and so it gives less to the investor, the real value of government bond has to go down. That means the price level has to go up, so inflation is going up.

Beckworth: Yeah, that's a very fascinating way of thinking about this. So the nominal government debt or liabilities are a residual claim to those future primary surpluses, just like someone holding stock has a residual claim to the dividends in typical stock price model. What's also fascinating when I think about people who've written on this and kind of used this analogy, I think of you. I think of John Cochrane. I think of Hanno Lustig. And there's others out there, but all of you, I think, come from a finance, a macro finance perspective as opposed to a traditional kind of just pure macro. So do you think that's a part of why this has emerged, is people who look at finance and macro thinking through these issues?

Brunnermeier: I think that's one, but of course there's also big contributions by Chris Sims, Woodford, Eric Leeper. So they were the initial-

Beckworth: Okay, fair enough.

Brunnermeier: ... people from the macro side pushing that. So it comes from both directions. It's actually always useful if finance and economics work together in pushing a common theme.

Beckworth: Yeah, all right. So let's go to this bubble term. And I really like this, because I think in some ways this brings us back to a ... I'm going to say this, may be wrong, a quantity theory of sorts. Because when you bring in the bubble term, what is the bubble term? Let's start there before we jump to my theory of what this means. What is the bubble term?

Brunnermeier: So typically I should say economists have a tendency to dislike bubbles. And then there are imposed transversality conditions and all these things to get rid of that. And I have a liking for bubbles, so I come from a different reaction, in a sense. And essentially when the interest rate, the real interest rate is smaller than the growth rate of the economy, the transversality condition doesn't need to hold. And when is this the case that the interest rate is very low? One reason is, for example, if there's a lot of uninsurable, idiosyncratic risk. So people want to save a lot for precautionary reasons. That pushes the interest rate down. And then it might push it down below the growth rate of the economy. And then suddenly a bubble term can emerge. And that's what you see in Japan. The interest rate is very low. And you also see it mostly in the US, that R is smaller than G. And Olivier Blanchard was pushing that. So you can this in the OLG model, where you have this. But you also can have it in incomplete markets model with uninsurable risk, where people save for precautionary reasons a lot. That pushes the interest rate down. And then suddenly the bubble term pops up.

Brunnermeier: And people haven't put this together, so that's what we did, essentially in this paper, say, "Okay, let's just look at the situation where there's a lot of idiosyncratic risk, uninsurable, idiosyncratic risk. People save for precautionary reasons. The real interest rate goes down below G. Suddenly the bubble term pops up. And then the asset pricing equation has this bubble term, which typically is ignored in the literature. And that's led to this view which we have that might actually rescue. To explain, you have to be able to rescue. Essentially now we can explain it also for Japan and other countries. And therefore we have a positive view on FTPL, in the sense I think this helps, actually, the fiscal theory of the price level.

Beckworth: Yeah, I completely agree with that. And I think one of the interesting research questions moving forward will be how to measure that bubble term and quantify it. And I know you guys, one of your papers, you take a stab at that. Debt is a safe asset, money in the bubble. But I think it'll be a great macro empirical research efforts going forward. What exactly, how big is it? Can we measure it? All those things, but also on that bubble, you motivated it by talking about ... There's the technical side, the transversality condition.

Beckworth: But also I think the more intuitive story for me, at least, is this precautionary saving motive, because there's idiosyncratic risks. And as you put in your papers, people want to hold onto government bonds, because they provide a transaction service. They have some kind of liquidity. And when I hear this story, Markus, what I'm hearing is a monetary story. These government securities are more than just some investment vehicle. They're actually a transaction asset. So this bubble term, I may be stretching it here, seems to be resurrecting some money demand and putting that back into the FTPL equation. Or am I being too monetarist here when I say this?

Brunnermeier: No, I think you are right. And so the way I would say, it's like a safe asset role for government bonds. So if you're a monetarist, you would say money is a very narrow concept. So it's like cash and amount deposits and all that. And then essentially you relax the double coincidence of wants problems. Typically we modeled this with a cash-in-advance constraint or money in the utility function or something like that. I have a broader perspective, where government bonds can also help us to save for precautionary reasons, because it's a safe asset. And so this gives you a formal convenience yield, which is way bigger with the safe asset aspect. So that's what we argue. If you look at this convenience yield coming from this safe asset perspective, it is what we call a service flow. So you get the cashflow from holding a government bond, and you get the service flow. And the service flow comes from the fact that this government bond is a safe asset.

Brunnermeier: And the way I define a safe asset is by two characteristics. I say a safe asset is like a good friend. It's around when you need it. Like a friend indeed is a friend when you are in need. So whenever you have a personal crisis, you have a healthcare bill or something, you can actually sell of this asset and then get some extra funds. So, for individual shocks, it's very good. And also for aggregate shocks, it gains in value in aggregate shocks. And that is self explanatory in this world. So you live in a world where you have all this uninsurable, idiosyncratic risk we all face. There's a lot of risks we face. And in times of crisis, actually the idiosyncratic risk is going up. So the value it provides, the service flow is actually becoming more important in times of crisis.

A safe asset is like a good friend. It's around when you need it...So whenever you have a personal crisis, you have a healthcare bill or something, you can actually sell of this asset and then get some extra funds. So, for individual shocks, it's very good. And also for aggregate shocks, it gains in value in aggregate shocks.

Share this

Brunnermeier: So you can see that, in times of crisis, this asset gains in value, and it gives it a negative beta. So if you think of a kappa and beta, it gives a negative beta. So it's like insurance instrument, and that's why the government has to pay such a low interest rate, because it issues an asset with a negative beta. And this negative beta comes from the fact that the bubbling term expands in times of crisis. And you can only do this if you have a sound, safe asset. If you issue a sound, safe asset, then you have an asset where the bubble term expands in times of crisis, so the asset becomes a negative beta asset. And that allows the government to issue, typically, the asset at a very low interest rate. And paradoxically, in times of crisis, they can issue an even lower interest rate, which is a great asset to issue. If you're a government who can issue this safe asset, it's fantastic.

Beckworth: It's great to be king and have a safe asset in your country, when you're in that situation. Now I want to tie this work to some other work that's, I think, really related. I've had Hanno Lustig on the show before, and I've talked to Arvind Krishnamurthy and then some other folks along with him. They've done a lot of work on the convenience yield on treasury securities. So these treasury securities in some ways act like money. They provide a transaction service, and that's ... I think you can take that idea and wed it to your bubble idea, correct?

Brunnermeier: That's correct. So essentially I would say the service flow has three elements, at least. So one is like standard convenience yield, because it has some transactional role. It overcomes some double coincidence of wants constraint. It might be due to some collateral constraint. It will act as a collateral constraint. So you have a Lagrange multiplier essentially relaxing a collateral constraint. And the new element we bring into the picture is that it allows precautionary savings and reinsuring yourself to idiosyncratic risk. So there is an economy like a Bewley economy or an Aiyagari type economy. Everybody has this idiosyncratic risk, uninsurable, idiosyncratic risk. And whenever, that's why you want to hold, for precautionary reasons, this safe asset. And we think this third element is quantitatively big.

Brunnermeier: So even when John Cochrane was all, "The convenience yield is tiny, can't make a difference," we think the third element makes a big difference. And that really makes it what's really the value of a safe asset. And that's this huge privilege you have, if you are able to issue this safe asset like US government is. And at certain points in time, even private companies will be able to issues some relative quasi-safe asset, which the safety might be lost in times of crisis. So you have to be careful not to lose the safe asset status. And that's what I want to come back, is that's why it's a bubble. Bubbles can burst, and the safe asset status can be lost. So I call it the safe asset tautology. A safe asset is safe, because it's safe. It's like a multiple equilibrium story. It's safe, because it's perceived to be safe, and you have to be careful in printed word to preserve the safety status. So that also leads to then what fiscal policy you want to do. You want to maintain the safe asset status, because it's such a privilege. You don't want to give it up.

Beckworth: Yeah, so I think there're a number of things that have happened over the pandemic that has increased these three roles, these three components to the bubble. But I want to also go back and ask this question about safe asset maybe in general. Do you think it's difficult to create a new, safe asset that can compete with the dollar system? And I want to go back to a paper by Arvind Krishnamurthy. And I'll put it in the show notes, but he and some coauthors make the argument that investors go to the dollar, because there really is no other alternative in terms of scale and size. There's kind of this path dependency, this locked in effect. Now you can still lose it, but when I look at the magnitudes, look at the net investment position of the US, look at the portfolio side, it's close to 30 trillion. Now that includes some stocks, but you look at fixed income, it's about 20 trillion. And that's what we owe to the rest of the world. Those are liabilities to the rest of the world. You look at the BIS's global liquidity indicators, they show trillions of dollars being created outside the US. And I'm just wondering, how would an upstart safe asset even be able to compete? Because you'd have to scale up so dramatically, what are your thoughts on that?

Brunnermeier: So there're two aspects. So one is there's always the argument, the safe asset has to be large enough. There's a shortage of safe assets. I think what you are seeing, US government is issuing a lot of that. I don't think there's so much shortage out there. I think the bigger problem is that, from a global perspective, a safe asset is provided asymmetrically. So whenever we go in a flight to safety modes or risk on, risk off, everybody is rushing into the US or in the German boom or Japanese yen. And that creates a lot of instability in global financial architecture.

Brunnermeier: So what ideally we would like to have is to have this safe asset be provided, where you rush in in times of crisis to be more symmetrically distributed. Right now, it is the case that everybody rushes into US Treasury, and then the US monetary policy has to react and shovel it back through swap lines, shovel it back to the emerging economies. You need a very active, interventionistic approach. I like architecture that is self-stabilizing, where you don't have to intervene all the time. I have made some proposals, how to do that, but that's-

Beckworth: Well let's actually go there. I know we were going to talk about this when we go to your book, but that is a question I have. And that is it seems the Fed is increasingly relied upon to save the global dollar funding system. So 2008, 2020. I'm glad that the Fed did step in and save the global dollar funding markets, because it could've been very ugly if it hadn't. But it does seem like, man, are we going to rely on this one institution in a haphazard way, kind of step in when things happen? What is your systematic way to solve this problem?

What ideally we would like to have is to have this safe asset be provided, where you rush in in times of crisis to be more symmetrically distributed. Right now, it is the case that everybody rushes into US Treasury, and then the US monetary policy has to react and shovel it back through swap lines, shovel it back to the emerging economies. You need a very active, interventionistic approach. I like architecture that is self-stabilizing, where you don't have to intervene all the time.

Share this

The Fed’s Role in Stabilizing Dollar Funding Markets

Brunnermeier: So first of all, I think the Fed did a fantastic job in March 2020, when it has to say that to save the global financial system. And even the US Treasury as a safe asset, people wanted to get out of that, because for a safe asset to be a truly safe asset, I should add, it has to be easily tradable. So it has to have a low bid-ask spread, and it has to be informationally insensitive. It can't be that there's a lot of asymmetric information attached to it, because you get the lemons problem kicking in and things like that, where some people have informational advantage over others, so then you get a low bid-ask spread. So you want to have easy trading, little bit of low bid-ask spread.

Brunnermeier: And they have stepped in as a market maker of last resort, so the market making pushed the low bid-ask spread down again. And then it worked again as a safe asset. So in general, as long as the Fed is doing that, it's fine. You have elastic currency, in a sense. There's some huge demand for dollar assets, and the Fed is just providing it. It's only if you at some point say, "Okay, we don't want to have a system where you have to intervene activity." And the Fed has to often take on a global perspective. It has to save the global economy. And it have a US mandate and have a global perspective, that's the ideal perspective it should have. So it's like a central bank for the whole global economy. Here's a tension there. The legal mandate is to focus on the US, while from an economic perspective, it should focus on the whole global economy. And in order to minimize this tension, you might want to have a different architecture, where the Fed doesn't have to intervene so aggressively when there's a flight to safety.

Brunnermeier: So what is my proposal in this regard? My proposal is actually that emerging economies can also issue safe assets. Okay, and so this way, when there is a flight to safety event, it will not only rush out of the emerging economies into the US. It might also rush into the emerging economies into the safe asset emerging economies issue. And what is this particular safe asset? I call it GloSBies, or Global Safe Bonds. What you would do, you would pool some government debt from the emerging economies and then branch it in a senior bond and a junior bond. And the senior bond would be protected by the junior bond, so it would be very safe. And then when there's a rush to safety, there will be still some rush into safety in the US Treasury, but there would also be a rush from the junior bond into the senior bond. And the senior bond stays within the emerging economies. So there will be less rush into the US and then hence less need for the US to react in order to counterbalance the flight to safety into the US or out of the emerging economies.

Beckworth: So what would be needed to make that happen? What barriers are there to that actually coming into fruition?

Brunnermeier: So at first what needs to be done is to set this up. You need some international agency to set this up in order to have some commitment on behalf. So account itself cannot do this branching, because there's a commitment problem. It cannot, so it has to be some international agency, multilateral agency to set this up. And it also has to be the case that that senior bond is regarded as a safe asset. As I said earlier, it's a bubble-safe asset means people have to believe in it. It's a safe asset. You have to jump to an equilibrium where whereby coordinate beliefs and high order beliefs and all these things that this senior bond is also a safe asset. So this requires quite some policy push and some understanding that the world is actually behind that. And investors will then flock around it. So it requires some belief coordination to jump into this equilibrium. But I think it would be a more stable world, and it would help, essentially, to minimize the flight to safety capital flows in times of crisis.

My proposal is actually that emerging economies can also issue safe assets. Okay, and so this way, when there is a flight to safety event, it will not only rush out of the emerging economies into the US. It might also rush into the emerging economies into the safe asset emerging economies issue. And what is this particular safe asset? I call it GloSBies, or Global Safe Bonds.

Share this

Beckworth: Do you remember Mark Carney's proposal, the 2019 Jackson Hole, his solution?

Brunnermeier: Yeah, so you can relate it to that. It has also some digital flavor to it, but you can relate it to that.

Beckworth: But it sounds like yours is more endogenous during a crisis as opposed ... His is let's just create this stock of alternative, synthetic, hegemonic currency asset. And I guess the challenge I had with that, and I actually had him on the podcast, we talked about this, was again, how do you scale up to compete with a dollar? But your argument is, in a time of crisis, you can maybe endogenize it. People, you create this safe asset when you need it. Is that right, or am I misstating?

Brunnermeier: So you have to establish it in normal times, and it will be beneficial in times of crisis. And it will require some coordination, and there has to be some multilateral institution saying, "We are behind it." And probably the US has to support it, saying, "This is a new architecture, global architecture we're going for. It's a more stable system. It benefits everybody, essentially." And this way, it will require, literally speaking, a champion equilibrium. So you design it first, and then you still need a jump that we believe that this senior bond is a safe asset and enjoys a safe-asset status. If you want to have a public component on it, because it could still be that the bubble is only on the US Treasury. But it's not on the senior bond created from emerging economies. So this still requires some handholding and perhaps initially some push from some official institutions.

Beckworth: But if this did come to fruition, the idea is that many of the problems we now see would be at least minimized, such as the so-called global financial cycle that Helene Rey talks about, the fact that the dollar is so far reaching, so dominant that, when a dollar fluctuates in value, the fed changes monetary policy. It affects everyone in a very distortionary way. So this would hopefully minimize that, correct?

Brunnermeier: That's correct, and it also would allow the US Fed to focus more on the domestic policy exclusively, because it doesn't have to deal with what are implications for the rest of the economy, which then has some spill-back effects on the US as well. But it would give more freedom for US monetary policy to focus on the domestic challenges.

Beckworth: No, I think that's a real benefit, because you think of some of the optics, bad optics that can emerge from the Fed bailing out parts of the world, at least that's the perception. I shouldn't say bailing out, but the Fed backs topping global funding markets, say, in China. There's this potential there for political blow back and for the independence of the Fed to be undermined, if this is perceived the wrong way.

Beckworth: All right, let me circle back to fiscal theory of the price level. This is where we started this part of the conversation. I want to end our conversation with your book. So the last thing I want to ask about the fiscal theory of the price level, if we take this new approach that you've presented, you take the primary surplus and the service flows, the net present value of both of those terms, how can a central bank use it? Or does it have any bearing for monetary policy? Because it is still focused on the fiscal side, but there is a bubble term there. And so is there any practical application for Jay Powell or other central bankers?

Brunnermeier: First of all, it's a different mindset. You have to take the fiscal side very seriously into account. So there's a belief out there, say central banks are totally independent. They can just do it and are totally in control. It works at the moment, but at some point it might bite you. We might be in this fiscal dominance region. So you really should take this seriously, and so that's one thing. The other thing is, I would like to add, on top of fiscal dominance, financial dominance. So you might be, as a central bank, not be willing to raise interest rates, because you might bring the fiscal side into difficulties. The government will squeak and cry that you ruin the budget, and that might be political pressure. But you might also, when you raise interest, you might create some financial crisis. And that also might prevent you from raising interest rates, if you have to in order to keep inflation under control. So both fiscal dominance and financial dominance, I think, are real threats to this space monitoring policy has. You have to take this fully into account, and I think this framework allows you to think about these things. And in particular, if you're worried that, if you raise interest rates, you create the financial crisis, you might not raise interest rates enough, if it warrants it for inflationary reasons.

Beckworth: Okay, well that's been a lot of fun. And I could go on another 30 minutes, but we've got to save the time for your book. So let's move to your book. It's just released, and the title of your book is “Resilient Society.” So give us the history of the book and maybe the executive summary of it, before you go down more closely into certain parts.

What Does a “Resilient Society” Entail?

Brunnermeier: So the history of the book is actually the two sources. One, you know I have worked on systematic risk and volatility paradox and the paradox of prudence. And so these are all the concepts I was thinking for quite some while. And then when the COVID crisis hit, I was actually setting up a webinar series. And I thought I had all these great speakers giving me insights, and I thought, I have to condense it. And actually it was Arvind Krishnamurthy said, "You have to write a book about that."

Brunnermeier: So I said, "Okay, let me do that." And then I wrote it, and then I said, "I need a common theme, or it has to have a red line going through the talk." And then I said, "Okay, resilience is actually a common theme." It's not only resilience be resilient society or resilient economy. How do we set up the economy? And then around this theme, I was then writing this book to get a better sense about resilience but drawing on all this work I did on the systematic and the volatility paradox and other paradox I have worked beforehand.

Beckworth: Okay, so give us the executive summary, the bird's eye view. What does this book argue? What's the main takeaway?

Brunnermeier: So the main takeaway, first it's like a definition. What is resilience? So, we should actually focus on: we cannot handle all the shocks. There will be a lot of shocks coming towards us, and some are unforeseen. Others are not unforeseen, so we have to really focus on, once we're hit by a shock, we take the shock. But we have to manage it in such a way we can bounce back. And that's different from robustness. Robustness is like you avoid any shock. You are like rigid and stick to it.

Brunnermeier: And I have an introduction. This parable from de La Fontaine, he's a French writer in the 17th century, who says, "Okay, you know there's an oak." An oak is very strong. All the wins, and it will not break, and it will be very rigid, not moving a little bit. And on the other hand, there's a reed. And the reed is constantly moving around back and forth. And it seems so weak and so volatile. But when the wind becomes extremely strong, the oak falls over while the reed bounces back. So essentially something which seems weak and volatile might be, at the long run, if it has the ability to bounce back, it actually might be a more resilient way of doing, a more stabler thing. So rigidity is actually not the way. What's really what's important is agility, being able to readjust, having some redundancies but flexible redundancies, which you can redeploy whatever the shock might be. And then this way you'll bounce back. And that's related to this volatility paradox that I mentioned earlier.

What is resilience? So, we should actually focus on: we cannot handle all the shocks. There will be a lot of shocks coming towards us, and some are unforeseen. Others are not unforeseen, so we have to really focus on, once we're hit by a shock, we take the shock. But we have to manage it in such a way we can bounce back.

Share this

Brunnermeier: Something which seems weak and volatile might be, at the end of the day, more stable than something which seems very rigid and very strong. But then when it falls over, it can't bounce back. So that's essentially the gist of the book. And then it applies it to all the different concepts in macro economics and global economics and also to the health aspects of the COVID crisis.

Beckworth: Well let's take that robustness versus resiliency comparison, and let's just apply it to some cases that come to mind. The great moderation, well known in macroeconomics. Do you think that's a case where the volatility paradox applies? We were too stable. We created expectations of stability, which then set the stage for the great financial crisis.

Brunnermeier: Yeah, to some extent that's true. Essentially what you do is you push the risk in further, into the tail. And then it, when you realize it, it becomes much bigger. So it seems very stable. It's like the oak. It's a little bit like a debt contract. The debt contract is very stable. It doesn't move. An equity contract is much more volatile, but an equity contract protects you more against crisis. While a debt contract, in the extreme, when it defaults, then it cause a lot of havoc. So it's very stable, very calm, like the oak. But if there's a big shock, you have to default, and then it creates a lot of difficulties, while an equity contract seems very volatile, constantly floating around and going up and down. Seems, wow, is this a risky economy. But the whole system is actually more stable if you're more equity financed than debt financed.

Beckworth: That's interesting. So some of the proposals that came out of in 2008, 2009 were more state-contingent debt contracts, which make the debt contracts like equity. And I guess that's the motivation there, right? There's more, better risk sharing between creditors and debtors, and the world is a better ... I mean John Cochrane wants all money to be equity. He has these proposals, same. But of course I guess the pushback I would give is there seems to be some appetite for fixed nominal debt contracts, right?

Brunnermeier: So this leads us back to informational sensitivity. Debt contracts, they're not informationally sensitive. So I don't have to worry that you have more information than I have. So there's no lemons problem, no adverse selection issues. We can just swap this amount of money. It's very standardized, and there is value to having something that is standardized. Now if I buy something from you, and then you offer me some shares, I have to not only value, I buy a shirt from you or something. What is the value of the shirt? You might take advantage. But then I pay you with shares, and then you have to think about, my god, he's selling with his shares, because he thinks the stock is not worth so much. And having a standardized, informationally insensitive asset as money, and that's what money is, it's a common measure. And we agree all upon it. We don't disagree on this. I think that has its value too, and that's why I don't think money can be replaced with equity. The informational insensitivity of money is a value on its own. That's why it's a unit of account. Once we have to bargain about what I pay with, then it makes it much more complicated.

Beckworth: Yeah, I think there'll always be some demand for some kind of fixed nominal safe asset debt type contract. They'll always be there. That's one reason maybe TIPS hasn't been as popular as people had hoped they would be. Let me throw one more thing out there on this better risk sharing, turning debt into equity. And you may know I'm a big fan of nominal GDP targeting. In fact, as a guest of the show, you'll be getting a nominal GDP targeting coffee mug later. And a lot of our guests love to drink out of it, and they look at the symbol, and they think about nominal GDP targeting. But one of the arguments, one of the more recent arguments, there's a large literature on it. But kind of since 2008 has been that nominal GDP level targeting actually leads to some of this risk share. And it's a substitute for state-contingent debt contracts and this following since.

Beckworth: So if a central bank were to perfectly be able to stabilize or to offset demand shock, so the only thing that was left were supply shocks, then what you would have is you would have inflation, countercyclical inflation. So during a recession, inflation goes up. During a boom, it goes down. And what that does, it makes pro-cyclical debt payments, which leads to more or at least better risk sharing. So during a recession, debtors get relief. But during a boom, say there's some unexpected boom, creditors get some higher real debt payments from their debtors. So they too benefit in the boom. So any thoughts on that from a macro perspective?

Brunnermeier: So I agree with you. You would like to have a better risk sharing, but if you have a nominal price level targeting, the problem is, you have to make a few distinctions between demand shocks and supply shocks in terms of ... And that's what you alluded to. That's, one has to be careful. And so if we read it the same way, it might be a little bit risky. The other thought I have is that if you go to average inflation targeting, the question is, over how many periods do you take the average over? Is it just the last three years or the last 10 years? Or it can't jump at all, there's an undershooting for the last 10 years. And that's, if I take a long period, then average influence targeting is like nominal price level targeting. And so the Fed is going into your direction, in a sense.

Brunnermeier: And the reason in the paper, what I worked on, what is actually the optimal duration, which you take the average over. And that depends very much on your debt maturity. So if people have a very long-term debt, then you would like take a longer average. If they are floating debt, then you take a shorter average, if the debt is very short term. So that's a little bit what, in order to get closer to that, so it depends very much how interest sensitive the debt is in order to get the right risk sharing going through monetary policy.

Beckworth: That's interesting. So you're saying that, based on whatever the current average debt maturity structure of the collective US economy is, you need to base your nominal GDP target, parametrize it around that fact. That would be a key fact.

Brunnermeier: That's right, yeah.

Beckworth: If you want to do the risk sharing element, that's interesting, great point.

Brunnermeier: In other countries you have floating mortgages. You might go for a different length of this window. And in the US, it's risk sharing in one direction. Or if you have a mortgage, it's fixed. It can't go up, but it can, through refinancing, can go down. So that makes the whole thing even more complicated. But optimum monetary policy framework has to take this into account in order to optimize the risk-sharing elements.

Beckworth: Absolutely. And we're going way off script here, but I'm going to go ahead and play on this point, because it's so fascinating. So one of the, to me, evidences of what a nominal GDP target could do, and just to be clear, nominal GDP target would be equivalent to a nominal income level target. Same, from an accounting perspective, same thing. But this past two years, the past year, 2020, what we saw was a massive fiscal and monetary policy response. And one of the things that it accomplished is it stabilized nominal incomes.

Beckworth: In fact, personal incomes are a little above trend. I think they're coming back down to trend. Nominal GDP is in a pre-crisis trend, whereas in 2008 there was this kind of unit root permanent shock that permanently lowered the trend path. But anyway, the point I want to make is, because we maintained and restored, stabilized the levels of incomes, nominal incomes, we didn't have a massive financial crisis. People were able to make their obligations, their mortgages, their leases. By stabilizing nominal incomes, we avoided some of the worst parts that we saw in 2008. Now again, this was a very unique situation. I can't always hope to replicate this in every crisis. But to me, this is a good at least case study, N of one, a good case study of, if you stabilized nominal income, the financial pressures are much less. Is that a fair assessment, you think?

Brunnermeier: I think it's a fair assessment, but you rightly pointed out, we can't just always add 20, 30% of extra GDP every crisis, so that may not work. Fiscal response is very different during the COVID crisis compared to the earlier crises in US. But also in particular in Europe, the case of ... So it was very aggressive and helped us a lot to overcome the crisis. The other argument is often made, and then coming back to resilience, that the global financial crisis was much less resilient. It took us a long time to get out of it, while from the COVID crisis, we bounce back much more quickly. So one argument is because the fiscal response is very different, and the monetary response is also much faster in the COVID crisis.

Brunnermeier: Another argument is it's the nature of the shock, that once you had some imbalances spilling up during the housing boom, the housing bubble and all these things. And then it takes a while to work out these imbalances, where in the COVID shock, we didn't have these imbalances. There was an exogenous shock hitting us. And then as soon as the vaccine came, we could remove the exogenous shock. And then the economy was in the right place already, so it's a different nature of a shock. In the one, you have to correct some imbalances, which were building up over a decade. The other one, it was just an exogenous shock coming, and then with the vaccine, we could remove it again.

Beckworth: Okay, well let's get back to the script here on your book. So we talked about resiliency versus robustness. We've touched on risk a little bit. We've talked about your volatility paradox. Sometimes we don't want to eliminate all risk, right? Otherwise we have the potential, the build up of instability. Maybe speak to that a little bit more.

Brunnermeier: Yeah, I think that's very important. But if you talk about resilience, mathematically it's mean reversion. And if you talk about risk, it's about standard deviation or variance. So one is a static concept. Risk is a static concept, and resilience is really, over time, some dynamic concept. And you can say, "Of course, in the long run risk, the resilience makes the world less risky as well." But what you really have when you look at risk, you have to categorize risk, whether it comes with resilience or not with resilience. So I am not worried about shocks. We should go into some risk. If you have to do some innovation and go into some risky environments, as long as we can control the resilience. What you have to watch out for is resilience killers. I call them resilience destroyers, or some risk which lead you into a trap. Once you're hit by a shock, you're trapped. You can't bounce back. So these risks are treated very differently from risks where you can bounce back, or you have some feedback loops. And I call this, in a society, you have feedbacks from externalities combined with strategic complementarities.

Risk is a static concept, and resilience is really, over time, some dynamic concept...what you really have when you look at risk, you have to categorize risk, whether it comes with resilience or not with resilience. So I am not worried about shocks. We should go into some risk. If you have to do some innovation and go into some risky environments, as long as we can control the resilience. What you have to watch out for is...resilience destroyers, or some risk which lead you into a trap. Once you're hit by a shock, you're trapped.

Share this

Brunnermeier: I call this feedback externalities. It's like you do something. You cause a negative externality on me. And now comes the strategic complementarity. I will react to that, which causes an externality back to you, a spill back. And then you will react further to it and causes more externality on me. So if you have not only a simple externality but an externality combined with the strategic complementarities, which leads to this feedback externalities, they are really to watch out for. So if you have a shock which will then translate into this feedback externalities, that's really ... And a third resilience destroyer or killer is the tipping points.

Brunnermeier: If you were a shock, and then you're pushed over a tipping point, then you can't come back either, and you will actually spiral out of control. And all these three things, they are really make ... These are risks, because they don't come with resilience you have to watch out for. But if you have other risks, where you bounce back from more easily, then you should take this risk. So on the one end, we have to work to make the whole economy and the whole society more resilient. And this requires some redundancies or some agility and flexibility and all that. On the other hand, we also have to watch out for risks from which it's very hard to bounce back from because of these three reasons I've outlined.

Beckworth: Yeah, that brings to mind an example from India. India was a very closed economy for many years, opened in the early '90s back up. And one of the industries that was protected was the auto industry. And I think it's Hindustan Motors, is the name of the company, but they had the same model in 1990 that they had like in 1950. The car had not innovated at all, because there's no risk taken. They were completely insulated from any kind of incentive or urgency to innovate and change.

Brunnermeier: Yeah, it's true. Actually these cars are still running. They are very robust and rigid.

Beckworth: Well there you go.

Brunnermeier: And the same was in East Germany. There's a Trabant. There was one car, it never innovated upon. You want to enthuse people to go into require and have a good bankruptcy court and things like that, that people go into risk, because there are benefits to society if people go into some well-calculated risk, as long as there's some resilience combined to that. And if you look at the US GDP, and you blot it on a log scale, it always bounces back to the same growth path, except for the global financial crisis. The growth path actually slows down, and there's a step down. So typically regular recessions, especially Fed-induced recessions, where you hike interest rates in order to control inflation, you bounce back to the old growth path. If you are financial crisis, typically you don't bounce back. You see it all the way clearly all the way in Japan. There's a huge break from the growth path beforehand. So that depends a little bit which crisis. Some crisis are more resilient than other crisis. And financial crisis typically don't have high resilience. They are hurt you in the very long run, because you don't bounce back to the old growth path.

Beckworth: There's more I could ask, and I wanted to bring up the Robert Lucas comment about we shouldn't really worry about business cycles. But I'm going to have to move on for the sake of time here. Let's talk about social contracts. What role does that play in resilience?

Resilience, Social Contracts, and Innovation

Brunnermeier: So the question then is, so by the way, Bob Lucas article is mentioned in the book. I go into that. But the social contract, I think, should be designed to make the society more resilient, which means they should be allowed to go into some risk. And if they fail, they can bounce back. We don't stigmatize them, and then they can bounce back. But there're two aspects to it. The social contract should also limit these feedback externalities I was talking about earlier. So what's different about resilience in a resilient society? Resilience can be about individuals, so there's a lot of psychological literature about it, who is resilient, how to become more resilient as an individual. But also societies can become more resilient. And societies become more resilient, if you control these externalities and these feedback externalities from others. So that's the social contract, has to limit these externalities and also provide a certain sense of social insurance as well.

The social contract, I think, should be designed to make the society more resilient, which means they should be allowed to go into some risk. And if they fail, they can bounce back. We don't stigmatize them, and then they can bounce back.

Share this

Brunnermeier: So which these are shocks, and I call them, you can also interpret them as externalities from mother nature. A shock is like a externality somebody else, mother nature in this case, imposes on you. And typically an externality is just I impose an externality on you. But the shock is just from mother nature. And you want to provide some basic support on that as well. And what's the difference between insurance and resilience? Resilience a little bit, it helps people to come back, not just by throwing money at them but give them the ability to bounce back. So you reskill them. You help them to bounce back, be part of society again rather than ... You have negative shock. You lose some money, and I give you money. That's a little bit simplistic. Resilience, I have a broader sense of helping them to come back to society.

Beckworth: And you have, I believe, three mediating institutions under the social contract idea. You have social norms, governments and markets, and they all have a role to play. Is that right?

Brunnermeier: That's correct. So essentially, then, you ask, "How do you implement the social contract?" And we economists, we always think there's a government. There's market failure we implement. But in reality, social norms play a huge role. You want to be part of society. There's a common identity to identifying with society. And so social norms, I talk a lot about Japan. You never have to enforce that people wear masks. It's all self reinforced by people around you. And there's, of course, a strong force. So one way to enforce the limitation of the externality is to limit the externalities and feedback externalities to have strong social norms. The problem with social norms is that they don't adjust in time, quick enough. So you see, if a COVID crisis, and suddenly we have to readjust and behave differently, the social norms are not changing so quickly. Laws can be changed much more quickly. Social norms are very nice, because they're self enforced. You don't have surveillance by the government. It's done by the people.

Brunnermeier: So one drawback of social norms is the slow adjustment. And also, sometimes in order for society to be resilient, they need mavericks. And these strong social norms, they don't allow so much their mavericks. And I like a society where mavericks can invent and think outside of the box. And that helps us, often, to bounce back. We need these mavericks. If you think about Elon Musk and others, they help really society to speed ahead. And you have to be tolerant of people thinking differently. And that's, I think, where social norms are still limiting.

Brunnermeier: And a drawback or the advantage of having the government imposing laws and internalizing externalities is, whilst you need a huge surveillance and all these things and enforcing with a stick and might also cause a lot of uneasiness in other dimensions, but that's another way to do it. And a third way is to say we trade property rights in terms of pollution permits and other things. That's another way to internalize externalities. Typically market solutions lead to more innovation, and I think often we only bounce back to innovation. It was just fascinating how the vaccine development, how you saw certain companies, BioNTech and Pfizer. BioNTech was doing MRNA at a switched on just from doing cancer treatment. They switched the whole company to do, in January 2020, before it really came to the rest of the world, the virus, they moved the whole company just develop this vaccine. And that's what you need, essentially, this agility to be resilient. The private sector is often more agile in this dimension.

Beckworth: Yeah, it was very inspiring to see these big pharma companies, some of them at least, being so agile and so willing to take on the risk. And now the government made it easier with the operation warp speed, the financing. But man, how inspiring. Elon Musk, go back to him also, I like this point you make about we need mavericks. We need people to once in a while break the china in the store, so we can see things from a different perspective. I mean social norms are important. We want to have civil society. We don't have breakdown, but yeah, you want that.

Beckworth: I recently had a guest on, talking about nuclear fusion, which is something people have hoped for many years. We have nuclear fission. We don't have nuclear fusion. But he argues we're really, really close. And one of the takeaways I got from this discussion in his book is that the private sector is really engaged in this now. It's been national laboratories, governments. But now there's a race to see who can be the first one to get break-even energy from it and then to commercialize it. And it reminds me of the space race we're seeing now, Elon Musk and different billionaires. So I'm hopeful that this is the kind of resiliency that we see, and it will benefit all of us, if we have more of that.

Beckworth: But let me take this idea of resilience. Let me apply it to some situations in the time we have left. There's a lot more in the book, and I encourage the listeners to get the book, provide a link to it, obviously. But some recent events and some that haven't happened but some recent events that suggest there wasn't resiliency there, and the first one I wanted to speak to is our electrical grid system. So there was that severe winter, I believe last year in Texas. And it couldn't handle it. So this is an example where they need resilience, correct?

Brunnermeier: Yes, so that's something where you, if something goes down, it's fine if it goes down for a few hours. But it has to come back on, though, and it shouldn't be. So that's, when I come back to the ... I didn't mention about the three layers of resilience. One is individual resilience, and then there is this network resilience, which is a more systematic resilience. And then there is this society resilience. And there's network going down. That's essentially there's network resilience, where you have to build in extra buffers. And typically the private sector puts too little emphasis on the resilience aspect. So there might be something, what we can learn from the financial crisis. So one of the big innovations of the financial crisis was to have stress tests done. So we might say, "Okay, we do some certain stress tests. Our companies have to do certain stress tests to develop more resilience." And some people will argue also for closed value chains and all this. We should do some stress tests.

Beckworth: That's a good point.

Brunnermeier: Or at least as the companies, do some stress tests and report it to us. And that's one of the lessons, what we have learned with global financial crisis, that the stress test can be a very powerful tool. Because it's not only about regulation, but it just forces the companies to realize on their own, actually there's a problem. And they report it, but it's more for self reporting, in a sense.

Beckworth: So going back to those mediating institutions, you got social norms, government and markets. In the case of electrical grid, sometimes the government has to step in, because there's a coordination problem that the market is not going to be able to solve by itself, something on that scale, something national or, in the case of Texas, the entire state. The state had to step in afterwards and bring reforms.

Beckworth: And I guess this leads me to a next, maybe, example. But it's more hypothetical, but these tail events in general from nature, you mentioned nature. So we could have a solar flair from the sun that could destroy our grid. There could be a weapon, an electric, magnetic pulse weapon that could go off. Asteroids could hit Earth. There's a lot of things that we don't ... Pandemics, we just went through a pandemic. There's a lot of things we probably don't think a lot about. I know there's some people doing it. I know NASA is looking closely at asteroids coming our way. But they don't have any good way to stop it, even if they identify it. Why is it that we have a hard time with these tail events that have huge ... If they're positive, they're huge, huge cost to society.

Brunnermeier: Yeah, because so it's not such an event which always is repeated. We can react. We know how to react to that. So often you have to just say, "We'll prepare for it as good as we can, and then we react flexibly." That's to be agile, as I come back and then try to bounce back, because we cannot prepare for all the crises. So I might add more of these aspects. We do some, probably, genetic modification at some point with humans. There might be some tail even there too. How to deal with that as the whole climate change aspects, there are a lot of antibiotic resistance. That's probably something coming for sure. So we have to prepare for that so the whole list of things which can happen. And once we all have some brain implants, computer brain implants, which will probably happen, there might be something going wrong with that. And so we have to think about potential things, but we can't prepare for everything. So essentially we have to have an attitude that, once a shock hits us, we will this deal with that. We prepare to the extent we can but then also have an attitude to react to that accordingly.

We have to think about potential things, but we can't prepare for everything. So essentially we have to have an attitude that, once a shock hits us, we will this deal with that. We prepare to the extent we can but then also have an attitude to react to that accordingly.

Share this

Beckworth: So, Markus, we are running near the end of the show, but I wanted to ask one more question to you, and that's about innovation. How is resiliency in innovation, which boosts long run economic growth, related?

Brunnermeier: So, I think the first thing is, innovation involves risk taking. So we have to take some risk in order to be innovative, and in the context of the COVID crisis, it might also help us. The crisis itself often opens up some pathway to go to a better innovative outcome. So there's this famous qwerty problem, which you're probably familiar with. It's like the keyboard is aligned in a particular way, so the keys are aligned this way. And that's because of old typewriters, where the hammers come in. And to minimize that these hammers get smashed, that's why the keyboard is designed the way it is. That makes no sense anymore, in modern computers to have that. But you're just locked into this keyboard. Everybody is familiar with this keyboard, and everybody sticks with it, even though a new design would be much better.

Brunnermeier: And there's many of these phenomena, where we're stuck in old ways of thinking. So working from home was stigmatized. Now COVID threw us in a different environment, and suddenly working from home is fine, or at least one day or two days a week. And so that's why COVID has some long run impact on innovation, it’s like a trend accelerator. But it also has some scarring effects. So on the one hand, you have this long run boost of innovation. But on the other hand, you also have the long-term scarring effects going on.

Beckworth: Okay, well on that note, our time is up. Our guest today has been Markus Brunnermeier. Markus, thank you so much for coming on the show.

Brunnermeier: Thanks a lot, David. It was a pleasure, good talking to you.

Photo by Win McNamee via Getty Images

Support Mercatus

Your support allows us to continue bridging the gap between academic ideas and real-world policy solutions.Donate