Marc Giannoni on the Fed’s Framework Review, it’s Independence, and the Future of R-Star

Marc Giannoni is a managing director and the chief US economist at Barclays Capital. In Marc’s first appearance on the show he discusses working on the 2020 Fed Framework Review, the troubling issues of Fed independence and fiscal dominance, the future of long rates and r-star, his influential 2006 paper about what good monetary policy looks like, and much more. 

Subscribe to David's new Substack: Macroeconomic Policy Nexus

Read the full episode transcript:

This episode was recorded on August 28th, 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 Marc Giannoni. Marc is a managing director and the chief US economist at Barclays Capital. Marc is also a veteran of the Federal Reserve System and the Swiss National Bank. Marc joins us today to discuss the Fed’s framework review, Fed independence, the future of r-star, and more. Marc, welcome to the program.

Marc Giannoni: Thank you very much for having me, David.

Beckworth: Well, it’s great to have you on. Now, I think I first ran into you at a Hoover Monetary Policy Conference. You may not remember this, but I was sitting down on those round tables for a lunch or dinner, and you were there and you introduced yourself. I was like, “Ah, Marc Giannoni, I loved your 2006 REStat paper, ‘Has Monetary Policy Become More Effective?’”

We’ll get to that later in the program because that was one of my favorite papers in a long time and deals with identification and the challenges we face as macroeconomists. Such a clever and, I think, such an important paper when people are doing economic analysis. I met you there, and then I think I ran into you at the Jackson Hole 2019 conference. You were the research director. Tell us about your career, your journey. You were at the Federal Reserve. Now you’re in the private sector. Walk us through. You’re from Switzerland originally, I learned. Fascinating.

Marc’s Career

Gianonni: That’s true, yes. Very good. Again, thanks very much for having me, Dav. Pretty early when I started economics, and as you said, this was in Switzerland, way back when. I was really fascinated with monetary policy. I dreamed of working one day in a central bank. Actually, my very first job out of college was at the Swiss National Bank. I started out as a junior economist. I was tasked with covering basically Italy and France.

Right away, right after I started, I was thrown in the deep end, I guess. We had the European monetary system crisis shortly after I began. I had to make sense of that and report to the big bosses about that. That was quite an exciting time. I worked about five years at the Swiss National Bank and kept wanting to learn more about economics. That pushed me to the US eventually to pursue a PhD. I was fortunate enough to be able to go to Princeton University. That was actually an incredible time in macro, in monetary economics. I think the field there was undergoing a major shift.

I had the privilege to be working with some amazing advisers. I had Mike Woodford as a primary adviser and Ben Bernanke, Chris Sims, as well as being part of my committee. It was a great, great, very inspiring committee. I worked for several years as Mike Woodford’s research assistant also, while he was working on his book, Interest and Prices, which has since become a cornerstone, I think, in monetary economics. Really, really interesting times. I was really fortunate, I think, at that time.

Then continued to work along with Mike on several exciting research projects for many years, actually. I focused on optimal monetary policy. This was mostly technical work, but work that influenced a lot the way I think about monetary policy now to this day. That was my Princeton days. 

Then, after that, I joined, again, a central bank. I went to the New York Fed in the research team there, also just in time for the 2001 recession, which was also quite something, quite interesting. A couple of years later, I moved back into academia. I joined the faculty at Columbia Business School. I spent nearly a decade at Columbia between teaching and doing research, continuing this research project with Mike and with Jean Boivin, the paper that you were referring to before, but continued to work also closely with people at the New York Fed. Eventually, I returned to the New York Fed full time and focused on monetary policy issues, r-star issues, as you alluded to, and policy advice by briefing the president.

Then came a call from Rob Kaplan, who was president in the Dallas Fed. He asked me to join him and serve as director of research and chief policy adviser. Off I went to Dallas, basically, leading a pretty large research group, working closely on FOMC issues, and engaging also with the board of directors there and businesses across the district, which brought to me a very different, very new set of interesting issues to work on and people to talk with.

Now, five years later, I’ve found myself back in New York, this time in the private sector, as you alluded to. Joined Barclays as chief US economist, where I am today. There, I lead a US economics research team, and we focus on the US economy still, monetary policy, trying to share our insights with Salesforce, traders, clients, policymakers, the broader public.

Beckworth: Yes, I’ve seen you on TV a few times in that role, and I have a few of your research notes. I didn’t realize, Marc, that you were a part of what my former colleague Scott Sumner, he has this paper, he calls it the Princeton School of Macroeconomics. That late ’90s group, Ben Bernanke, Paul Krugman, Mike Woodford, I guess a fellow grad student at the time, Gauti Eggertsson, Sims, all these big names we’re thinking long and hard about. Oh, Lars Svensson as well. They’re all thinking long and hard about Japan and deflation. You got exposed. That’s pretty remarkable.

The other thing I want to highlight is, you went to the Dallas Fed, and I had a good friend there, Evan Koenig. You remember him. He’s a big champion of nominal GDP targeting. So am I, so we naturally connected. You’ve had quite the run. That’s really fascinating to see your career. You mentioned you were the research director at the Dallas Fed, and you were there during the lead-up to the 2020 framework review. You were part of that process. I want to get into the current one, but tell us about that one. What was your role? How did you see it unfold? Give us the backstory.

Fed’s Framework Review

Gianonni: I was called at the time to join in a so-called steering committee that involved four senior staff members across the system who coordinated the research effort from about 80 researchers across the Fed system that focused on that framework review. The other three members were, initially at least, Thomas Laubach, the late Thomas Laubach, a great friend of mine, head of monetary affairs at the Board at the time, who unfortunately left us way too early, Jeff Fuhrer at the time at the Boston Fed, and Dave Altig.

Together, we were part of the steering committee, thinking about how to structure the research program around that framework review and basically channeling all the work from all these wonderful researchers across the system, and then present it to the FOMC. For the good part of basically the second half of 2019, early 2020, we spent quite a bit of time briefing the FOMC at every FOMC meeting, presenting new results and moving forward the agenda, which then gave rise to that 2020 statement on longer-run goals and monetary policy strategy.

That was quite an interesting episode as well. Of course, that came, or at least toward the end of that process, we went through the COVID period, which brought a whole set of new challenges, but we were very much focused on that reform and that statement on longer-run goals and monetary policy strategy.

Beckworth: Do you think the 2020 framework, which is called FAIT, popular name is FAIT, Flexible Average Inflation Targeting, do you think it reflects the legacy of the Princeton School of Macro because it has a makeup policy term in it? The whole notion that if you go way below, you’re going to catch up. I think it’s important. Now, we don’t always need makeup policy, but if you have something like 2008, the Great Depression, I think makeup policy is essential. You guys were able to bring it in.

In fact, Rich Clarida, I think he framed this as a version of Bernanke’s temporary price-level targeting, and Bernanke, of course, is part of that Princeton School of Macro. How do you think about that 2020 framework?

Gianonni: I would say Rich Clarida was instrumental during the process. He was vice chair at the Fed, of course, and we were reporting directly to him. He was basically leading this whole effort at the time. Yes, I do think this FAIT feature was really important at the time. I still do think it’s important in general. I guess as we talk about the current reform, we can talk again about the importance or not of including FAIT as part of the policy strategy. I do think it’s basically an automatic stabilizer that helps the Fed provide accommodation when you need it and when the typical instruments like the funds rate are being constrained by the effective lower bound.

Beckworth: Let’s jump into the 2025 framework, where you just announced, Chair Powell announced it. Of course, it didn’t get a whole lot of attention because there’s other things going on now. I know I was closely watching that part of his speech. Everyone else was watching the first part of his speech about rate cuts and Fed independence issues. I think this is important because, like you said, makeup policy or level targeting or forward guidance, all those elements are essential if we get stuck in a world like we were in the 2010s. On the surface, at least, it looks like we went from FAIT to FIT, so Flexible Average Inflation Targeting to Flexible Inflation Targeting, although there’s some caveats, some nuance. How do you read the new framework?

Gianonni: So far, the FOMC has released an updated version of its statement on longer-run goals and monetary policy strategy recently at the Jackson Hole Conference. This is a foundational document that specifies the FOMC’s policy goals and its approach to achieving those goals. Namely, it’s Fed’s reaction function. In a sense, it provides the basis for the committee’s monetary policy decisions.

Now, the FOMC, as I understand it, is working on other aspects  of its framework review at present. I expect it to come up with enhancement to its communication policy sometime later this year. That would include, I suspect, enhancement to its summary of economic projections, potentially its stock plot, something like that in the fall. Now, in regards to the statement on longer-run goals and policy strategy, it basically came in largely as we had expected. Actually, the FOMC had given us some advance notice in their minutes earlier this year, and I think it made a few changes that were quite warranted.

For one, it no longer refers to the fact that neutral rates of interest have declined. This was a feature in 2020. That’s what we had seen in the decades preceding the 2020 review. Now, since then, there’s a lot of uncertainty about the path of rates, but they have likely not declined further. I think it’s appropriate that the FOMC remove that part from the statement.

They also deemphasized very much the concerns about hitting the effective lower bound on interest rates, given the recent experience with elevated interest rates and elevated inflation. That’s not a concern at present, although it could be a concern in the future again. We could end up again at some point at the zero lower bound or at the effective lower bound, and the Fed policy may want to have a strategy in place to handle that.

Now, as you said, the FOMC removed the FAIT strategy from its statement. I think it’s accurate to say that it returned to the FIT, Flexible Inflation Targeting, without the averaging part. It has some merits in the sense that it’s simpler. It’s easier to understand. They are looking to achieve 2% over time, so there’s no ambiguity about that. I do think they lost something by removing FAIT. It was, as you said, intended to provide more stimulus at the effective lower bound, in effect, by generating higher inflation expectations when inflation expectations were drifting down, when inflation was below the target, and interest rates were constrained by the zero lower bound.

By generating higher inflation expectation, you are basically lowering the real interest rate. You are providing more stimulus. That, in a way, creates an automatic stabilizer for the economy. I think it has a very desirable feature that you see in other strategies, like price level targeting, for instance, that helps stabilize the economy quite effectively, I think, when you need it and when you are constrained by the zero lower bound.

Now, in its place, the FOMC revised the statement but notes that the Fed is prepared to use its full range of tools, as they say, to achieve the goals at the zero lower bound. They don’t really specify what that means, but I suspect this refers to the balance sheet and maybe forward guidance to some extent when they are constrained at the zero lower bound. Nowadays, some disagreement about how effective the balance sheet is, if you don’t have a strategy that aligns with it, I think, that’s to be seen.

I personally do think it’s unfortunate that the FOMC dropped the FAIT part. I fully appreciate it’s not relevant now. It may not be relevant for a couple of years. If we do return to a low inflation, low interest rate environment, it’s a tool that could potentially provide the stimulus when you need it. The strategy has relinquished its tool at this point.

Beckworth: Yes. Like you, I noticed that they seemed to hedge their bets with the “full range of tools at the zero lower bound.” They also got rid of the shortfalls, but they didn’t go back to deviations. They talked about, “Well, there may be times where the economy can run above maximum employment in real time and not have price stability.” I feel like they’re hedging their bets. They’ve gone back to FIT, but they’re allowing elements of FAIT lurking in the background. In some ways, I would say this is a little more confusing.

Now, I know FAIT was confusing because it was makeup from below, not from above. I think that was a hard point to get across to the public. That, Marc, is why I think something like a nominal GDP level target is a whole lot easier to communicate. We don’t want to tighten policy in the face of supply shocks. We want to see through them. In my mind, at least, it’s a whole lot easier if you frame this in terms of nominal income growth or spending growth.

In any event, they’ve hedged their bets. They’ve got this new framework. I guess, Marc, if you could wave a magic wand and you were given the power to design a framework, what would your ideal framework look like?

Gianonni: I think it has a lot of elements of that ideal framework in the sense that they specify the inflation target, 2%. That’s pretty clear. We can debate the number. Should it be higher? Should it be lower? I think having a number brings a lot of benefits ultimately. It helps anchor inflation expectations. I think it gives a good sense of the strategy. I think they could be more specific on the number of points.

I talked about my preference for FAIT in general. If you want to have a framework that’s robust to a broad set of environment and not just the conditions that we’ve been over the last three years, so other configurations. I do think it brings that feature. You alluded to the shortfalls. I think this is another part that, yes, it got dropped from the statement. Exactly as you said, there are elements of it still there.

I think they suggest that the committee will partially return to its prior strategy of tightening policy more preemptively when the employment is perceived to be about its maximum level, but not quite, because they left this option to not tighten policy when the unemployment rate is low, if the FOMC judges that it’s not feeding undesirable price pressure, let’s say.

This is, as you said, I think somewhat confusing. Obviously, we don’t know, and nobody knows what maximum employment is. Certainly, we don’t know in real time what maximum employment is. It’s something we can estimate over time, but there’s a lot of uncertainty around that. Obviously, we need to take these estimates with a grain of salt, and there’s judgment that needs to be applied. I think what a good framework has is clarity about how you deal with that judgment and what you are trying to achieve. To me, saying that, “oh, it’s sometimes okay to let employment go above the maximum level of employment, or at least what we in real-time estimate to be the maximum level of employment,” is somewhat confusing.

They also talked about, or reemphasized, I would say, the importance of having anchored inflation expectation. Obviously, that’s very important. I think it’s good that they added that, but it is a little vague as well. We’ve seen, for instance, this year, inflation expectations move up. Some estimates of longer-run inflation expectation move up quite a lot. Others have not. The FOMC participants have consistently repeated that longer-run inflation expectation remains anchored. I think that creates a little bit of confusion as well, as to what exactly are we talking about when we say that inflation expectations are anchored.

Beckworth: Yes. A lot of interesting questions here in this framework. It will be interesting to see five years from now, when they revisit this, what happens, any changes to it. You mentioned the 2% inflation target being explicit is very useful. Maximum employment, we don’t know what it is in real time. Do you think it’s better to leave it undefined since we really don’t know what it is?

Gianonni: I think it’s appropriate that the FOMC, as it does in its statement, give us a number for inflation, 2%, but they don’t give us a number for a natural rate of unemployment or what they think of maximum employment, what the level of that is or what the trend of that is, just because of the considerable uncertainty around that. I think as part of their communication, they could give us, and this may come later, who knows, they could give us estimates of what they think it is.

We know from the SEP, let’s say, what they think the longer-run neutral rate of unemployment is. We don’t necessarily know what their estimate of the shorter-run neutral unemployment rate is, but they could give us more information about that, about how they judge where the economy is relative to what they are trying to achieve, and in what direction they have to move the economy with their policy instruments.

Beckworth: We’re talking about the framework review. I think both of us think it’s an important discussion, and maybe more attention should be given to it by the media. At the time that this speech was given by Chair Powell, there were other things brewing. We had questions about whether it would be rate cuts or not. Then the other big lurking question, of course, is Fed’s independence.

Fed Independence 

Marc, at some level, I wonder when we have these conversations about the framework, it assumes we have Fed independence so they can make these choices, redesign its framework, do things. There’s been a lot of questions under the current administration, whether that independence is being compromised. I don’t want to get into all the details. Listeners know what’s going on. I do want to frame it a little bit different than I think many people are framing it.

Former Fed Chair Bernanke and Yellen had an op-ed in The New York Times. I believe Yellen had one just recently in the FT as well. Their angle on it was very much what I would consider political independence or the norms. The president doesn’t bully or push the Fed. I think of several forms of independence, there’s that political independence. There’s legal independence. The Federal Reserve Act defines certain things. There’s also financial independence. The Fed has seigniorage. It can finance, it doesn’t depend…

The last thing I will call for, maybe there’s a better term, but economic independence. The Fed is able to pursue its 2% target without facing fiscal pressures, without being called upon to keep the US government solvent. I bring this up because, Marc, when I look at what’s been going on, why all this pressure on the Fed? Why is Trump going after the FOMC? It’s very different than 2019. 2019, it was about full employment, jobs. Trump has been very explicit. He wants to lower the interest cost on the debt. There’s a number of other things that have been going on. I’ve mentioned this before in the podcast, but I’ll just throw them out there, which, to me, paint a picture of fiscal dominance coming soon or on the horizon.

One would be, again, Trump’s call for rate cuts to help lower the interest cost on debt. Secondly, things like the supplemental leverage ratio tweak, that’s been motivated in part by big banks can put more Treasuries on their balance sheets. Stablecoins, I think stablecoins is an exciting development, but it, too, has been motivated in part by more demand for T-bills. Senator Ted Cruz, he called to end interest on reserves. I question his thinking on this in terms of what actually would happen. Again, he’s motivated by fiscal pressures.

Then, even, we see a lot more T-bills being issued by Treasury, which it’s more cost effective because we’re getting all these pressures. Even Treasury market reforms that are going on, they’re all there to keep the market resilient and robust. To me, they’re all symptoms of something deeper. Do you worry that maybe we’re missing the forest for the trees here? The real issue is we have a fiscal problem that it doesn’t matter who is Fed chair, who’s on the FOMC. At some point, they have to become subordinate to Treasury and Congress just to keep the government solvent.

Gianonni: Look, in terms of the Fed, they’ve been assigned these goals: maximum employment, price stability, actually also moderate long-term interest rate. That third goal is actually viewed generally as a consequence of the first two goals of price stability and maximum employment. That’s why we talk about the dual mandate. The Fed is organized in a fairly decentralized fashion to withstand basically external pressures and the pressures to move policy rates one way or another.

It was also granted a fair amount of autonomy and operational independence for the conduct of monetary policy. I do think operational independence is important here in the sense that, ultimately, we are accountable to the public, to the people, and the Fed is as well. It reports to Congress. In terms of the actual conduct of monetary policy, it has been, so far, granted operational independence.

I think this independence is critical. There is plenty of empirical evidence by looking across countries over time that this kind of central bank independence brings big benefits to society through price stability generally, through more stabilities in economic activity overall and in employment. I do think there can be legitimate disagreements at all times about the assessment of the economy, about its outlook, about the risks around the outlook.

Different policymakers can come up with different conclusions about the appropriate path of policy, given their assessment of the economy. It is, I think, paramount that the FOMC be allowed to do its job to set the policy instruments in order to achieve basically the congressionally mandated goals: price stability, maximum employment. You alluded to a lot of changes that are going on. I do think the independence is being tested. We’ll have to see where that brings us. Does it bring us to a system where we’ve tested the limits of the independence, and it’s a more well-defined, clearly resilient independence of the Fed, or is it not? We’ll have to see where that brings us.

You mentioned about the fiscal situation here and potentially fiscal dominance. I think clearly it would benefit the fiscal outlook in terms of budget balance and in terms of the debt-to-GDP ratio to see longer-term rates be lower in the sense that the interest costs on the debt would be lower, so the budget deficits would be reduced and the debt-to-GDP ratio would not be growing as rapidly.

As I mentioned before, I think this can happen in a system with low inflation and maximum employment. In fact, I think the sure way to get to longer interest rates that are moderate, that are relatively low, is by having relatively low inflation and maximum employment. I think if the Fed were to cut rates too much with the intention to boost growth to lower the financial cost, if that were the objective, it may ultimately end up creating more inflation and ultimately end up creating higher longer-term rates, which ironically doesn’t really help so much with the interest expenses. I do think a lot of people recognize that, and that ultimately, it’s an important consideration to maintain.

Long-Term Rates

Beckworth: Speaking of those long-term rates, if I look at them now, it’s the one thing that makes me question my worries about fiscal dominance. The 10-year Treasury still seems relatively low. If, in fact, fiscal dominance is on the horizon, why is it at 4.2%, 4.3%? I have to at least acknowledge that the wisdom of the markets disagrees with me for now, at least, that we’re not at the precipice of a fiscal crisis.

Let’s talk about long rates in general because you’ve done some really fascinating work on this. You have multiple papers. I want to focus on two of them. The first one is a Brookings paper on economic activity in 2017. It’s called “Safety, Liquidity, and the Natural Rate of Interest.” I believe your co-authors are Marco Del Negro, Domenico Giannone, and Andrea Tambalotti. You do a really fascinating discussion of why r-star, or the real rates, had fallen. This was focused on the US. Let me begin with a very basic question. I know a lot of listeners will know this, but just to set the table, how do you define r-star?

Gianonni: That’s a great question. That was really a fun set of projects with these co-authors, at the time colleagues, at the New York Fed. As you said, Marco Del Negro, Domenico, who is not my cousin—we share almost the same last name, except for the last letter—Giannone, and Andrea Tambalotti. We were trying to figure out why interest rates were so low at the time. This is around 2016, 2017 period, pre-pandemic, essentially, and what that was telling us about the future.

We essentially approached this from two angles. On the one hand, we built a flexible statistical model to extract long-run trends in bond yields, inflation expectations. On the other hand, we had a much more structural model, so-called DSGE model, that has lots of moving parts there, but lots of details, lots of data as well, used to estimate this model to try to capture underlying economic forces.

I think what was pretty interesting, we found basically by working on this project, is that both methods gave us a fairly consistent picture. As you know, in the world of economics, that’s not obvious that if you take two different routes to answer a certain question, you get to the similar results. In this case, we did. We found that the natural rate of interest, the way we thought about that, there are different definitions of that. You could think of that as being the equilibrium interest rate, so that monetary policy is neither accommodative nor restrictive, so it’s neutral. We can think of that as the interest rate, or the real interest rate, such as saving, broadly defined, and investment, are in balance.

We can define it through the longer-run trends model as being basically where the trend for the real interest rate is, given all the data that we have, or in the very structural DSGE model, we can think of really what would be the real interest rates that you would have if the economy were to clear very quickly, if you were not subject to price rigidity, wage rigidity, and other frictions like that, what would be the equilibrium interest rate that you would obtain directly.

For those who are very much into the weeds of econ, remember the good old Phillips curve and ISLM model, you can think of basically where the IS curve is crossing the full employment, essentially. There are many different ways to think about the equilibrium interest rate. We had different models to try to assess that in slightly different way, both through the trend model and the DSGE model.

Beckworth: Now, what did you find?

Gianonni: Exactly. First, we found that this equilibrium interest rate had been declining quite significantly from the late 1990s. The real rate we assessed to be at about 2%, 2.5% had been there for several decades, right through the mid-1990s. Then from the late ’90s through the prepandemic period, it basically had been trending down pretty significantly by about 100 to 150 basis points or so. Then we tried to figure out what these forces were.

A key force we identified there was that there was essentially a rising premium for holding safe and liquid assets like US Treasuries. Again, this is prepandemic period. Investors really valuing Treasuries, not just for their pecuniary returns, but also for the unique attributes they bring in terms of safety, in terms of liquidity. Since then, we’ve seen a lot of models trying to explain why Treasury had these very attractive features that let investors all around the world want to hold them.

We found that this so-called convenience yield had increased pretty sharply, leading to a lower rate in terms of Treasury rates than you would otherwise have, basically since the 1990s, and therefore explaining a lower r-star than you would otherwise have, were it not for that feature of this Treasury. That contributed, in our estimate, to a good percentage point of the drop in the equilibrium interest rate.

Second, we identified slower trend growth, both in productivity and consumption, during that period. We think that contributed as well to some of the decline, although less so. We had also demographic factors that played a role with the aging of society as well, contributing to the lower rate, but less so, and so on. The bottom line, I think that there were these deep structural shifts and this increased attractiveness of Treasuries at the time, from the late ’90s to just before the pandemic, basically contributing to more demand for Treasuries from around the world, and this lower equilibrium interest rate.

That’s what I think contributed as well to real interest rates being lower, but then monetary policy being potentially more constrained by the zero lower bound. It underscores, I think, the importance of fiscal policy as well, financial policies, given that the supply of safe and liquid asset is quite central where this equilibrium interest rate settles.

Beckworth: You noted that the decline in the equilibrium rate, the real rate, r-star, starts late ’90s. Basically, the 2000s up to 2020, which one could look at, oh, it’s globalization, it’s a global demand for safe assets, maybe it’s the aging of the planet. In your mind, what has fundamentally changed since then, why we won’t return, at least in near term, to a zero lower bound environment? What has altered that trajectory?

Gianonni: Look, I do think there is a clear distinction between short-term r-star and longer-term r-star. This is something that, let’s say, the DSGE model, for instance, that I was alluding to before, can help distinguish. The other, more statistical models, have more trouble distinguishing that. When I think about longer-run interest rate, I think about the factors I described: demographic factors, productivity growth.

Now, productivity growth is really hard to assess. If you’re an optimist, you’ll say, “Well, that’s maybe accelerating going forward, and who knows, maybe that’s going to contribute to higher r-star.” I’m hopeful about the effects of AI, but I don’t know how much of that is going to realize and how much is going to contribute to productivity just yet. The demographic factors, this is something we know that’s contributing to lower equilibrium interest rate. I think it’s going to likely continue for some time.

The part that we identify as being quite important, this convenience yield part, arguably, there’s a little more uncertainty about that at this point, given the fiscal outlook we discussed, given the fiscal trajectory. I guess there is some uncertainty about the trajectory going forward, but I think it’s going to be an important factor to think about when we think about the evolution of the longer-term r-star.

Then short-term r-star, I think of that as more supply, demand imbalances, imbalances between saving and investment in the shorter run. What we’ve seen in the last few years, since the pandemic, initially during the pandemic period, a massive shift of resources from the public sector to the private sector, which boosted consumer spending tremendously through the checks and various other stimulus programs. Also, an increase in investment afterwards and the need for investment, whether it’s the energy transition or now AI investment or whatnot.

These forces that are basically really increasing the demand for investment at the same time as the saving from household being relatively muted, as households feel wealthy, having gotten these checks and went to spend them. We had, for several years, very robust consumer spending. That, in equilibrium, basically leads to a higher, shorter-term r-star. The way I think about it is that in the short run, r-star is currently quite elevated. If you look at the New York Fed’s DSGE website, it’s going to tell you that in the short run, r-star is probably in real term around 2% now.

It’s projected to gradually come down to something like 1.5% or 1% over the coming year. To me, that makes quite a bit of sense to think about the trajectory of r-star. We’ll eventually converge to a lower r-star again. We recently went through a period of fairly elevated r-star.

Beckworth: That’s very helpful. It really answers my question. Did the world fundamentally change in 2020, other than a whole lot of debt that was added? What you said is, no, it hasn’t. Population dynamics are the same. If you think of a basic Ramsey growth model or economic growth model, productivity growth, if it goes up, it raises the r-star because the return on capital goes up. If population growth goes up, if the return on capital also goes up, it’s positive as well. Those two things, really, as you said, haven’t changed.

We know the one for sure, at least, it’s going to get worse. Population growth is going down. That would suggest lower r-star. Maybe you cross your fingers, AI will change the world. That was the context of my question. What really fundamentally changed? You’re suggesting, maybe things haven’t changed as much as the headline numbers show us right now.

Gianonni: Yes. Again, that distinguishing longer term and shorter term, is important. I think the longer-run drivers maybe have not as much yet. We’ll see what AI does. We’ll know. Maybe that’s where I think I agree with Chair Powell. We’ll see by its works where r-star is at the end. We’ll probably see that. It’s very hard to predict that part. I think short-term r-star, in my mind, if I look at models, if I look at structural models that take information from investment data, equipment data on investment, on productivity, on consumer spending, on employment, so on and so forth, and rates, these models, like the New York Fed DSGE models, for instance, would point out that short-term r-star is probably quite elevated, and has been for the last three years.

Beckworth: Let me ask you a really speculative question about interest rates here. Let’s just imagine, for the sake of argument, that AI does take off. In fact, rapid growth. We have AGI, artificial general intelligence. For the sake of argument, again, this is extreme, let’s say we get trend real GDP growth of 10%. Really extreme. That would suggest real rates would go up quite a bit. Now, that would be a great problem to have because the amount of income growth would be fantastic. It would be a wonderful world to be in.

However, if you were Treasury secretary and you knew that growth was going to accelerate and real rates would go up, would you be worried about the debt more or less? Because on one hand, you might say, “Man, we’re going to have to finance with higher real rates.” On the other hand, there’s a bigger tax base to draw income from. Can you help us think through wild scenarios like that?

Gianonni: I haven’t run that exercise. Just thinking like that, I think it’s a great problem to have if productivity growth brings you to a 10% real GDP growth. I think you’re going to bring a lot of tax revenues, and people are going to be very happy to pay these tax revenues. You are going to improve the fiscal situation a lot, even if the interest cost on the debt is elevated. We saw some of that to a much lesser extent, of course, in the late 1990s, early 2000 years, where for a while, we had Treasury surpluses and federal government surpluses. 

This is the time where, by the way, I joined the New York Fed. One of the tasks here was to figure out, how do you conduct monetary policy in a world where there is no more public debt? That was the big question of the time because we thought at the time that maybe the Treasury supply would come down so much that you wouldn’t be able to use government debts to conduct monetary policy. Times have changed, but that gives you just a sense of how, with rapid growth of the late 1990s, there were quite a bit of tax revenues coming in.

Beckworth: I remember some of the discussions from back then, and talking to Bill Nelson, who was at the Fed at the time, he mentioned they were talking a lot about doing more discount window, lending more assets from the banks to replace the Treasury. Very fascinating. All right. Let’s go briefly to your other paper that I wanted to touch on. Same set of authors, I believe. This is “Global Trends in Interest Rates.” This is Journal of International Economics 2019. Maybe summarize that for us.

Gianonni: What we did here was to use a similar structure in terms of this trends model and now looking across different countries, US with a range of other European countries, trying to figure out what were the trends in interest rates and account more specifically for the linkages between these different trajectories for interest rates. Unsurprisingly, we found similar evidence to even incorporating data from all of these other countries and having a structure that allows for the dollar to play a role, the convenience yield in different regions to play a role, interest rate parity conditions.

Thinking hard about those relationships and how to model those relationships in that setup. Found essentially similar results that longer-term rates had moved down across the board, across all these different regions, and was constraining basically the path of policy by the end of this period of the 2010s, basically before the pandemic.

Beckworth: We’ll provide links to both of these papers in the transcripts. Just one comment about this global paper. You have a chart in there. It’s so interesting. It shows the global rates before and after this inflection point, where they’re all over the map, very different, but then there’s this sharp convergence. They’re not just all going down, but they’re converging, which is super fascinating, a lot of work.

Has Monetary Policy Become More Effective?

Now, we are running low on time. Marc, I’m going to jump to a topic I’ve been dying to chat with you about. This is your REStat, your Review of Economics and Statistics paper of 2006. The title is, “Has Monetary Policy Become More Effective?” As I was preparing for the show, I found an earlier version of this, which is interesting, the New York Fed version. It was like, “Has Monetary Policy Become Less Powerful?” I see what you’re doing there, but maybe walk us through this paper, and then we can talk about it some more.

Gianonni: This is a set of projects with a great friend of mine, former colleague as well, Jean Boivin. We were in grad school together. We were at Columbia together for some time, and now he’s at BlackRock. We were basically motivated by a few puzzles here. One is something that we saw. First, the economy looked a lot calmer after 1980s through this whole great moderation period than it had in the prior decades in the ’60s and ’70s. If you did this so-called impulse response functions, trying to figure out if the Fed raises interest rates, how much of an effect does it have on output and inflation, it appeared that after the ’80s, basically a 25-basis point increase in the Fed funds rate did not seem to have much of an effect anymore on inflation and output. A lot of people were—this is early 2000s—talking about that, suggesting that, “Oh, monetary policy is no longer effective. Look, the proof is in the pudding. You just raise the interest rate and look at what happens. Inflation doesn’t change, output doesn’t change, so monetary policy is therefore not effective on the economy.”

We were like, “Hold on, wait a second. Is it really that the economy has changed to the point that it becomes less sensitive to, let’s say, policy changes? Or is it that actually, policy is doing exactly what it’s intended to do, and it’s more effective at actually stabilizing inflation?” If you think about it, if the Fed is really successful at stabilizing inflation, let’s say at 2% all the time, whenever there is a shock, the Fed is going to change policy rates in order to bring inflation at 2%.

What you should observe in the end is the interest rates moving around a lot whenever there is a shock, but if it’s really successful, inflation should not move. It should stay at zero. You should find that the correlation between the interest rates movement and ultimately the goal variable, inflation, falls to zero. Then the question is, when you do find that correlation is zero, is it that the Fed is successful, or is it that inflation is really not responding to interest rate movements?

To sort that out, we used a small structured model, doesn’t seem very sophisticated by today’s standards, but at the time, it looked pretty innovative. Then run that through the machine and basically say, “Okay, let’s shock the system. Let’s now change structure of the economy, like we estimate before the 1980s, after the 1980s, see if that plays a role, or change the policy the way we estimated how it was before the 1980s and after the 1980s.”

We concluded, sure enough, that a large part of that change, or this so-called reduction in impulse response functions, was due to a more appropriate policy or a policy more effectively stabilizing basically policy rates. A pretty successful policy. This was, in our view, suggesting that first, you cannot just look at correlations. I think that’s always an important lesson, but second, that policy had done a pretty good job at stabilizing the economy. It’s important that it continues to focus on these goals.

Beckworth: Marc, in grad school, I did those vector autoregressions. One of the exercises was to show, pre-Paul Volcker, post-Paul Volcker, like, “Man, it’s a world of difference. The Fed does matter.” Most people intuitively know the Fed does matter at some point. I know there’s rational expectation models, a lot of things you could throw at it, but why isn’t it happening? I read your paper. It was so obvious after your work came out. It’s identification. One of the key things we deal with in macroeconomics, which is different than micro, is it’s really hard to identify exogenous changes that aren’t influenced by the system itself. That’s why we do these general equilibrium models.

We have to do structural models because we don’t know. In fact, I don’t know if you can see behind me here. I have all these little globes. They represent all the counterfactual worlds that macroeconomists have to consider if we’re thinking through all these things. It’s such a powerful insight that the instrument and the outcome are not going to be correlated if the institution, the Fed, is doing its job. It’s so intuitive, but oftentimes in casual analysis or on the news or on a blog, people say, “Oh, look, the Fed is not powerful anymore.”

I just want to throw out a couple of examples and see if you think this fits the billing. Let’s talk about the Phillips curve. Now, I know there’s many reasons why the Phillips curve was flat in the 2010s. Some would say it’s because inflation expectations were anchored. The Phillips curve, at least the reduced form, one you would estimate with data, would try to explain inflation by expected inflation term plus an output gap plus some kind of cost shock or supply shock term. That slope on the output gap got really flat or really small. One explanation could be, “Maybe it’s simply because the Fed is doing such a good job with inflation. It’s so stable, you’re not going to find a relationship between the output gap and inflation.” Would that be an example that fits your story?

Gianonni: 100%. I’m so glad you bring this up, actually, because that actually ties into another paper we worked on that talks exactly about that. This is with, again, Marco Del Negro from the New York Fed and another great co-author, Frank Schorfheide from the University of Pennsylvania. We were looking at, after the Global Financial Crisis, remember, obviously, we had the surge in unemployment, this collapse in output, but inflation didn’t fall very much.

At the time, there was a lot of discussion about, “Oh, they’re missing disinflation. Why is inflation not falling?” So much was written about how bad all this Phillips curve, where the Phillips curve was dead, made no sense to think about inflation depending on slack in the economy. There was a ton of literature about that. Again, coming back to that earlier insight, this didn’t make any sense to me in the sense that if you think about two equations working together, one is the Phillips curve, and the other one is a policy, basically, that tries to provide stimulus to the economy, and maybe some equation that relates activity, if you want, or slack, to interest rates.

If you are in a world with a forward-looking Phillips curve, what the New Keynesian literature would suggest, in this world, you can have a lot of slack now, but inflation does not need to necessarily fall at all if you think that you are not going to have much slack in the future. Why would businesses basically cut prices a lot today if the unemployment rate is high today, if they think that in the future, they will have to raise prices again because you are not going to have slack in the future, and it’s going to take time, and there’s some rigidity and some slack, and it takes time to adjust these prices?

Bottom line, in these forward-looking Phillips curves, as long as you had the expectation that you were expecting the economy to rebound at some point in the future, that the central bank was providing enough stimulus for that to happen, then it was not clear at all that you should be cutting inflation or cutting prices a lot. We put that to the test again in a model, New Keynesian model, Phillips curve with a fairly steep, actually, structural slope of the Phillips curve, shocking the system with a shock like the financial crisis, and trying to see how much inflation is actually falling in response to that.

Lo and behold, inflation did not need to fall at all very much as long as the monetary policy response was there to respond to the shock and promise to basically create more activity in the future.

Beckworth: The key insight or finding or point is that if the Fed is doing its job systematically, you should not find any relationship between your instrument and then your outcome, inflation, even real GDP. Let me take that insight and apply it to critiques of the money supply. Now, I’m not championing targeting money supply, but I do think it adds insight to why did the relationship between money and nominal income or nominal GDP break down? I think the answer it’s the same answer we just talked about. When the Fed’s not doing its job, that’s when you see the correlation. That’s when you see it.

If you go and look at growth rates of money and inflation or nominal income during, say, the ’60s and ’70s, it’s pretty strong. Same thing, post-’80s, there’s nothing there. People would say, “Oh, there’s been a breakdown in the relationship.” Again, I’m not advocating the targeting of money, but there is probably still a structural relationship. We’re just not seeing it in the data.

In fact, I would argue that what we saw in 2021, where money did explode, that’s the case—again, I want to be gracious to the Fed here because it was unusual. It was a big pandemic. It wasn’t a typical mistake, per se, but you could say the typical approach the Fed took broke down. We could see high money, high inflation, and such take place. To me, that’s the better explanation for why we don’t see a relationship between money and the economy today. Any thoughts?

Gianonni: I’m so glad you make these points. I wholeheartedly agree with that. I guess Charles Goodhart was making that point, the Goodhart’s law, emphasizing, again, if you try to stabilize something, if you’re successful at it, you’ll stabilize that thing. The instrument that you use to stabilize it may not be correlated with the outcome.

Beckworth: Yes. Let me throw one other example in because I got to do this example. It’s Milton Friedman. He had an analogy called the thermostat analogy, like you have an air conditioning system in your home. That thermostat is constantly operating to keep the temperature at a certain level. You set it at 72 degrees in your house. That AC, it’s going on, it’s going off. It’s responding to shocks from the outside. There is no correlation between what the thermostat is doing and the temperature in the room. Man, that thermostat’s been working really hard for you all day. It’s instrument, has been going back and forth. That’s a powerful analogy to what the Fed’s doing.

Gianonni: Beautiful. Yes, totally agree with that. I love it.

Beckworth: Marc, this has been a fantastic conversation. It’s been a real treat to get you on. I appreciate you coming on. Our guest today has been Marc Giannoni. Marc, it’s good to have you on the program.

Gianonni: Thank you very much again.

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.