Episode 30. Peter Ireland (Boston College Economics Professor) on Monetary Policy, Monetarism and New Keynesian Models

Podcast Interview Transcript

Peter Ireland (Boston College Economics Professor) joins the podcast to discuss his career as a monetary economist, his views on the history of monetarism, New Keynesian models, and the Shadow Open Market Committee which Peter sits on and celebrates its 50^th anniversary. (Listen to the full Capitalism and Freedom in the 21st Century Podcast episode here at the New Books Network which can also be found on your favorite podcast platform).

Jon: “This is the Capitalism and Freedom in the 21st Century podcast where we talk about economics, markets, and public policy. I'm Jon Hartley, your host. Today, I'm joined by Peter Ireland, who is a professor of economics at the Boston College Department of Economics. Peter is also a leading macroeconomist and is a member of the Shadow Open Markets Committee. Peter is also a leading monetary economist on central banking and has been an advisor to many regional Federal Reserve banks as well as the Bank of Canada. Today, we're going to talk about long-term trends in macroeconomics, including the rise and fall of monetarism, the popularity of New Keynesian models at central banks today, empirical macroeconomics, and the history of the Shadow Open Markets Committee. Welcome, Peter.”

Peter: “Hi, Jon. Thanks for having me here.”

Jon: “Peter, I want to get into your background. You were a University of Chicago undergrad and a PhD student in economics in the late 1980s and early 1990s, where you are Rob Townsend student. Where did you go up and how did you first get into economics?”

Peter: “Yeah, thanks for asking, Jon, and thanks for having me on the show, by the way. I grew up around here where I am now in the suburbs of Boston. It was never a real explicit goal of mine to come back home, but it is nice to be back in the same place that I grew up. But just to address your question, at Boston College here, the Jesuits have this thing they call vocational discernment to help undergraduates find, you know, their right career and lifetime choices. And the nature of the exercise they put students through is they asked them to consider first, potentially disjoint subsets, you know, A, what things interest you, B, what are you good at, and then C what activities can you make a decent living at. And then they challenge the students to try and bring those three sets together, like a Venn diagram, and find the intersections. So, it ensures that they find the career path and the lifetime path that, you know, helps them live well while at the same time having something they can also make a living at in the economic sense. I really like that idea and it resonated with me because I sort of stumbled on the same thing more or less on my own. When I got to the University of Chicago, I had several potential fields in mind. I really liked math. I really liked history and I really liked economics. So, I took courses in all three of those fields. And what I realized is I like math but doing innovative research pushing out the frontiers in math as opposed to just learning what other people have done. That just requires a talent and a skill that I just don't have. So, I realized early on, I'm not going to play in the National Basketball Association, and then I also realized I'm not going to be a professional or academic mathematician. History too, I'm interested in history. I like leading history, but I'm no good at memorizing names and dates, and I get confused whenever I talk about them. Economics is a teenager, I liked economics. I was a teenager in the 1980s at the start of the bull market and was really interested in the stock market too. And when I started taking courses in economics, I realized that was sort of at the intersection of that Venn diagram. Yeah, I liked doing it, I was pretty good at it. And thanks a lot. make decent livings both in the academic world and in industry doing economics. So, you know, that's how I gravitated towards the field. And then once I was there, I mean, University of Chicago was just a really exciting place at that time.”

Jon: “Well, it's amazing. I'm sure there are a lot of University of Chicago, Chicago School luminaries who were still there at that time. I know Friedman moved to the Hoover Institution in 1978. I'm sure many like George Stigler, Gary Becker, Harry Johnson, and Rob Townsend were still there. Amazing that you were part of that tradition, both as an undergrad and as a graduate student. Now, you joined the staff of the Richmond Fed in the early 1990s. At that time, I know it was a very well-known regional Fed department in terms of the staff where a lot of people went on to have very big careers in monetary economics and macroeconomics. Then you moved to Rutgers briefly and then for a couple of years and then to Boston College in 1998 where you've been since. I'm curious, how would you describe your overall research agenda, your main ideas, and your macroeconomic views broadly?”

Peter: “Yeah, thanks for asking about that too, Jon. That again, I think it wasn't planned, but it went along very much the same lines in this gradual process, as I said, for vocational discernment. You mentioned before the chair of my PhD thesis committee was Robert Townsend, and in fact, my PhD thesis contained in it a very Townsend-esque monetary model where patterns of trade arose endogenously reflected traders' optimal responses to overcoming frictions in the underlying economic environment. So, some transactions took place with credit in the event that credit relationships could be sustained. In other instances, the absence of commitment or enduring relationships necessitated the use of money. Today, that modeling spirit lives on in the branch of monetary economics that uses search and matching theory, so Kiyotaki, Lagos, and so on and so forth. Again, I really have a great deal of admiration for Townsend's models, for the Kiyotaki, Lagos style models as well. Doing good work along those lines, however, I think required a set of skills and a set of creativity. It just wasn't my forte. And I was really fortunate to get my first job at the Richmond Fed, which was really like a kind of second round of professional training for me. At the Richmond Fed at that time, you know, Marvin Goodfriend was there, Mike Dotsey was there, Bob Hetzel was there, also visiting scholars like Bob King and Bennett McCallum. And when I arrived and got to know all of those guys, I saw how they were able to arbitrage, is the word I would use, between policy issues and economic issues that we as academics, as PhD economists, could address with the help of our, you know, analytic models both empirical. I mean, Marvin Goodfriend is probably the leading example of somebody who made his career out of taking models and methods from the academic world and applying them specifically to policy problems at the Fed and other central banks around the world. Today, there are a lot more people that do that, you know, you're a PhD student in macro and on the market, Fed Reserve jobs are really attractive options, partly for exactly that reason. You get to apply the models to exciting real-world problems. At the time, that was less obvious. If you think back to the early 1990s, the models and use, you know, in those days, the real business cycle model, RBC models have important lessons to tell us for how monetary policy should be conducted. But on the other hand, those policy lessons operate at a kind of higher level than the sort of detailed answers to specific questions that, you know, a Fed Reserve Bank president might ask. That's unruly to an FOMC meeting. So, I really enjoyed and discovered that I was good at listening to policymakers talk about the issues that were preoccupying them and concerning them at the time, and then going back and asking myself, "Okay, so how can I use what I learned as a PhD economist to address those questions?" And, I'll leave it to others to decide how fascinating my research turned out to be. What I do, what I wouldn't argue though, is that I think my research turned out to be quite a bit better and quite a bit more interesting. Doing that, then it would have, if I had stuck to, you know, what I would call pure monetary theory.”

Jon: “That's absolutely fascinating. And we're gonna talk a little bit more about monetarism. But I'm just going to talk about regional Feds and the dynamics between the staff and the president at various regional Feds or central banks in general. In your mind, and you interact with a lot of central bankers through your activities at the Shadow Open Markets Committee, where you often have a regional Fed bank president speak. In your mind, do regional Fed bank presidents and central bankers actually pay attention to models in coming up with their policy recommendations? Do you think that the views of the median central banker are actually influenced by the output of the house DSGE model? Or are these things nice to have things that people have and at best keep maybe keep at the back of their mind? At some level, I think VARs still do a better job of forecasting than DSGE models. I had Larry Summers recently on the podcast. We talked a bit about the underwhelming legacy of DSGE models. And Larry Summers had a long exchange with Prescott in the 1980s about this, about what should be the role of models versus data. And I'm just curious, given that you've seen a lot of interaction between staff and central bankers. And maybe the Richmond Fed's kind of an outlier here, but like my sense is that this is the case (the decline of models), especially in many of these cases where you have people coming from industry rather than academia, which seems to be on the recent trend with central bankers, you know, both Jerome Powell and Christine Lagarde are lawyers, for example. In your mind, are models like DSGE models still having a lot of influence and policy decisions made at central banks?”

Peter: “Yeah, I think so. I would disagree with Larry Summers on this point. First, let me just say, I mean, you're right, central bankers come from different backgrounds, and depending on those backgrounds, they're going to be more or less interested in hearing about, you know, the academic details of the work that their policy advice do. So, you know, somebody from a banking background or a financial markets background might not want to hear about the details of your latest DSGE model, but others, you know, Jim Bullard maybe or Loretta Mester, you know, might indeed be, you know, who come from a more academic background might indeed be interested in the details of the model. And John Williams, for example, at the New York Fed is a contributor himself to the New Keynesian DSGE literature with his work on price level targeting and the zero lower interest rate bound. So, different personalities within the FOMC will have different interests and different willingness and abilities to, you know, hear directly about the details of academic style work. But stepping back and asking about the contributions of, you know, DSGE models, you know, whether the models are referred to explicitly or not in the policy conversations. I mean, take for example, the New Keynesian divine coincidence, which describes a particular set of circumstances under which by stabilizing the price level or the inflation rate first, the Fed Reserve can also achieve stabilization of the output gap. So, when you take a look at the FOMC's 2012 strategy statement which formalized the Fed's flexible inflation targeting strategy, I mean, right in there was the idea that in many instances, the two sides of the Fed's dual mandate are complementary, not computerized goals. I mean, that is reflected, and it comes out of the same academic tradition that gives us the New Keynesian divine coincidence. Just another example of this, because you mentioned Ed Prescott at the board in the 1980s and the late 1980s, you know, the so-called FRBUS model, the Federal Reserve Board econometric model, underwent heavy revisions. Most of those revisions, my understanding is, were directed towards providing a much more detailed description of the model's supply side and a much more detailed description of disturbances that could generate stochastic fluctuation or random fluctuations in what we would call the natural rate of growth, quite apart from what monetary policy and other policy-oriented initiatives directed at aggregate demand we're doing. So, for sure, in my view, work on real business cycle theory, emphasizing the supply side, work in new Keynesian economics giving us the divine coincidence. I think that's actually had a tremendous influence on central banking, not just in theory, but in practice. Now, one more thing that you mentioned along the way was sort of some hint at the notion that maybe in many central banks around the world, they're kind of backing off from... what used to be a stronger academic orientation. I've noticed that as well. And I think that that's a bit of a mistake. I agree that it helps to have on a policymaking committee, a diverse set of people with backgrounds in banking, in finance, you know, in other areas in economics and in politics but I also think it really helps to have academics like Bernanke or like Jim Bullard or like Loretta or John Williams who are more capable of bringing academic insights to the attention of students committee and to bear on the important issues that monetary policy committees address. So, this goes both ways. A PhD in economics doesn't guarantee that you'll be a good policymaker and it's not the only skill that would make somebody into a talented central banker. But I do think that it helps to have PhD economists not just on research staffs, but on policy-making committees. And, you know, I hope that central banks around the world will kind of move back in that direction in years to come.”

Jon: “Yeah, it's fascinating. And part of me wonders what degree central bankers are maybe increasingly relying on like empirical macroeconomics or getting back to it, since the 1980s, the... rational expectations revolution, [macroeconomics] became very, very theory dependent. But I feel like we've gotten away from the good old sorts of reduced form regressions of things like prices on money or good old regressions of thinking about the Phillips curve. Obviously, I mean, the Phillips curve has really broken down a lot; that relationship empirically at least has broken down a lot. And part of me wonders, I guess part of the question too, is like, to what degree can you continue to trust a three equation New Keynesian DSGE model if it's the case that when inflation has been falling the past couple of years, but unemployment hasn't been going up as the model would suggest. And I know there's some people like Nakamura-Steinsson who say the Phillips curve has been flat recently until I guess just recently [until the pandemic]. Some would say that the Phillips curve is nonlinear and that all these recent fluctuations are evidence of this, that inflation falls when unemployment rises, but inflation doesn't necessarily go up when unemployment falls. So, it's sort of asymmetric in that sense. I think that's sort of the view of maybe Gauti Eggertsson and others. I mean, do you have any sort of thoughts on just empirical macroeconomics in general and where that's been?”

Peter: “Yeah. Well, your questions seem to touch on two separate issues. You know, one is just the role of macroeconomics and in particular, empirical economics and informing monetary policy decisions today. And actually, I think there's a lot of exciting work along numerous dimensions being done right now along the way. I mean, the work on estimated New Keynesian DSGE models continues certainly compared to the previous generation of real business cycle models. New Keynesian DSGE models allow us to come to address in a much more direct way the issues that are of principle concern to monetary policymakers on a meeting by meeting basis, heterogeneous agent models or models with, you know, elaborate network structures, let us get at issues that the representative agent model never could, you know, having to do with wealth effects on consumption, having to do with issues concerning supply chain disruptions, which, for obvious reasons have been of paramount concern going back to 2020. You mentioned Nakamura and Steinsson, I mean, they've been at the vanguard in moving away from pure time series analysis and looking at disaggregated data. I think that's another source, a rich source of potential information on how the economy works and the effects that monetary policy is having on the economy, I really applaud that.”

Jon: “I think an identification, too, that's, I guess, a huge part of it. This has, I guess, been the criticism of applied microeconomists for the past 20, 30 years of macro, and that you don't have that plausibly exogenous variation. And I think that's kind of the bit that part of this new wave and empirical macroeconomics won't say. sort of, I think, small in comparison to all the theory work that's being done, is maybe a growing and increasing the influential one potentially.”

Peter: “For sure. And the other line of work that I find quite exciting and wanted to mention is work with what, for better or worse, I'll just call big data. So textual analysis, this analysis of trying to gauge trends and economic activity by looking at data from websites or internet searches. Again, this is a source of information that macroeconomists and policymakers have never had access to before. And that can be very, very, very helpful. Now, you also, in your question, put special emphasis on the Phillips curve and Phillips curve instability. And that is a big problem right now. Embedded into mainstream New Keynesian models is a specific view of the transmission mechanism that relies importantly on the Phillips curve. In times of Phillips Curve instability, those models, it's a challenge to apply them and understanding what is actually happening. Whereas I do find that there is a lot of exciting work being done in macro empirical and theoretical work these days along numerous dimensions, one thing I wish there were more of, and maybe we can get into this later. In the conversation is, you know, other views of the monetary transmission mechanism, besides those that have to rely heavily on the Phillips curve because clearly Phillips curve instability is a problem right now.”

Jon: “Yes, absolutely. And [the Phillips Curve] such a, you know, a fundamental, you know, part of the New Keynesian DSGE model in all derivatives. I want to sort of just pivot to the sort of the other sort of type or brand of macroeconomics that you've worked on and in part represent in sitting on the Shadow Open Market Committee (which we'll talk a little bit more about that the history of later) because it in and of itself has its own historical importance that you know, it's 50 years been around for 50 years and was it started by Allan Meltzer and Karl Bruner, who are, you know, themselves very famous monetary economists. I just wanted to ask you what is monetarism, precisely? You know, we have this famous adage that inflation is always and everywhere a monetary phenomenon, you know, with long variable legs is, I think, how the Friedman quote goes, we have this equation of exchange, mv equals pq, or that money times velocity has to equal price of all goods times value of real output." That's an identity that's something that's always true. But how each of these aggregates sort of move in relation to each other, I think, is the core of the idea around monetarism, which is and monetarism sort of says that if we increase the money supply M, we can temporarily increase real output Q or if velocity V doesn't drop or prices don't increase immediately such that it completely offsets the increase in money supply and that money isn't necessarily neutral in the short run that is you know M increase in M isn't totally offset by some increase in prices or a drop in velocity. Hence, there's this argument that we should be using money supply as a policy instrument. Is all this a fair description of what monetarism says or am I missing something there in that sort of statement about what monetarism is?”

Peter: “Yeah, this gets back actually to the point I was making at the end. And when you see monetarism, a school of thought within economics, we can talk a little bit more about that perhaps later on. I think it's a bit broader and a bit more nuanced, let's say, than the issues that your question specifically refers to. When you talk, though, about, you know, MV equals PY and monetary policy actions being reflected first and foremost in movements in the money stock. And I think what you're getting at really is, you know, more specifically the monetarist view of the monetary transmission mechanism. So, to explain a little bit more, let me just start by saying, you know, the monetarist tradition, the Keynesian tradition, these are long and rich intellectual traditions in macroeconomics to try and summarize either, let alone both in just a, you know, a minute and a or a minute and a half. I mean, you're never going to be able to do that to anyone's satisfaction. But, you know, if you'll indulge me on this point, you know, if you think about what I would call the Keynesian view of the monetary transmission mechanism. This is the view that's embedded into New Keynesian dynamic stochastic general equilibrium models. The view is the Fed conducts monetary policy by managing interest rates, nominal interest rates, because of short run rigidities in nominal wages and prices. When the Fed manipulates nominal rates, that translates into sympathetic movements in real interest rates, which then induce households and firms to intertemporally rearrange their spending patterns over time. And then as they do so, that impacts on today's rates of aggregate resource utilization, the output gap or the unemployment rate. And then finally working through a Phillips curve, those movements in resource utilization generate movements and inflation. So, taking a step back when you ask what are the key elements of the Keynesian view of the transmission mechanism, it's the idea that policy movements start, policy actions start with movements in interest rates, and then ultimately transmit their effects to the economy working through the Phillips curve. But the monetarist story is different in each of those two ways. Instead of focusing on interest rates, the monetarist view would say the Federal Reserve conducts monetary policy by exploiting its monopoly supply over base money. That is by conducting open market operations that either increase or decrease the supply of current debt as well as bank reserves. Then the story goes on to describe how those changes in base money ultimately give rise to changes in broader monetary aggregates like M2, the kind of M that you would include in an empirical implementation of the equation of exchange. And then, you know, so long as... reasonably stable, those changes in the nominal money stock translate into change as a nominal income, P times Y. The monetarist description of the transmission mechanism admits in a sense that economists still don't have reliable ways of exchange or understanding how changes in nominal spending break down into real and nominal components in the short versus the long run. But that was kind of the point that I was trying to make at the end of my answer to your previous question. I think what is needed in macroeconomics these days is what I wish there was more of is work that helped us understand that, you know, last part of the puzzle, you know, it could be coming from a monetarist tradition, or it could even come from, you know, the Keynesian tradition, but trying to cope with the fact that, you know, any Phillips curve that anyone has ever written down, you know, ultimately, or ultimately encounters challenges in confronting any given data set. But sure, the way that I would describe the monetarist view of the transmission mechanism would be centered around the equation of exchange and, you know, the story would be monetary policy actions that start by changing the supply of base money, filter through first to changes in the broader aggregates and then to changes in nominal spending.”

Jon: “That's saying how monetary aggregates are sort of back and I think that's sort of popular discussion and they've always sort of been a bit in the popular discussion, but they're very rarely found in academic macroeconomics papers. Papers really since I think Woodford sort of brought in this sort of the idea is you know you didn't really need your money anymore and you could just sort of have you know have interest rates and we'll get there. I just want to get sort of more into the history of monetarism here and just so sort of we get sort of a complete picture here of what is. So, we have my sense of the history here. We have Friedman and Schwartz's "A Monetary History of the United States" in 1963. The work of economic history, largely no equations, makes this argument that the money supply fell right before real output did in the Great Depression. Hence, the Fed was to blame and the Fed caused the Great Depression to some degree. And this is, you know, the famous, you know, Ben Bernanke at Milton Friedman's 90th birthday party in 2002, saying “Milton, you were right”. You know, all that. Of course, there were some mantras before that. And I think, you know, George Tavlas was a great book out on monetarism and the history of it. There were many, I think, in the early Chicago School days, in the early part of the 20th century, that preceded Friedman in Schwartz. But I think in terms of monetarism coming into sort of popular discussion, I think was largely in monetary history of the United States sort of did that. You also had things like Friedman's K-percent rule, which says that the Fed should grow the money supply at a fixed rate as rule, something like two percent, sort of leaves no room for discretion or stabilizing the business cycle really, the money supply should just be fixed in order to keep inflation in check. This argument I think was highly influential, especially as maybe inflation got out of control in the 1970s, 1980s. Then at that time, much like now, there was this debate over what was causing inflation. There were oil shocks, there was the price controls enacted during the period, or it wasn't money. And there was this big debate during that era, largely the 1970s, so much so, and I think ultimately money ends up winning out. And then what happens is the Fed sort of adopts, after defeating inflation following the Volcker shock and Volcker raising interest rates in the early 1980s, the Fed sort of adopts this targeting of monetary aggregates for a very brief time in the late 1980s. But then the Fed sort of realizes that it can't really target monetary aggregates as a policy tool. So very quickly, the Fed and central banks around the world quickly move to adopting interest rate targets and then they quickly adopt inflation targets in the early 1990s starting with the Reserve Bank of New Zealand. This is where sort of targeting 2% inflation comes in. There was a better debate in the early 1990s as to like, should it be 0% versus 2 %. People like Volcker and John Crow (the then governor of the Bank of Canada) we're on the side of sort of 0%. And then others sort of in the from the New Keynesian School weere arguing for 2% because you don't want deflation that that's not something that's desirable because it could lead to deflationary spiral. Am I getting the history right here with the rise and fall of monetarism in the 60s and 70s and it's fall in the early 90s, is that right?”

Peter: “I think everything that you've said is right and all of the episodes. intellectual and economic from a historical standpoint, you mentioned are important ones. I would just add maybe two comments to dig a little bit deeper into, monetarism is not just description of the monetary transmission mechanism narrowly, but, narrowly, but, you know, more of a broader approach to macroeconomic analysis. And just to highlight some of the deeper themes from monetarism that I think your summary, however accurate, has left out. You're right that the sort of defining feature of monetarism is, you know, that the thrust of monetary policy, expansionary or contractionary, can be seen most clearly by looking at the behavior of the monetary base and the broader monetary aggregates. But the flip side, or the second part of the monetarist critique of Federal Reserve policy and conventional or non-monetarist descriptions of economic history focuses on interest rates as a potentially misleading indicator of the stance of monetary policy. So, this theme comes out to some extent in Friedman and Schwartz but it's developed even more explicitly in Karl Brunner and Allan Meltzer's descriptions of what went wrong during the Great Depression. During the 1930s, according to their story, interest rates were very, very low. This led policymakers and outside observers to conclude that Fed Reserve policy was doing all it could to help pull the economy out of the Great Depression. Instead, according to the monetarist view, those very low interest rates reflected deflationary expectations that instead were the product of a highly contractionary, a perversely contractionary monetary policy, the lesson being, again, that had the Fed been focused on the monetary aggregates as opposed to interest rates, you know, that they could have done more to end the Great Depression. And then in the 1970s, you had exactly the opposite happened. Interest rates were very, very high. Nominal interest rates approaching 20% in the 1970s. Some observers concluded, well, monetary policy must be, you know, excessively tight doing all it can to bring inflation back down, that inflation is not coming back down, just suggests that it has non-monetary origins and, you know, commodity price shocks and so on and so forth. But again, the monetarist's response would be that those high interest rates instead reflected expectations of higher interest rates coming. And, so, you know, real interest rates, on the other hand, were quite low, despite high interest rates, monetary policy was highly expansionary. And again, that's reflected more clearly in by looking at the behavior of the money stock. The reason I bring this up is that, you know, you sort of talk about the rise and fall of money and I think you're partly right about that. Virtually all economists today and I think I would include myself in this group would say that, you know, the Fed shouldn't adopt Friedman's constant money growth rule it can do better than that. But what we have to remember is looking back, let's say, to the late 1970s when Friedman's constant money growth rule it can do better than that. experiences in Friedman's lifetime were, first, the Great Depression, and then the Great Inflation of the 1970s. And I think, even though I might concede, you know, we can do better than a constant money growth rate today, I would also argue that, yeah, Friedman was right. If the Fed had focused on stabilizing money growth, the Great Depression would have been nowhere near as severe and as long as it was. And the great inflation of the 1970s would have been nowhere near as severe as it was. So would things have been much, much better under a constant money growth rate rule from the beginnings of the Fed Reserve history through, you know, the early 1980s, I think the answer to that question is yes. So again, we can get additional insights by thinking about, you know, the unreliability of interest rates as an indicator, as well as focusing on the behavior of the money stock. But I also wanted to touch on another monetarist principle. And this has to do with a view of the workings of the market economy. Again, this is a generalization, perhaps not completely fair, but I would ask you to indulge me on this point too. I would characterize this as a generalization. The macroeconomy: it's summarized by Keynes' own idea of animal spirits, the notion that the free market is inherently unstable and is buffeted around by exogenous disturbances and that it's the role of the Federal Reserve and other macroeconomic policymakers to find policies that help stabilize the market against, you know, in the in the face of these exogenous disturbances, the monetarist view, again, this builds on the important key insights from Friedman and Schwartz, instead views the market economy as inherently stable and interprets depressions, inflations and so on, is a sign that macroeconomic policy is itself a source of macroeconomic disturbances and it is itself to blame for most of the volatility and output in prices and employment that we've seen historically. This is important, I think, in drawing links to the work by the Wicksellian approach that Michael Woodford helped revive and extend and that also New Keynesian economics is elaborated importantly on. The whole idea behind the Wicksellian approach is that as real disturbances buffet the economy, the price system responds in a way that is stabilizing. So, the real interest rate moves up and down over time in order to stabilize the economy, not as a source of instability. So, in this more contemporary view, it's not a problem if central banks focus on using an interest rate instrument as opposed to a monetary base instrument. If the central bank chooses an interest rate instrument, however, it needs to adopt procedures that allow its interest rate targets to track the natural rate and to allow the price system to stabilize at the end of the day, it's not a problem if central banks focus on using an interest rate instrument as opposed to a monetary base instrument. monetarism because it focuses on interest rates. I would actually argue to the contrary that the Taylor rule doesn't body a number of monetarist principles. Number one, it's a simple predictable rule. So, it focuses first and foremost on removing policy itself as a source of economic fluctuations. But also, the Taylor principle, which follows from New Keynesian economics, and it's closely related to the formulation of the Taylor rule. And I believe the Taylor principle was a term that might have been coined by Michael Woodford. I mean, this is the principle that tells us that if a central bank does want to target interest rates, it can do that successfully provided that it adjusts interest rates in response to shocks. So, it's to stabilize inflation. So, I would argue that, you know, when you dig more deeply into monetarists, principles and just step a bit back from the narrow point that money matters, let's say, what you see is that the influence of monetarist ideas, it's still there, it's present in the New Keynesian divine coincidence. It's present in Woodford's Wicksellian approach. It's there in the formulation of the Taylor rule and in all. that, you know, something like the Taylor rule constitutes optimal monetary policy.”

Jon: “Fascinating points about the Taylor rule and how indispensable it is to both the New Keynesian model, you know, closes the New Keynesian model provides equilibrium determinacy. And it's interesting too, how like, I guess you can see how endogenous policy rules are present in also things like fiscal theory of the price level, and other areas too outside of the traditional New Keynesian three equation model. I want to talk just briefly about monetarism in the context of the Great Recession era, that being the zero lower bound on normal interest rates, hitting that in 2008 immediately after the collapse of Lehman and entering into this era of quantitative easing or QE. Milton Friedman in 1998 said, talking about the Bank of Japan, when it was hitting the lower bound in the 1990s, that to quote Friedman, he said, 'The Bank of Japan can buy government bonds on the open market.' That's exactly what the Fed did in the Great Recession with QE, buying government bonds while issuing excess reserves that are held by banks to fund those purchases. But we didn't see much of an increase in M2 or other monetary aggregates, nor an increase in inflation following several rounds of QE. I'm curious, why didn't we see any M2 spikes or inflationary spikes then? Is the issue that reserves were replacing bonds? You know, QE really just all it's doing is just marginally bringing long-term interest rates down by some number of basis points, maybe somewhere up to a percent or so. You can see the QE event studies that Arvind Krishnamurthy and Annette Visting-Jorgensen have done for the Great Recession QE or the event sites that I've done with my coauthors on the COVID era QE. But is that kind of all we can hope from QE? I just remember so much of the macro discussion in the early 2010s was about, will QE spike some inflationary spiral or not? And it clearly didn't. What's your account of this episode in the context of monetarism?”

Peter: “Yeah, that's a great question. Well, it goes back to the same old themes. Another way of sort of asking your rhetorical question would be to say, look, in the aftermath of the financial crisis, and the Great Recession, the Fed did all it could to lower short and long-term interest rates. It promptly lowered the federal funds rate to a range near zero. It used quantitative, used a rather forward guidance to promise to hold interest rates lower for longer, and then it used quantitative easing in an attempt to lower long-term interest rates as well. And you would say, surely that should have been enough to bring inflation back to the 2% target. What exactly went wrong? I do think that a monetarist analysis can provide good answers to that question. As you said, quantitative easing, the first three rounds of quantitative of quantitative easing in the aftermath of 2008, 2009, never really did generate persistent and robust growth in the broader monetary aggregates. So, a monetarist would say, you know, that's the first signal that perhaps whatever it was the Fed was doing was not supplying nearly as much monetarism accommodation as, you know, is Fed Reserve officials and other outside observers would have suggested. You then look at, on the other hand, at the size of the Fed Reserve's balance sheet, and particularly the supply of bank reserves, which did expand, you know, something like by 400% over the period, you know, clearly. the Fed did a lot from a monetarist viewpoint to increase the supply of base money. So, then you have to ask again, well, what exactly went wrong? What went wrong in my opinion was that the Fed began paying interest on bank reserves. This is a kind of peculiar, but particularly of interest historical episode. The interest on reserves was introduced by the Fed in late 2008 and the reason the rationale for beginning to pay interest on reserves was that in 2008 the Fed wanted to conduct and had to conduct large-scale emergency lending in the wake of the failures of AIG and Bear Stearns. Believe it or not, it's hard to remember this at this point in time. But in 2008, actually, the biggest monetary policy concern within the Federal Open Market Committee, as opposed to issues related to the financial crisis, the macroeconomic concern, it was actually over a higher, the threat of higher inflation, not a prolonged period of low or low inflation or even outright deflation. 2008 was the year when every time you drove past the gas station, the price of gas had gone up another 50 cents a gallon. Upward movement in energy prices and other commodity prices were putting upward pressure on headline rates of inflation, and the FOMC was really concerned that those increases in headline inflation would spill over into core and into expected inflation, you know, signaling a loss in the Fed's credibility in fighting inflation. So as to try and do both of these things at once, to conduct the emergency lending that had to be contained but also, to avoid an overly expansionary monetary policy that the Fed started paying interest on reserves as a way of inducing banks to hold the additional reserves that the Fed was supplying instead of triggering the chain of lending and depositing that gives rise to you know the money multiplier and growth in the broad monetary aggregates. So, there was a rationale for interest on reserves when it was first introduced. But then everybody forgot about that. The Fed was paying banks to hold excess reserves. Regulatory policy was pushing as work by Bill Nelson has taught us that you know the entire was doing all it could to induce the banking system to increase its holdings of reserves. So, from a monetarist perspective, sure, the supply curve of reserves was shifting to the right, but regulatory policy and paying interest on reserves was shifting the demand for reserves also to the right; two curves shifting at once, how can you tell which effect dominates? You need to look at the effect on prices as well as quantities. The fact that the increase in base money never did generate robust growth in M2 and never did return inflation to 2% is indicative of the fact that the rightward shift in the demand curve was overwhelming, the rightward shift in supply, as hard as that might be to believe in the abstract. Just to finish the story, though, you know, the last wave of quantitative easing, which took place in the aftermath of the 2020 economic shutdowns and continued into 2021. I mean, did generate robust growth in M2. And this time around, we did get quite a big increase in inflation. So, going back to your original question, I blame interest on reserves on the fact that QE did not work as well as some people hoped. I think QE is a potentially quite powerful tool through which monetary policy can affect output in prices, even at the zero lower interest rate bound. And I would just admit, I think 2020, 2021 proves the point.”

Jon: “Passing, I agree. I mean, there are some people out there who say that monetarism is back. You know, there was all this fiscal spending on transfers during the early 2020s, the CARES Act, the American Rescue Plan (ARP), and so forth. People's cash balances went up a lot. There were also unemployment insurance checks, PPP, all these things. People's cash balances went up, so did M2 just before inflation started rising in mid-2021. So, some people argue that there is a causal link there. I mean, if you look at long-run cost-country regressions of M2 against inflation. There's a pretty strong case that big inflations when they do happen are almost always correlated with big jumps in the money supply. But I'm curious, in this case, in the early 2020s inflation case, was it largely just the fiscal spending or what role did QE play? I mean, all these things were concomitant happening at the same time, so it's time. So, it's really difficult to separate these things. But I'm curious, like, to what your diagnosis is in terms of, like, what this all means for QE. And, you know, we have these sorts of big balance sheet sort of issue that, you know, all these central banks around the world, you know, have these big balance sheets, they don't kind of know what to do. When they've tried lowering the size of the balance sheets, occasionally they hit these sort of liquidity issues. Think like the 2019 repo crisis in the US. I'm curious, with this latest episode in the early 2020s, what have we learned about fiscal policy versus monetary policy specifically QE in terms of their role in causing inflation?”

Peter: “Yeah, that's a great question. The way I would answer it may not be exactly the way that you're questioning was phrased, but in my eyes, the latest experience confirms that all of the lessons that monetarist economics teaches us about the mistakes that happened during the 1970s and can be avoided with the help of, you know, better policymaking, they still apply today. I mean, one way of looking at it, just would be to say, the Fed should have been monitoring M2 growth in 2020, 2021, and realize that their monetary policies involving zero interest rates and quantitative easing this time around were overdoing it, were highly inflationary and likely to lead to problems which they did down the road. But you can also see this by looking at things from the perspective of interest rates. And again, the mistake of using nominal interest rates as the main indicator of monetary policy. My Shadow Open Market Committee colleague, Athanasios Orphanides has a great paper on this. It's called something like “nominal growth targeting is a source of robustness in monetary policymaking”, something like that. But his point is that if you go back to 2021, by the beginnings of 2021, it had become pretty clear that, thank goodness, the sort of worst-case scenarios that we all worried about in the spring of 2020, and you're going to have to go back to the beginning of the year and you're going to have to go back to the beginning of the year and you're going to have to go back to the beginning of the year and you're going to have to go back to the beginning of the year and you're going to have to go back to the beginning of the year first had to shut down, that those worst-case scenarios were not the ones that were playing out that the economy had begun to recover unevenly but sustainably from the shutdowns and then is 2021 war on, it became increasingly clear that, you know, the recovery was enjoying enough momentum that highly expansionary fiscal and monetary policies were no longer needed. Athanasios's point is that by keeping interest rates at the zero lower bound, even as inflation and inflationary expectations were moving up strongly. Again, you would say, well, the Fed isn't changing its policy because it's all being changed rate constant, but that's the nominal rate. If inflation is rising at the same time, real interest rates are falling. And so actually, it's not just that the Fed is maintaining a given accommodative stance of monetary policy. It's that by holding interest rates fixed, it's adopting progressively more and more accommodative monetary policy in the face of rising inflation already. You know, this is the beginnings of the kind of inflationary spiral that in theory happens in the New Keynesian model when the rule is not satisfied. And again, it comes from the same mistake as was made during the 1970s. when systematically the Fed would be too slow to raise rates when the economy began to recover, leading to an overshoot and inflation. And again, that's exactly what we got. Now, what role did fiscal policy play as well? My own criticism of fiscal policy just focuses on... fiscal policy alone. I think that the gigantic stimulus packages that were passed and enacted in 2021 were not needed for macroeconomic stabilization purposes. They were largely done for political reasons and they've added to what was already a problem with the U.S.-run budgetary situation. As far as what role they played in generating inflation, I do think probably another one of the lessons of modern macroeconomics is that expectations matter. And I think excess fiscal stimulus fed into an overall feeling in 2021 that, you know, the Fed had not only, you know, done whatever was necessary to prevent the 2020 shutdowns from, you know, turning into a longer depression, but in fact, we're clearly overdoing it. And you know, that may have led to shifts and expectations that then, you know, fueled, you know, the incipient. me, the main lesson is that monetary policy was inappropriately calibrated in 2021 in exactly the same way that repeatedly it was inappropriately calibrated during the 1970s, and the result, higher inflation, it turned out to be exactly the same.”

Jon: “Got it. So, I guess really broader lesson for central bankers in the context of sort of monetarism here is that the really monetary aggregates should be a better barometer of inflation that we should be paying attention to a bit more while still, it's not a great policy tool in the sense that I don't think we're any better at predicting money demand now than we were in the late 1980s, so it's not a policy instrument that we could use as opposed to interest rate targeting. Are you an advocate of including money growth more so in our models, or are you recommending that we still stick to interest rate targets both in terms of modeling and in terms of what central bank policy targets should be?”

Peter: “Yeah, that's an excellent question. I would answer in two ways. I think it is a mistake for central bankers and for macroeconomists more generally to simply discard the view that there is valuable information content to be found in the monetary aggregates. It's a real shame that no one at the Fed was monitoring and taking seriously the surge in M2 that occurred in 2021. That would have been another indication that the Fed was falling behind the curve. So, if as a result of this historical experience, economists both inside and outside the Fed go back to paying at least some attention to money, I think that would be a good thing. When the European Central Bank first started, it had that two-pillar approach where one of the pillars was sort of Keynesian macroeconomic analysis of the kind that still happens at the ECB and still happens at the Fed Reserve as well. It always will be at the core of monetary policy of the house at central banks around the world. But the second pillar was to use monetary analysis as a cross-check to make sure that serious mistakes weren't happening. And I think the ECB made a mistake by jettisoning that second pillar. And I think having some like that at the Fed would be a big plus. But let me also say too that, again, the difficulty with the Fed reserves policy in 2021, it was that it's fine. I think that history from the great moderation and theory, the Taylor principle, Woodford's work, New Keynesian economics, it's clarified that the problem that central banks encounter when they choose interest rate instruments, it's not, it doesn't really have to do with the choice of interest rate instrument per se. It has to do with the failure of the Taylor principle. If a central bank is going to conduct policy by managing interest rates, it needs to be prepared to adjust interest rates vigorously, especially in response to developments that threaten to push inflation higher. So, you know, the advice that I would give on the most practical level to FOMC members is to say, not, oh, you should follow a constant money growth rule, it would simply be to say. that the preemptive approach to interest rate management, according to which the Fed would raise interest rates before inflation began to rise, not after they begin to see inflation in the statistics. That's the key to conducting successful monetary policy and the big mistake in 2021 and the big mistake that's sort of embedded into the, you know, 2020 revisions to the FOMC's strategy statement, you know, the flexible average inflation targeting framework is this idea that, you know, it's okay to wait until inflation actually appears before raising interest rates. If you do that, it's too late, and then the measures taken to correct the mistake or way more costly than putting up for a little while with inflation that's just slightly below target.”

Jon: “Yeah, sounds like a persuasive argument. So, pay attention to monetary aggregates in some in some form in the sense that the Fed and some banks fall all sorts of indicators. from the labor market, trying to understand labor market slack, but also when thinking about inflationary pressures, maybe not being totally reliant on things like u-star, r-star and narrowing concepts like that, which are very poorly defined. Output gaps are often defined in sort of an ex post-fashion and there's really no clear agreement on what even how to measure output gaps and so forth. And there's all this work that seems to be done on that, but doesn't seem to have produced too much fruit. There are things like hysteresis as which complicates that, is the cycle the trend and so forth. And some downturns have hysteresis, some don't. But, you know, looking at things like monetary aggregates, just as a simple thing to have on the dashboard, bringing that back makes sense to me while still maintaining the existing framework of interest rate targets and sort of Taylor Rule like, you know, policy making that central banks have been doing for the past 30 years, makes sense to me in terms of avoiding things like the inflationary spiral of 2021 or being able to address it sooner. So, you're a big advocate for divisive monetary aggregates, is that right? I know there were sort of a concept popularized by Bill Barnett since the 1980s. I mean, there's all these different types or different ways of measuring monetary aggregates. And there's like M0, which is base money. And you've got M2, which includes things like demand deposits and other things. But then you've got this like division M4. Can you explain to our listeners what the division monetary aggregates are exactly and what do you think the advantages of using them say on our dashboard?”

Peeter: “Sure. The easiest way to explain would be to say that the official traditional monetary aggregates like the Federal Reserve’s M2 are monetary aggregates that are based on accounting principles. So, official, so-called simple sum M2 takes the dollar value of funds held in the form of currency, adds on the dollar value of funds held in checking accounts, savings accounts, money market deposit accounts, and simply adds the dollar value up. The divisia aggregates, by contrast, are constructed according to economic principles, specifically economic aggregation and index number theory. So, the insight that economics can provide, economic theory can provide into the construction of a monetary aggregate. The easiest way to see it is by analogy. This is what we teach college freshmen in principles, how real GDP is measured. You have one apple, you have one luxury sedan, and you have one aircraft carrier, you wouldn't want to say that means that GDP is equal to three. You would want to weight each of those objects by its relative price in order to come up with an aggregate that accurately captures the effect that much more resources go into producing an aircraft carrier or a luxury sedan than a luxury sedan. So, the principle of divisive aggregation is likewise to recognize that different liquid assets provide their holders with different amounts of liquidity services. Currency is sort of like the aircraft carrier. Currency is the most liquid asset. You can use currency in practice any transaction with checking account balances. Still, you can write a check or you can do an electronic funds transfer, but it's a little less convenient. You have money market mutual fund shares, again, a little less liquid, less flow of monetary services from a given dollar volume held in money market mutual fund shares. So, divisive monetary aggregation is an approach to forming a weighted sum as opposed to a simple sum of the funds held in different component assets where the assets are weighted according to their degree of moneyness or liquidity as measured by the interest rate spread between the interest rate on an illiquid bond and the own interest rate paid by the individual liquid assets. So, from the perspective of economic theory, the aggregation is done in a more sensible way. In forming the monetary aggregate, the more liquid asset gets the higher weight, the less liquid asset gets the lower weight. Now, all of this would be just trivia if it weren't for the fact that this can matter empirically. And let me just mention two ways in which it has mattered empirically. In the early 1980s, in the immediate aftermath of the Volcker disinflation, the simple sum monetary aggregate grew at very rapid rates. So famously, Milton Friedman had an editorial, arguing that the Fed was backsliding and re-adopting inflationary policies. The simple sum aggregates he was looking at did not account for, however, is that the growth in those aggregates was coming mainly through the growth of interest-bearing checking accounts and money market deposit accounts and money market mutual funds. Had those components been appropriately downweighed as they are in the divisia aggregates, the money stock, the growth rate of the money stock wouldn't have looked nearly as alarming. So, there is a leading instance in history of a case where this measurement issue does matter. That's... money stock was not growing from an economist's perspective at nearly the rate that the official simple sum aggregates would suggest. More recently, simple sum and divisia M2, the differences between the two have been less dramatic. But where divisia monetary aggregation can really help, you mentioned this implicitly in your question. When you think about including more and more assets in a broader and broader monetary aggregate, in a separate literature, what a lot of financial economists have focused on in recent years is the idea that, you know, treasury bills, because they service collateral, provide important liquidity services to their holders. So, you might say maybe to some extent, Treasury bills are sort of more like M in our model, money and less like B, illiquid bonds. So, we should add them to a monetary aggregate. The problem is, if you just use accounting measures and just add the total dollar volume outstanding of Treasury bills to everything else in M2, basically because the supply of Treasury bills is so big, it just swamps the effects of all of the other assets. If instead you properly downweight the contribution of Treasury bills, you can form a monetary aggregate that includes them, but also accounts in a satisfactory way for fluctuations in bank deposits as well. So, recent research using the Divisia aggregates by people like Apostolos Serletis and John Keating have shown that Divisia M4, which includes not just consumer-oriented financial assets, but also assets issued by the shadow banking system has particularly powerful predictive power for economic activity during and since the financial crisis of 2008. The theory of divisia monetary aggregation allows us to consider those broader monetary aggregates that include assets issued by the shadow banking system in a way that simple sum aggregates never could. So, for both of those reasons, I think you know, the divisia approach to monetary aggregation is a valuable one and is in the end the right one to take.”

Jon: “Well, that's fascinating. I know, you know, there's a lot of, you know, fiscal theory advocates who do say, you know, one challenge with monetarism is what is money exactly? And there's sort of gray lines with public debt and how much public debt do you include? You know, is it just treasury bills? They have a lot of money-like properties and there's I think clearly some sort of like monetary premium associated with it. But you do include longer-term, you know, treasury debt and so forth. I think it's an interesting question and I don't know if you have any particular thoughts on just that. But it may be some of these other measures. I guess M4 is trying to bring in public debt as well or at least on the shorter end of the curve. Which has some pretty clear monetary.”

Peter: “Agreed. I think macroeconomists and financial economists agree on the general principle. In our old fashion models, there was M, which paid no interest, but was highly liquid. You used it to satisfy a cash advance constraint. And then there were bonds, B, that were entirely illiquid, but paid interest at market rates. You know, see in today's deregulated financial system is a whole spectrum of assets that fall somewhere in between the M and the B in our old fashion theoretical models. I think William Barnett was the economist who really pioneered the theory of divisia monetary aggregation and the great contribution that Barnett has made is to give us as economists a coherent economic framework for saying, well, what M really is, is it's a flow of liquidity services. And it's that flow of liquidity services, as opposed to, you know, simply an accounting dollar value that really matters, and we can glean information about the liquidity of monetary services from any particular asset along the spectrum by looking at the comparison between the interest rate paid on that asset and the interest rate paid on less liquid assets. So, you can bring into your empirical work, certainly US Treasury bills, all sorts of assets that do pay interest in that sense like bonds. but also have characteristics of a pure monetary asset as well.”

Jon: “One last question here, and this is going to be about the Shadow Open Market Committee, or the SOMC. It seems like economists nowadays really, or the profession really shuns monetarism and really pays no attention to macro aggregates all. This has been the case largely since the early 1990s, Woodford and all the New Keynesian work that's been done supplanting it with interest rates. But I'm curious, the one place where monetarists can still be found is the SOMC. Now, you've been on the SOMC for a good number of years now. And I'm just curious about it’s history. It’s an outfit that was started by Allan Meltzer and Karl Bruner in the early 1970s. You’re about to celebrate your 50th anniversary. This is an organization that's been going on for quite some time. You meet a couple of times a year and make various monetary policy and other macro-related recommendations, sort of in a bit of a critical manner to what the Federal Reserve is currently doing. You have a big in-person audience. You often have a regional Fed president give a keynote speech; it's not just the FOMC members, like yourself, who give speeches, but also you have very prominent external guests. Could you tell us a little bit more about the SOMC and how it's evolved over time?”

Peter: “Sure. So, as you mentioned, the Shadow Open Market Committee (SOMC) is just a little bit more than 50 years old. It was founded in the fall of 1973 against the backdrop of an inflation rate in the United States that was reaching unacceptably high levels and against a political and economic backdrop where... the principal solutions to the problem of inflation were non-monetary in origin, things like, you know, either voluntary or legislative price and wage controls. So, the SOMC was founded at that time by Allan Meltzer and Karl Bruner, is a way of pushing back against the consensus that any of these jaw-boning or explicit price and wage controls were the way to end inflation and that the real solution to the problem of inflation was to bring money growth back down to more acceptable levels through the appropriate calibration of monetary policy actions. Now, regrettably, throughout the day that followed the 1970s into the early 1980s, the problem of inflation just got worse and worse. So, throughout the 1970s, the Shadow Open Market Committee consistently argued that the Fed should try and pre-announce a gradual approach to disinflation, bringing inflation back down gradually to acceptable levels, by gradually reducing the rate of money growth. As it became more apparent moving into the late 1980s and the early 1990s, the Shadow Open Market Committee's advice didn't change, but because the conduct of monetary policy improved, you started seeing Shadow Open Market Committee members not only criticizing what they didn't like about Fed policy, but also supporting statements by Fed Reserve officials that reflected the advice that was traditionally given by the Shadow Open Market Committee. When Alan Greenspan as Fed Reserve Chair, for example, explained to Congress that, you know, the best way that the Fed can provide for maximum sales growth is by stabilizing inflation first. I mean, he was echoing the message of the Shadow Open Market Committee and that, you know, the SOMC happily acknowledged that fact. So today we continue in the same tradition. Sometimes the Fed Reserve does things that we can support. Sometimes it does things that we have issues with. I think one of the things that I'm proudest about regarding the Shadow Open Market Committee, this is something that I think is sorely lacking in other venues in the public arena these days. We try to keep the focus on policies, not personalities. So, if there's something about what the Fed does that we don't like, we try and explain in economic terms what it is we don't like about it and propose specific alternatives. This isn't about criticizing particular individuals or questioning their motives. What we really want to do is to contribute to improved monetary policies, monetary policymaking, fiscal policymaking, and bank regulation. I hope that's what we continue to do. But it's always, I think, in the spirit of, you know, healthy, respectful dialogue with Federal Reserve officials. So, you know, as you say, these days, we like to invite FOMC members to speak at our meetings. And, you know, we always find that the insights that we get from, you know, central bankers, you know, interesting and informative.”

Jon: “Well, always a treat to get to hear from the members of the SOMC and congratulations again on the 50th anniversary. Looking forward to celebrating that further with you and really want to thank you so much, Peter, for joining us today. It's been a real honor. Having you on really a joy to talk about your career, your vocation, as well as these really important issues in central banking, New Keynesian economics, and monetarism. Thank you so much for joining us. Today, our guest was Peter Ireland, who is a professor of economics at the Boston College Department of Economics, a leading monetary economist and a member of the Shadow Open Markets Committee. This is the Capitalism and Freedom In The First Century Podcast, where we talk about economics, markets, and public policy. I'm Jon Hartley, your host. Thanks so much for joining us.”