Tyler Muir on How to Understand the Fed’s Quantitative Easing

Tyler Muir is a professor of finance at UCLA. In Tyler’s first appearance on the show, he discusses how he became a leading scholar on quantitative easing, what things the Fed can learn in responding to crises, why QE matters, how QE transformed the bond market, the new “Tyler Rule”, QE’s role in the COVID Pandemic, and much more. 

Subscribe to David's new Substack: Macroeconomic Policy Nexus

Read the full episode transcript:

This episode was recorded on January 8th, 2025

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

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

Our guest today is Tyler Muir. Tyler is a professor of finance at UCLA and is a leading scholar on the Fed’s large-scale asset purchase programs, better known as quantitative easing or QE. He joins us today to discuss what we now know about QE. Tyler, thanks for joining the program.

Tyler Muir: Thanks so much. I’m really happy to be here.

Beckworth: Well, it’s great to have you on. Now, you truly are a leading scholar of QE. In fact, I had the privilege of listening to you present a paper this past year at the Atlanta Fed’s Financial Markets Conference, and I was really blown away by some of the counterfactual paths that you drew for yields relative to where they would have been had there not been QE over the past decade or so. I was excited to get you on.

Also, listeners, we’ll get to later in the show the possibility of something called a Tyler rule, which is similar to a Taylor rule, but for the balance sheet. If you’ve ever wondered why isn’t there a Taylor rule for the Fed’s balance sheet, well, this is your lucky day, because we’ll talk about something along those lines. Although Tyler is too humble to call it that, I’ll call it that for him. It was great meeting you at that conference and hearing your presentation.

Tyler’s Background

Tell the listeners a little bit about yourself. How did you get into finance, economics, Fed’s balance sheet, all that good stuff?

Muir: Yes, almost in some ways by accident. I was an undergraduate math major at Berkeley, and I didn’t really know what I wanted to do with my life. I took a couple classes in economics and finance, probably the end of my junior year. I thought this seems really exciting and amazing. It combined some of the math background that I have with something that felt really real-world and tangible and important.

Almost on a whim, I just applied to PhD programs. In some sense, the rest is history. I was pretty lucky that I ended up doing that. I graduated undergrad in 2008. Had I not gone and done a PhD program like most of my other friends—they struggled with pretty significantly finding jobs, or if they had them, they didn’t have them for very long. That was sort of a fortunate coincidence.

Beckworth: You cut your teeth on the Great Recession. You witnessed firsthand the casualties, the challenges of being a graduate, of being a student, going on to grad school. You had some great people on your committee. I was reading the letter that Arvind Krishnamurthy wrote for your Fischer Black Prize. We’ll talk about that in a minute. You had some great people around you. Tell us about that experience.

Muir: Yes, it was an amazing time. It was an amazing time to be at Northwestern when I was there. Speaking of the financial crisis, I remember during our orientation for PhD admission in September of 2008, we got the news that Lehman Brothers had failed. From a researcher perspective, it was just a fascinating time. You have this whole PhD life ahead of you. You’re in one of the most interesting periods that we’ve had in a very long time from a research perspective.

Then on top of that, I was incredibly lucky to have people there, my advisers and other people I took courses with, that were experts on those issues. My committee, Arvind Krishnamurthy was my chair, Andrea Eisfeldt, Ravi Jagannathan, and Dimitris Papanikolaou. I’m still in close contact with all of them. I remember Arvind, in particular, was pretty interested in how the financial sector shapes asset prices. That wasn’t something that was in mainstream asset pricing at the time. I took his course, and it was much more of a theory course. Then I was also taking an empirical course on data and asset pricing. 

Most of my dissertation was just, well, I could combine those two and take an empirical approach to looking at financial crises and intermediation and how that affects asset prices. I was just fortunate to have the right people there at the right time and have it all work out the way it did.

Beckworth: While you’re working on your dissertation, your graduate classes, were you getting interested in the Fed’s QE programs? What ultimately led you down this research agenda, where you are one of the leading experts now on this topic?

Muir: QE wasn’t as much directly on my radar during the 2008–2010 period. I started really getting interested in the QE research during COVID. Part of that was just when the Fed came out during COVID, it was QE infinity, we’ll buy an unlimited amount. They started getting involved in corporate bond markets. The response was just everything from the 2008 playbook in a week, and then potentially much, much more. I just thought, “Okay, this is a huge deal. I need to understand this better.”

Beckworth: All right. Now, I had mentioned that you won the Fischer Black Prize, and this was in 2025, so you just got it. This comes from the American Finance Association, which awards the Fischer Black Prize to the top finance scholar under 40 who best exemplifies the namesake Fischer Black’s commitment to rigorous research that significantly influences the practice of finance. Tell us about that. It must have been exciting. It just feels great. You’re a young person. Your career has taken off, and you get this amazing reward.

Muir: Yes, it was incredible. It’s incredibly humbling. If you look at the people that are on that list, it’s incredibly humbling. It’s one of the biggest honors in finance that you can receive. It’s a little bit odd in some ways because it is awarded to an individual person. When I look at my career, it’s all collaboration with co-authors. I owe them a huge debt of gratitude for getting me there. Yes, just in the nick of time for being under 40. I actually had several people comment to me afterwards, “It’s really well deserved, but I had no idea that you were under 40.” 

Beckworth: That’s awesome. One more personal side of you, Tyler, before we move on to your research. I found out that you are an avid surfer. In fact, there’s some images of you surfing online. Tell us about that.

Muir: Yes, absolutely. It's something I've done most of my whole life. I grew up, actually, in Southern California. I probably started surfing when I was around 13 or so. It’s something that I do fairly regularly. It’s great exercise, but it’s also something that has also allowed me to travel and see a lot of the world. I’ve done surfing in Indonesia, and Chile, and Mexico, and Europe, and Canada, you name it. It’s always interesting to go new places and do that there.

Beckworth: Are you one of those surfers who goes out looking for the biggest wave? Do you keep a record of how high you’ve gone?

Muir: I wouldn’t say that that’s exactly what I’m seeking out. If I happen to be somewhere and the waves are really big, then I’ll go for it. I’ve surfed some fairly big waves on the North Shore of Oahu in Hawaii and in Indonesia as well. I wouldn’t say I’m all about seeking the thrill of it.

Beckworth: Yes. It’s great to hear these stories. I recently chatted with Atlanta Fed President Raphael Bostic. He’s an avid birder. For him, it’s a hobby. It’s a passion. It’s also therapy for him. I imagine for you, too, getting away from all of this and being able to escape into your world of surfing is probably good for the soul, as well as traveling around the world.

Muir: Absolutely. Then when you’re out there, there’s no phone, there’s no device. You’re not thinking really about anything else except for what’s right in front of you. That’s huge.

Financial Crisis and Risk Premium

Beckworth: All right. Let’s jump into your research. I want to begin with the 2017 QJE paper titled “Financial Crises and Risk Premia.” This is an interesting paper because you really make the case that financial crises are different than your garden-variety recessions. There’s ordinary recessions, and there’s financial crisis-driven recessions. Walk us through this paper and why we should be more vigilant and careful when we see financial crises emerging.

Muir: The main argument in the paper, it’s studying financial crises—and I’m finance person, so it’s really about their effects on asset markets, on asset prices. Obviously, that’s a big part of how they transmit through the macro economy as well. There was some confusion or some debate—again, this was all happening around the 2008 financial crisis. Well, we see asset prices and asset markets being really depressed. Is that just because we have some bad macroeconomic shock, or is there some amplification of that effect coming through the fact that the financial sector is particularly hit, credit provision is impaired, all of these sorts of things? 

The idea in that paper to try to do that is to say, well, let’s look at financial crises and see what happens in markets during those periods and compare it to other really bad macroeconomic events. Garden-variety recessions, particularly deep recessions, wars, natural disasters, other things that have shocks to the macro economy. The big thing that you see in the financial crisis episodes is asset prices fall by much more, and they fall by much more than is warranted by fundamentals. 

In the other episodes, you still see a big drop in asset prices and asset markets being affected. Most of that, or the large share of that, seems to be just because expectations of cash flows are falling. Firms are going to be less profitable, earnings are going to drop, things like that. Financial crises, it looks like you get this huge extra amplification. That’s this extra risk premium channel on top of just what’s going on with the macro economy. It’s a way to try to separate those two things out a bit.

Beckworth: With the 2007–2009 period, the Great Financial Crisis, would this be an example of that type of experience?

Muir: Yes, absolutely. The paper that I wrote put together a large number of financial crises historically across a bunch of different countries going back 100 years. You need a big panel of data to get enough of these to really say something statistically meaningful. An interesting thing is, then, well, how does 2008 compare? It’s very much in keeping with the higher experiences that we saw there.

Beckworth: Now, does this help shed light on why we had such a sluggish recovery from the Great Financial Crisis?

Muir: This, on its own, I would say, it’s not that obvious from this evidence alone. One of the things is, you see very clearly in financial markets, these big drops in asset prices, and they look like they’re malfunctioning during the acute phase of the crisis. That takes some time. On average, in the historical episodes, it takes about two years for that effect to fully go away. The drops in economic activity that we see tend to be a bit more prolonged than that. In some sense, the asset markets look like they recover and rebound a bit faster than the macroeconomic forces would imply.

Beckworth: Okay. I bring this up because during that time, or maybe right after that time, in the 2010s, there’s all this discussion. Well, this is different because it was a financial crisis. They tend to take longer to heal, to get through, to walk through. It’s interesting to think about that period. I also want to bring this up because the fact that you went back and did this big panel study, cross-country, cross-time, on financial crises, reminded me of another paper that I really enjoyed, I found at least useful. It’s a little bit outside my lane, but still, I think it’s insightful. Putting your two papers together really is, I think, a rich story to tell. 

It’s a paper in the, I believe, European Economic Review 2016 by Manuel Funke, Moritz Schularick, and Christoph Trebesch. It’s titled “Going to Extremes: Politics After Financial Crises, 1870–2014.” What they show is when you have a financial crisis, there is this pattern of populism emerging, nationalism emerging, and it sticks around for a while, and it persists.

To me, man, it really shows what we went through is no different than some of the other historical cases. Then, throw your story on it as well, this is a pretty persistent pattern. It was just fascinating to see this overlap between what you did, what they did. Again, two different measurements, dependent variables you’re looking at, but financial crisis is the common thread here.

Muir: Absolutely. One of the things taken in their paper is one of the reasons that financial crises tend to be really costly and persistent is that while the financial markets themselves might recover within a couple of years, it creates all this backlash against finance. We saw this with the Occupy Wall Street movement. It creates a lot of political uncertainty. It creates all these things. Those things can drag on for much longer and have macroeconomic effects for much longer.

Beckworth: Financial crises really matter. We need to take them seriously, do what we can to come up with good policies to prevent, and if we have them, to respond as best we can. 

Intermediaries and Asset Prices

I have a question for you. Have there been any changes in financial crises? Now, again, you noted they’re very similar in terms of their responses, but you have other research that points out that the market macro structure, as you call it, has changed. There’s been some big changes, at least in recent periods. I’ll highlight three things that you bring out: the growing importance of financial intermediaries, their balance sheets versus household balance sheets, the rise of passive investing, and the growing dominance of central banks. Maybe walk us through those, and particularly that first one, the growing importance of financial intermediaries. You have a paper, “Intermediaries and Asset Prices,” and you really push this point that their balance sheets, what they do, is almost more important than what households do, despite the standard view of finance.

Muir: Yes. Just to walk through that last one, maybe first, a little bit of the standard macro models that are used to think about asset prices basically have some representative agent that’s determining the prices of the losses. Now, what that really means is effectively every agent in the economy is paying attention to the stock market and the bond market and individual stocks and individual bonds and their prices at all points in time and making a decision, is that fairly priced? Should I buy more of it? Should I buy less of it?

When we really look at the data, that just doesn’t seem like it’s very realistic. For a lot of asset markets, households don’t really even know how to value a mortgage-backed security or a credit default swap or anything like that. A lot of them probably don’t even know what those things are. Even for something like the stock market, a lot of households are investing in it quite passively. They’re just putting money into a 401(k) every month. They’re not really thinking too hard about exactly what the prices are.

Prices end up getting determined by the much more active players in financial markets, which tend to be much more like financial institutions. Depending on which market we’re talking about, you could think about banks, you could think about hedge funds, all sorts of different financial intermediaries. Now, if there’s a shock or there’s something that’s going to affect their balance sheets, that’s going to have a first-order effect on financial markets directly. In my mind, at least, this all seems relatively obvious or straightforward, but that was not incorporated into asset pricing and asset pricing macro type of models before this.

Then, to get back to this issue of what we call—and this is work with Valentin Haddad, who’s my co-author here at UCLA—this issue of market macro structure, we’re basically saying if you look broadly at financial markets, you really want to think about the organization into those markets, who the key players are, and what it is that they care about or risk constraints that they might face. QE is a great example where the Fed really didn’t used to be much of a player in bond markets before 2008. Then, as of a couple of years ago, their balance sheet peaked at almost $9 trillion.

They became a huge owner of Treasuries, mortgage-backed securities, and these things. When you have a big shift like that of a new player coming in and absorbing a big fraction of a market, it’s going to affect bond yields or asset prices more generally, pretty directly. It’s important to keep those shifts in mind: how the financial sector has grown, the way in which people invest, which has changed a lot over time, and then new players, like the Federal Reserve stepping into markets.

Beckworth: This is so interesting, thinking about market macro structure, because these trends that you’ve highlighted that there’s rise of passive investing. So households, you and me, we put our money in retirement accounts. In fact, we’re told, just don’t look at it, leave it there, right?

Muir: Exactly. That’s the advice that we give to—

Beckworth: You give to your students.

Muir: —MBA students here at Anderson, which, for most of them, is good advice. Then by definition, they’re not buying and selling based on the price and all of the stuff. They’re blindly doing it, so the prices are determined by someone else, though.

Beckworth: This is interesting because that development, the growing passive investing, at the same time that the world’s becoming increasingly financialized—if you look at, for example, global debt to global GDP, it’s almost like a hockey stick. There’s more and more finance of the world. I was looking at this, preparing for a show, looked at some IMF data: 1980, you see this point where it goes up, and then it goes up even more in the ’90s and 2000s with globalization. On one hand, we’re becoming more financialized. On the other hand, many households are becoming more passive.

Somebody’s got to step in and fill that void, and you’re saying it’s these intermediaries, their balance sheets that are more and more important. Which is a nice segue to the next thing I want to talk to you about, and that is QE, because they’re going to play an important part of your story that you’re going to tell us. I want to go here, Tyler, because as a macro economist, I learned from people like Michael Woodford and some others, and they tell us that maybe QE doesn’t really even matter from a macro perspective, but we know that it did.

Looking back, it seemed to have mattered. Your research shows that it did matter. I want to invoke this famous line that Ben Bernanke shared. I believe he was Fed chair when it came out. He said, “The problem with QE is it works in practice, but it doesn’t work in theory,” which is a great line. What he’s invoking is this idea of Wallace neutrality, which is an application of the Modigliani-Miller theorem applied to QE. Maybe I’ll let you flesh that out. Let me give, in my mind, what would be a concrete example of this or an application of this.

Michael Woodford has also had something similar, his irrelevance proposition for QE. Him and Gauti Eggertsson for many years were publishing papers saying QE won’t matter unless you do certain things. Let me make this concrete example, and tell me if this is a good application or not. It may not be. Let’s just say it’s normal time, so it’s not a crisis. The markets aren’t in distress. If the Fed were to go and buy up some asset that has more risk on its balance sheet—let’s say it’s a corporate bond, let’s keep it simple—it’s taking that risk out of the private sector and putting it on its balance sheet. 

You might say on the surface, “Uh-huh, it’s reduced risk in the economy and the private sector.” Then the macroeconomist says, “Hold on, the Fed is a part of the government. Who ultimately backs up the government? The taxpayer.” You really haven’t removed the risk. The taxpayer, who’s the private sector, still is going to bear that risk. You really haven’t done anything other than moving around on balance sheets. It’s still there. You haven’t eliminated it. Is that the idea behind that Wallace neutrality?

Muir: Yes, that’s exactly how I would put it. The Modigliani-Miller view is, if you’re not creating or destroying risk by doing this, you’re just reshuffling it around. If I consolidate the balance sheets of everyone far enough, it’s not really going to matter. It’s left pocket, right pocket stuff. The way that you put it is probably better than what I just said. Yes, I agree.

Beckworth: At some level, it’s intuitive. Again, if you’re working with a representative agent model, this result falls out pretty easily. Of course, there’s been a lot of research, yours, there’s been others like HANK models, these Heterogeneous Agent New Keynesian models. They also bring in frictions that modify this result. 

QE and Why It Matters

Let’s talk about your work here. I want to draw from a recent paper of yours from the AER, “Whatever It Takes? The Impact of Conditional Policy Promises.” I want to use this to help us understand, Tyler, why QE does matter. What did you guys do in this paper? What did you find? How can you use it to respond to the Wallace neutrality or the idea we just shared about you can’t really get rid of risk?

Muir: This paper, what we really are doing is we’re arguing that QE should be viewed as a state-contingent policy. It seems obvious that the Fed doesn’t just do QE or asset purchases broadly, or randomly. They’re going to tend to do it during crisis periods. 

That paper focused mostly on COVID in 2020. It focused mostly on the announcement of corporate bond purchases. There, the evidence is really stark. The Fed comes out, I believe, March 23 of 2020. They announce this new facility that they’re going to purchase corporate bonds, or that they’re willing to. You see a recovery in corporate bond prices right at the second that they make those announcements. That, in total market value, is something like half a trillion to a trillion dollars, depending on exactly what window that you use. Right away, that market impact is telling you, well, apparently, this announcement does matter. It must be that these purchases have some bearing.

The more interesting thing in that paper is if you look at what the Fed ultimately purchased, it was something like $13 billion. They did that a few months later. They purchased almost nothing. The question in that paper was really, well, how can they have such this huge impact when they make the announcement when they ultimately really didn’t do much in terms of purchases? Our argument is that it’s this state contingency.

The really big, important thing now was they sent a signal to the market, we are willing to backstop this market. We’re willing to come in and purchase corporate bonds. The market’s thinking, well, they’ll do even more of that potentially if the situation goes south, if things get significantly worse. The market views this backstop, or that they’re willing to do whatever it takes to keep that market functioning. That will immediately trigger a huge response.

Beckworth: The title of your paper, “Whatever It Takes,” we know the famous ECB president said that about the ECB bond market. Maybe use that example as well.

Muir: Yes, that’s another good one. There, this promise to do whatever it takes, no matter what happens, was explicitly evoked in order to try to stabilize markets today. That was in Europe with the ECB. There, the situation was really about sovereign bond yields, especially for countries like Greece, and Italy, and Portugal. We have another paper that’s looking at that. Again, this is joint work with Alan Moreira and Valentin Haddad. All of these QE papers are written with that team. There, we have a paper looking at that, and we find that sovereign spreads are really collapsing when those announcements are made.

Beckworth: Now, this ties back into this point we were discussing earlier about the importance of intermediaries and asset prices. You have a paper on this. It’s their balance sheets that this is really responding to, and why it really interacts there. That’s where it really packs its punch.

Muir: Yes, that’s right. Going back to your earlier point, which I think was a great one, if the Fed comes in and buys corporate bonds, that risk doesn’t disappear. It gets transferred to the Fed’s balance sheet, which ultimately, if you trace it back, ends up being a risk onto the taxpayers. The point is that in those standard models, the reason why it ends up having no effect is because the taxpayers realize that.

They’re participants in all of these asset markets, and they go try to dump their corporate bonds. In reality, they’re not really participating in those asset markets. The people that are participating are the financial institutions. For them, you really are taking risk off of their balance sheets, and by doing that, you are going to affect asset prices quite directly.

Beckworth: All right. Is it fair to say what you're finding is kind of an asymmetric effect or even asymmetric policy response, that QE really packs a punch when you have market dysfunction, when there’s a lot of stress on balance sheets? Is it less effective during, say, normal times because of that connection?

Muir: There can be asymmetric effects about when you do the purchases, for sure. The big thing that we’re finding is it can have an effect even during normal times because of the expectation that you will come in and purchase a lot of corporate bonds, Treasury, whatever it is, if we end up in another crisis or market dysfunction, like we saw in COVID. 

If I think about Treasuries, a lot of the reason that people like to hold Treasuries is because they’re the ultimate safe asset. People know they can always sell them for a good price, no matter what’s going on in the economy. If they need to raise cash quickly, they can sell it for a good price. They care a lot about being able to selling Treasuries usually for a good price during the middle of a crisis. Well, this type of policy is really going to support that. What’s that going to do? That’s going to increase their demand for these assets up front, the demand for Treasuries up front, and lower their yields even in normal times.

I would say that it’s true that there’s more of an effect, in some sense, of the direct purchases during the crisis periods. The fact that you’re now used to the Fed stepping in during these crisis periods to do that will mean you’re going to have an effect on things even during normal times.

QE and the Bond Market

Beckworth: It’s more of an appearance that the effect wears off, when in fact it’s still there and continues. You just see it in the midst of the crisis. In fact, another paper we’re going to get into in a minute, but I’ll jump ahead, your AER ’24 paper, “Asset Purchase Rules: How QE Transformed the Bond Market.” If I read it correctly, you have 10-year yields, your estimate falling to 115 basis points, 40 basis points from the purchases, but 75 from that insurance, that contingent belief. That’s almost double what the actual purchases did.

Muir: Absolutely. That effect is sort of unconditional, even during normal times. That effect is all coming from this thing that you know that the Fed has communicated to you, effectively through its actions and words, that it will come into these markets during crisis periods or during periods where the markets are not functioning very well. That unconditional effect we’re finding is really big and, in fact, is sort of the main effect of QE, not just what you’ve purchased so far.

Beckworth: Well, let me ask this question, Tyler, because I’m someone who’s gone to bat for QE. I think one could be critical, maybe ’21, ’22, that it was a little excessive there. I would say in normal times, the Fed was responding endogenously to the macro environment. Clearly, during crises, 2008, 2020, markets were collapsing, the Fed had to step in. Then how do you explain QE2, QE3, beyond 2020? In my mind, it’s always the Fed responding to the circumstances.

I have pushed back against people who’ve said the Fed artificially lowers the equilibrium real interest rate. What does your research suggest? Is the Fed actually doing that, or is the Fed just maybe removing some risk premium? That 75 basis points over that period, how do I think of it? Is it artificial? Is it what needs to be done? Where do you land on that?

Muir: In what we find, it’s all on the term premium. It’s all about removing the term premium. It’s unclear. Now you’re getting to questions like, is this a good idea or a bad idea? How much is beneficial or how much is not?

Beckworth: Fair enough.

Muir: That starts to get a little bit tricky for some of the issues. It seems fairly clear to me that you want to intervene during these market malfunction states to keep the market from breaking down. Even if people know that ex ante, that you’re going to come in and respond during these market malfunction states, if that’s a force that’s lowering the yields ex-ante or during normal times, that seems pretty reasonable to me.

Beckworth: To be fair, it’s 75 basis points. They didn’t reduce it from 5% down to 0%. Magnitudes are important to keep in perspective here. One last question about this paper. Typically, you hear three channels in this literature: the portfolio balance channel, the signaling channel, the market functioning channel. How should we think of those in light of this finding in your paper about this contingency, this state dependency rule, or approach to QE?

Muir: We’re finding that a big effect is coming through this portfolio balance channel, but it’s not just the portfolio balance channel that happens when you make the purchases. It’s ex ante, knowing that they’re going to do these actions during these states, is going to affect your portfolio balance response even today. We find a bit less of a role for the signaling channel. It’s hard to separate the market malfunction from just this overall removal of duration risk that the Fed is doing.

The Tyler Rule

Beckworth: Let’s get back to this other paper I just mentioned, the 2024 AER paper, “Asset Purchase Rules: How QE Transformed the Bond Market.” Listeners, this is where you get to learn about the Tyler rule. Tell us about what you did. This is the paper, or some derivative of it, that you presented at the Atlanta Fed Conference. Is that right?

Muir: That is right. Yes. This paper is really about trying to quantify the effect that I was talking about in the earlier paper. In some sense, we’re always writing the next paper because of the things that we were bugged about in the last paper. In the prior paper that I was discussing, we really documented this state contingency effect very clearly. What we weren’t able to do very easily is say, “Well, if people know or expect that the Fed is going to come in during these states, how much does that insurance effect that the Fed is providing matter for yields even in normal times?” That was really the question that we wanted to go after. 

We had a couple of ways of trying to do that. One way in which we did that was we took yield data up until just before QE started, correlated with a bunch of macroeconomic variables, then projected forward what yields would have looked like, the counterfactual, versus what they actually looked like. That’s where you see this big gap of about 100 to 150 basis points, where the actual yields are much lower than they should be based on these prior relationships.

Now, what’s interesting is that entire gap you can explain just by the yield response of Treasuries on the days where they make these few QE announcements. It says that QE news really looks like it’s explaining why yields are so much lower than they are.

Beckworth: Interesting. Again, the result, 115 basis points, super fascinating. What about the policy rule? I call it a Tyler rule. You’ve talked about it, but is it something like an investor or an academic can pull from your paper? Is it comparable, in other words, to something like a Taylor rule?

Muir: It is comparable. It’s a lot more vague in the sense that the Taylor rule really specifies the output gap, the inflation. You’re really specifying very clearly what the Fed’s reaction function is for moving short-term rates based on these two things that are going on in the economy. I think for QE and for the Fed’s balance sheet, we have a broad sense that the Fed will do QE during crisis periods, maybe at the zero lower bound when they run out of other tools if they want to lower long-term rates.

We don’t have a sense of exactly what those stake variables are that we have for the output gap and for inflation. Part of that is because it’s also much harder to measure those things. If I want to measure the future path of short-term rates, I have a lot of data in which I can do that. I’ll have Fed funds futures or all sorts of other things. Then I can see when inflation news changes, how does that path change? You can back out these rules and standard monetary policy much more easily.

There’s no equivalent of clear expectations of the Fed’s balance sheet. You can’t get the conditionality of that very easily, as different variables in the economy change. It’s not as clear exactly what the rule depends on, or what the reaction function of QE depends on in terms of very clear measurable indicators. I think it’s still clear that something like that is there and that the effect that people are anticipating what you’re going to do next, whether it’s QE or QT, that seems very clear that the market cares about that, and that’s going to affect bond markets today.

Beckworth: Well, you have brought clarity to this issue, so we appreciate that. Has anybody else called it the Tyler rule yet, or is it just me?

Muir: My three co-authors tried to push that. I thought it unfairly pushed credit onto just me and not them, but I do think it’s pretty funny.

Beckworth: I’ll tell you what, we’ll call it the Tyler et al. rule, so listeners out there, please start citing it as the Tyler et al. rule, just so we can have something comparable to the Taylor rule. Look, and to be fair to you guys and your rule, even the Taylor rule has some unobservables. It has output gap potential real GDP, some implicit neutral rate, which we haven’t observed. Every attempt has something.

Now, clearly, they have more that they can work with than you guys, but I think it’s great that you guys are pushing this and I’m sure it’ll continue to push, and we’ll get more clarity as time goes on. Just quickly, I want to go back to the COVID QE or the pandemic response. It does seem much more aggressive, unique. As you mentioned, they went into markets they hadn’t gone into before the bond market.

When Selling Goes Viral

You have another paper: “When Selling Goes Viral: Disruptions in Debt Markets in the COVID-19 Crisis and the Fed’s Response.” I think you’ve touched on that a little bit, but why do you think they tried so much more? What’s going on here? Why go into corporate bond markets, whereas before they were just limited?

Muir: Yes, great question. I think it was a couple of things, actually. I think one is the speed and severity of COVID was really unprecedented. The financial market reaction was so fast and so severe that I really do think that they felt like we have to do something right away to respond to this. I think there’s a second thing, which is coming out of the 2008 crisis, I think there was a sense of policy in general that we didn’t do enough, and we didn’t do it fast enough.

There was some sense of, well, we don’t want to make that mistake again. It’s better to go too aggressive than it is to go too soft and get this really long, slow macro recovery. I think it was a combination of those two things. You have to remember, going into these other markets as well, the corporate bond market was something that the Fed had said that they wouldn’t do beforehand.

You look at the situation that you’re presented with and the corporate bond market was completely collapsing. There were massive dislocations in corporate bonds. We give an example in that paper of a bond issued by Google, who was AA rated at the time. They have enough cash on their balance sheet they could pay off all their debt the next second if they wanted to. That was still trading at a massive discount at the time. That just gives you a sense of how disrupted it was. There was a sense that if we don’t get this fixed soon, we’re going to turn this already big crisis into another financial crisis.

Beckworth: Well, lots of good stuff there for Fed officials and others to chew on. Actually, Tyler, you’ve been doing some work as well with the ECB, Bank of England. Is that right?

Muir: Yes, that is right. We’re applying a lot of these similar concepts to the European experience. There, the big thing is what investors in the euro area have really gotten used to is the ECB tends to want to intervene to keep sovereign spreads relatively tight. Anytime you see sovereign spreads blow out, they will come in and intervene.

We’re basically making the same argument there, that if you get used to that for long enough, then that’s going to be a force that, ex ante, in normal times, is going to keep sovereign spreads really compressed because you know if they were to increase, the ECB will come in with policy that reduces them back down.

There, we’re arguing that that’s a force that’s kept sovereign spreads really compressed. We have a bunch of event studies in that paper that are the outlines of what we’ve done here.

QE During COVID

Beckworth: All right. We look forward to seeing that. Now, I want to ask one more question about the pandemic QE. This is a big question. This may be a little bit beyond what you guys are doing in your papers. One might look at the pandemic QE, I’ll call it QE4 for short. If we go beyond March and April of 2021, the markets were crashing. Go beyond market QE and we look at the longer QE that lasted, I believe, until 2022, it sure seemed to have a much bigger effect on the real economy and inflation. Whereas QE1 through QE3, I’m sure it had some marginal effect, but there’s all this debate in the literature, how big of an effect on inflation, as opposed to what you’re showing is very large effects on asset prices, asset yields.

You have this, at least in my mind, this stark contrast between QE4 and what happened before. Let me give my response to that. Why is that the case? I’m going to see what you think. I’m going to do that by invoking something that Ben Bernanke said in his presidential speech at the AEA meetings. I believe he said this as a Fed chair at some point in the past. This is from his talk in 2020. He gave it at, again, the AEA presidential address. 

He goes, “The popular use of the term QE blurs a useful distinction. QE, as practiced by the Fed, was importantly different from the QE undertaken by the Bank of Japan before the crisis. The former emphasized the effects of buying longer-term assets on longer-term interest rates. The latter, the effects of purchases on bank reserves and the monetary base. In general, increases in reserves, per se, should have limited benefit in the liquidity trap, being only a swap of one sort of short-term liquid security for another.”

He’s pointing out what he called, I believe before, credit version of QE or financial asset version of QE versus some monetary, maybe old-school monetarist view of QE, focusing on the liability side. In the late ’90s, early 2000s, there was a lot of work being done by people like Michael Woodford, for example, I mentioned. They talked a lot about what’s going on in Japan and worries about it coming here. One of the things they really pushed, and this is where I delved in a little bit, is level targeting. As a listener of the show, you will know I love nominal GDP level targeting. I’m willing to settle for price level targeting.

One of the key elements in that, or what it implies, is a permanent increase in the monetary base or a permanent increase in the Fed’s balance sheet that’s expected to persist. Michael Woodford and others, they criticized the original Japan QE from 2001 to 2006 because it was temporary. It was focused on the liability side, it was temporary.

When I look at QE4—here’s the story I’m going to tell—it was a permanent increase in the Fed’s balance sheet. It was basically a helicopter drop. It was both fiscal policy, monetary policy, effectively, a fiscal drop, an unfunded fiscal drop. It was facilitated through the Fed’s QE program. To me, that’s the fundamental difference between, say, QE4, or at least the latter part, and QE1 through QE3. That’s why we got inflation and everything else. What are your thoughts on that?

Muir: Yes, those are great points, I think. You do have to be careful because there was a lot going on the fiscal side at the same time. I definitely agree that both the size of the increase in the Fed’s balance sheet during the COVID episode and the persistence are probably playing a big role.

Now, the Fed’s balance sheet has shrunk somewhat. It’s still at $6.5 trillion. It’s still really big, and I don’t see that going back down to zero anytime soon. I agree with you that that should have more long-lasting effects.

Beckworth: That is my takeaway. If you look at some of the conversations, and I was a part of them after 2009, 2010, one of the critiques you heard a lot about was the Fed needs a permanent increase in the monetary base. In fact, I believe it was 2012, Michael Woodford talked about this at the Jackson Hole speech that he gave.

The problem is, if you look at this from a consolidated government budget constraint, the only way to get a permanent increase in the monetary base is you got to commit to running deficits in the future, you’re not going to raise taxes. It requires both sides of the macro policy equation to cooperate. Clearly, we got that in ’20 through ’22. Massive deficits, it was signaled that it was going to be unfunded. It was clearly non-Ricardian. In other words, households, when they got those checks, they got the unemployment benefits, they truly viewed it as wealth. They had no inclination whatsoever they got to pay it back in the future. Anyways, that’s my takeaway.

I guess if I had to make a policy point out of it, I would say, well, use that a little bit more carefully. I think they overdid it, but that’s something I think useful to think about going forward. 

Now, in the time we have left here, Tyler, I want to go on to one other point about QE. We touched on this, and that’s the large balance sheets that have emerged out of that, out of the pandemic QE. Other places as well, but the Fed in particular. Does it concern you that it’s going to limit the Fed’s ability to do QE in the future if there is another financial crisis, if there are other events that come up?

Muir: Yes, that’s a question: What’s the total capacity of this? There must be some. I don’t think we know exactly what that number is, but it doesn’t seem like you want to preserve that capacity relatively carefully, so that the next crisis, you’re again able to come in and intervene and help solve it. I’m not sure if we know where those limits exactly are or how much of that they’ve eaten up.

I do think it’s clear that they probably did go overboard during the COVID era, but how much that costs, I think is a super interesting question. I’m not sure that there’s a really clear way to get at that.

Beckworth: To be fair to the Fed, I think part of the challenge they face is the fact that we’re running primary deficits as far as the eye can see. At some point, there’s pressure upon them to keep interest rates under control in a particular range for the target, which means they’ve got to intervene in Treasury markets and there’s all this supply coming in.

I think Darrell Duffy’s been really good on this. He’s been like, man, the supply of debt coming into the market and the capacity for broker-dealers is shrunk because of Dodd-Frank and regulations. The Fed has to step in and intermediate that. In some sense, the Fed’s hands are tied, and we’ll have to use QE going forward. It’s not an enviable position to be in for them.

Muir: Yes, I agree with that, both from the fiscal side and from what they communicated to markets a bit. The last thing you want to do is have another crisis happen. Markets expect you to come in and then you don’t come in. That’s going to be a force that would make things worse as well. I agree that there’s probably a feedback of that both ways, in that the Fed doing a lot of QE being a force that’s lowering long-term yields, that’s effectively creating some fiscal capacity for the government as well.

A lot of that ends up potentially getting used. It’s tempting to use it. Would we have issued as much debt in the last few years as we have, had the long-term yields been a lot higher than they otherwise would have been, which our paper argues that would have been the case had QE not happened. That feedback cycle can go both ways a little bit.

Beckworth: Yes, definitely facing the fiscal pressures. It’s amazing to me, though, that we still don’t see Treasury yields elevated any more than they are. Ten-year Treasury, as the time of this recording is just over 4 percentage points. It’s amazing times to be alive. 

Shrinking the Fed’s Balance Sheet

Okay, one last thing, I promise this time. What about the future of the Fed’s balance sheet before the next time we have to QE? They have shrunk it. They’re going to start growing it, but they can’t adjust what they hold on the asset side. They’ve talked about eventually going to all Treasury bills. Do you have any sense of what’s a good mix for them to hold as assets going forward?

Muir: It’s hard to come up with the optimum mix, but all of our research is suggesting that that mix absolutely does matter. The effects on the long-term yields are going to be driven in large part by the total amount of duration risk that they’re going to have on the asset side. If they switch to just holding bills, I would expect you to see a big increase in long-term yields or duration-sensitive assets coming from that.

Optimal mix is hard to know exactly what the right thing is there, at least from my perspective. That’s probably beyond any of the research that I’ve done. That mixture is going to matter for term premia and the premium on duration risk that’s out there in the market. I’m sure that that’s something that they’re thinking about and aware of.

Beckworth: That is interesting. They have a lot to take into consideration as they do move forward on the trajectory of their balance sheet. Do they want higher 10-year Treasury yields? Do they want higher mortgage rates? This is not an easy question to have. 

Tyler, part of the conversation about QE is, okay, how do you unwind it or how do you recover once you’re out of the crisis and you don’t need QE anymore? There’s been this discussion about quantitative tightening. Is it the same thing as QE, but in reverse? There’s been a big debate on that. Where do you land?

Muir: Where I land on that is that QT is probably closer to QE in reverse than I would argue a lot of people think. Now, in practice, it’s not quite, but that’s often because QT is done very slowly and gradually. The reasons for that are in part because I think that the Fed doesn’t want to surprise markets when they’re doing QT.

At times when we’ve seen them tighten when they have surprised markets—I have the Taper Tantrum in mind—we did see a really big effect on yields right away when that was announced. I think you don’t see as much in practice, when QT is happening, a big response in yields. I think a lot of that comes from the Fed being very careful to signal it well in advance and to taper it and do it at a very slow pace that’s not going to spook markets.

The way that they actually operate it is quite different in QT where they might buy really aggressively quite quickly. Other than that, I think if they were to sell it, if they were to unwind really quickly and aggressively like they do on the buying side, I think it would have really big effects on yields or you would see really big effects on yields.

There’s other differences as well, and I don’t want to diminish those that other papers have focused on and looked at. From my perspective, at least, that’s a big, important one.

Beckworth: That’s interesting to hear. That term premium, that 75 basis points that you guys found, that’s what could be affected if you take QT completely to its limit in the other direction. We may not see it immediately, but it would happen over time.

Muir: Yes, that’s right.

Beckworth: One last thing on QT I will highlight, and this is something I’ve been thinking about a lot lately, is this ratchet effect that some people have talked about. This is more, I think, on the liability side, less about the asset structure itself. Bill Nelson, who’s been a podcast guest several times, used to work at the Fed, he argues there’s a role for supervisors as they look at all the liquidity that banks are holding and it’s hard for them to go back.

There’s another paper by Raghu Rajan and his co-authors that’s been really fascinating, at least for me. They make the argument that when the Fed does QE, it’s adding all of this liquidity backstop on the asset side that makes banks more comfortable to fund with deposits.

It’s easy to undo QE, at least from the Fed’s perspective, but it’s very hard from the commercial bank’s perspective to unwind all that liquidity because now, you’ve got to have a backstop for it and people don’t want to lose the money, the spending ability. That’s what really is the constraint in their thinking. That’s another issue, I think, that needs to be further considered by folks like you, and maybe the Fed as we go forward.

Muir: Yes, I completely agree. I’m a fan of that paper. I think, in part, the stuff in our experience with it is new. You get these effects of what’s going to happen in terms of unwinding on the asset side, but then there are all these important effects about what’s going on with reserves and what that’s going to do to the system. I think we’re just getting an understanding of that now. That paper helped me understand that issue a bit better.

Beckworth: All right. With that, our time is up. Our guest today has been Tyler Muir. Tyler, thank you so much for joining us and shedding light on QE.

Muir: Thank you so much for having me. This has been a lot of fun. I’m a big fan. Thanks for having me on.

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