Ricardo Reis on the Field of Macroeconomics and His Defense of the Discipline

Ricardo Reis is a professor of economics at the London School of Economics and the editor of the prominent Journal of Monetary Economics. He joins Macro Musings to discuss the state of macroeconomics, which has recently come under attack from many commentators who claim the discipline lacks empirical rigor and has failed to accurately forecast economic conditions. Ricardo gives a nuanced defense of macroeconomics, arguing macroeconomic research is actually vibrant and empirical. Furthermore, he argues that although there have been shortcomings, macroeconomics has greatly improved over the past few decades in its ability to forecast and inform policy debates.

Read the full episode transcript:

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

David Beckworth: Ricardo, welcome to the show.

Ricardo Reis: Thank you very much.

Beckworth: It's a real treat to have you on. So as I do all of my guests, please tell us how did you get into economics and then particularly into macroeconomics?

Reis: Sure. So I studied in England. I'm originally from Portugal, but I came to England for my undergraduate. And here you pick your degree as you apply, unlike in the US where you pick after university. And the reason why I picked economics was that in high school, I'm sorry, my favorite subject was by far history. So I really enjoyed history and understanding what happened in the past and how it happened and so on. But I was a much better mathematician than I was an essayist. And so as a result, economics was a lovely way to think about history and why things happen and look at the past rather than actually doing history itself. And so I went to economics, which I thought was a great combination of what was probably my greatest strength in high school, which was mathematics and what was my real interest and passion, which was history. And so I went into economics because I thought it was a great combination of those and I wasn't disappointed because indeed it was.

Reis: Then, I guess, when I was in college, initially I interned at an Investment Bank. It was very common here in London where I studied to do summer internships to figure out what you want to do after. I interned initially in an Investment Bank, figured I didn't like that so much. And then I was very fortunate because I did an internship at the Bank of England where I was a research assistant on some research projects and I really enjoyed the process of it and so decided to apply for a PhD. And that's how I ended up being a PhD economist.

Beckworth: Very interesting. You've written a recent article that's titled, "There's Something Really Wrong with Macroeconomics," and as you know, the motivation for your whole piece is there's been a lot of blowback, a lot of criticisms of macro. Interestingly, there was a Wall Street Journal article earlier this year following the American Economics Association meetings that talked about the angst in the profession. I had a reporter on this show a while back, Josh Zumbrun, who wrote the piece and he makes the claim at the meetings there was just this sense of angst and despair in the air because of the new president and some of the potential jobs not being realized because of cutbacks. But also, just kind of the sense that economists weren't going to be taken as seriously in the new administration.

Beckworth: Your critique though, of course, is more focused on macroeconomics and you go through a number of areas in the article. So I'm going to work my way through them with you. Again, there are a number of pieces have been written by several authors questioning macro's relevance and other areas. Let's begin with the first one that you cover, and that is macroeconomic research. So what has been the critique of macroeconomic research and what is your response to that critique?

Critiquing Macroeconomic Research

Reis: So the motivation to write this ... Well, first of all, I wrote it because I accepted to take part of a debate at Oxford where several people were coming together and it was a little bit one of these debates where many of the participants were coming to state some of the different criticisms of macro, with which many of which I share insofar as I'm certainly the type of researcher who's eventually unsatisfied with the state of the field and always thinking of ways to improvement. And so I find in some ways those natural. At the same time, I thought that these debates have, over the last three, four, maybe even a little longer years, have I think exaggerated a little bit the problems. And so that was the motivation for this paper.

Reis: In particular, I do find it a little ... I was getting a little bit ... I say frustrated, I guess may be the word, at talking to students who would come to me and say, "Well, I just read that macro's in a very big crisis, professor, and there's all these problems." And I would say, "Well, I don't think that's really quite right." And being students, especially my master's and my undergrad students, they would say, "Well, but then what should I read for the other side or for defense or for this debate?" And I had to admit that there was really nothing for… so there was a little bit of the attempt at writing this piece. So that was the motivation.

Reis: Then the state of macro research, I think in many ways it's actually very exciting. And now, I'm certainly biased because I'm right in the thick of it, but I think people are very excited about questions. I think the great recession raised a tremendous amount of questions. People are making progress in them. I see macroeconomics being a much more diverse field nowadays where the questions are much more interesting. And there's a lot of young exciting people working on it. So I see a field full of vitality right now. And so in that sense, I have sort of a different perception from that of most of the critics who point to it as being in a field in, as you said, despair or so on.

Reis: And so let me do one particular example, because I think that's a good one. So when I finished my PhD about, I guess 13 years ago, and I was quite interested in topics of monetary policy and inflation and monetary economics, and back then you had two essentially branches of work. It was a theoretical work. This was a time in which we're digesting, especially the very important book by Michael Woodford and others. And this was a theoretical work that had to do with a little conceptually of how do we think still about it and concluding with those debates on why is there monetary neutrality, why does monetary policy have a real effect?

Reis: Then there was some applied work of trying to take this to the data, which was especially very important work by Larry Christiano, Marty Eichenberg, Charlie Evans, Paul… But there was also had a strong methodological component, how is it that we even take some losses data. And then finally, the policy world was super dominated by essentially should the parameter the Taylor rule will be 1.5 or 1.6. So it was I would describe the work as somewhat as track theoretical or when it came to the apply policy. I would even say, a little boring. And so the theoretical should be really 1.6, 1.75.

Reis: I'm not sure that's the most exciting question. You look at work on macroeconomics today and it's questions on how do we model liquidity? How do we think about liquidity now that we have certainty of liquidity in different assets? We have questions on a central bank balance sheet for guidance. What are these policies with a central bank the some things or if it does some others. So in some ways I see the, we see a lot of micro data together with theory being used both advance theory, you to understand with liquidity is as well as a tremendous amount of use of natural experiments as generated by the unconventionals, Paulson central bank. So I see I think the field is just much more exciting now. And I see a lot of progress being made relative to before.

Beckworth: You mentioned in your paper, a survey of some top recent PhDs that were hired. Also, you look at the sample issue of your journal, and you show there's a rich variety of papers approaches. They don't think kind of stereotype that you've seen the critiques, right? Of macroeconomics.

Reis: Precisely. I mean, I think a lot of the critiques are in my view better targeted to a little bit of what you see say in two classes to macro. And then those it's true that one sees a lot of representative agents. No inequality. Either agents know everything, or they almost know everything isn't the case rational expectations. There's really very little financial frictions or financial markets of any kind. There's no housing, little role for banks or for mortgages.

Reis: If you look at actual research and where people are doing, though, none of this really applies optimists are indeed, it's very common now to see models with a lot of heterogeneity with financial frictions with banks, where there is it's completely normal to walk into a seminar today and see someone present a model where there's no wrestles with nation but there's some least squares learning or sticky formation or some other model of expectations, and nobody really waves or thinks that that's the type of any kind and so I don't, I didn't really see the well I understood where the critics were coming from. I thought the critics was a more light to a very, very narrow view of what the state of macro is aspirants is taught in introductory classes, and not as what the actual research is and how varied and diverse that is.

Beckworth: You also note in your paper that a number of other indicators suggests that the field is healthy. You mentioned that publications have not declined that the share of journal articles going to macroeconomics is fairly stable. You also note that there's, yes, there is finance and macro intertwined and many of these articles that yes, this increases the crisis, but it's been there. I think you said since 2000, you also know that there's still plenty of macro economist getting jobs. So on many fronts than it appears macroeconomics is still healthy. Is that right?

Reis: So good. I don't mean, my training as an economist tells you that when you get into these debates, you look for evidence. And I just, there's just isn't this very clear evidence of this tremendous crisis, as judged by how other fields perceived macro how departments see macroeconomists, macro or others. And part of that, by the way, is also a certain pushback against the idea that macro is in a particular crisis, whereas other fields are doing just fine. I don't think macro looks any different from economics in many regards. And so in that sense, I don't see a huge crisis, I see a lot of excitement. I see a lot of people doing interesting work. And I think I'm not sure at all that when we look back at this time, we will see that.

Reis: Let me also add, because you raised that, which was pretty good one is the fact that if you read an enormous, too many of the criticisms of macro, say DSG, and describe essentially, the baseline DSG model and then go on to be very critical of it. Such that one gets the impression from reading some of these criticisms, the DSG is really dominate macroeconomics, and that is very far from the truth. So again, if you look at publication patterns, and I know this insofar as I was in the thick of it back in the mid 2000s. Actually, most of the DSG papers had a lot of… getting published to the top journals the some of the key top DSG developments were being published in field journals but certainly not in general interest journals.

Reis: I mean, the famous Smith founders models published in the journal European Economic Association, as part of it, invite a conference. And you look at many others of the work that was being done, and that was the heavy cited by Jordy and many others, a lot of it was not actually making it to the top journals of the profession, which were in worse. Instead, you look at those top journals back in the mid 2000s. And they have plenty of papers with macro and finance and different kinds of financial frictions, asymmetric information and even banks. So it was not the case. I think that at least when I was looking for the research frontier, that there was any sense that DSG was dominating worse financial frictions were being completely neglected. That was not at all the reality in which I was living in as a researcher.

Reis: What was true is that central banks especially… institution, founded yes you work very useful. They started using it a lot and applying a lot, and perhaps less so of the theory of financial frictions has been developed. And so I think the critics look a little bit of what was going on in terms of what parts of macro were being used and applied to policy debates, as opposed to the ones that are actually being published. Because I can tell you back in the mid 2000s, the New Keynesian DSG economist, and you can check that in so far as seeing the ones that were writing some of the very important papers from that area, there are now very highly cited. These were not people that were being treated very well in the professions as far as getting published and all the top journals and saying, well, this is the really what macro is. So the dominance of DSG was very far from the truth.

Beckworth: So the state of macro researches is healthy and vibrant and a budding young macro economist to take heart and that, let me ask one critique that's raised and I don't know a Smith is raises many times. So he writes for Bloomberg view, I know you cited, I think you cited his article, some of his article in your piece. So he knows that yes, there are some quasi natural experiments been done in macro. Yes, there's some maybe administrative data that's being used some empirical tinkering here and there.

Beckworth: But I think one of his critiques is and I've heard this other places as well, is that macroeconomics is fundamentally limited in what it can do empirically, because we can't do true randomized controlled experiments at a macro level. So some of the maybe the details, we can test them the frictions we might test, some of the maybe, some evidence pointing towards some issues on the margin. But testing which business cycle theory makes more sense. We just can't run recession. It kind of like the example that gets thrown out. We can't run multiple recessions on a country to test some theory. So what is your pushback against that?

Reis: So first of all, I don't want to push back and so far as I want to agree that I think there are the features of macro insofar as the unit of analysis is aggregated, and that we have limited time series data, and the inability to experiment does limit our ability to learn Or to test some mechanisms as well as we could in some other fields. So I do want to absolutely acknowledge that and agree with it 90% if you want. But let but again, for the sake of the debate, let me tell you that 10% where I disagree, I'm your price agrees that I think I'm yeah, no Smith, maybe this would be an interesting dialogue with him. But maybe it's under estimating a little bit the creativity of a lot of macro economists so far as again, while in theory, I would agree with this point, and very wholeheartedly agree.

Reis: But then look at some of these peoples whose paper I just, whose work I described in my paper, for instance, but we could look at others, and they're all coming up with creative ways to use microliters as this or that mechanism and to build macro models that incorporate this or that. And again, none of them is really contradicting Noah's point, which I think is a valid one, which is that we can ultimately test like you said, or your point better not Noah's but your point that you can test it is a cycle theory. But boy, there's so much you can learn using different data, different techniques, different ways to look at the problem and building up towards it. So I guess on the one hand, perhaps it's just because I'm less ambitious than you, David, that I see. I still see a lot of scope for us to learn with the data we have before we hit that limit, where we just can't do anymore because there isn't any data.

Beckworth: No, I'm very sympathetic to that view as well. Well, let's move on to another point you make in the paper. You're going through, and you're evaluating the state of macroeconomics, so that we've touched on the research agenda, part of macro. But also you touched on some other areas. Let's move on to macroeconomics, policy performance. So how have policy makers done with macroeconomic policy? What is your takeaway there?

Macroeconomic Policy Performance

Reis: So I mean, I think some of the limited, so there's two points that I like to make. The first one is macro economists. And by that and my would even add economists is in general have a much smaller influence in policy that often is scribe during debates in the press. So, macro economists say lots of things and most of them are not picked up at all. And thus, if macro policy has been good or bad, it's only through a small percentage that economists can be directly celebrated or blamed. So I think that's important to keep in mind. Most policies are driven not by macro economists at all.

Reis: Point number two, on the proposed macro policy, and I think it's hard after the Great Recession, after we've had such a tremendous software systems to say that that was very costly in terms of human welfare. So things are going well, of course, they're not going well, we'd like to do much better. So I think you have to, I think it's inevitable to simply say things aren't going well from that perspective. At the same time, let's think in terms of not going well, relative to what? Relative to the past. Oh, I mean, in the last 40 years, the business cycle has been relatively stable relative than before. You're not pretty 40 years, but sorry, a little less, 35 years, in spite even of the Great Recession, the fact is that we haven't had the great stagnation the way we had a little less than 100 years ago during the Great Depression, where session was bad, but it was not a great depression. So that's what's relative to history hasn't been that horrendous.

Reis: Two, if you look at a different macro variable, not recessions with a controlled inflation, last 16 years have been amazing, remarkable. I mean, inflation has been essentially, both the federal reserve the ECB, Bank of England, the Bank of Canada, they've all kept inflation very close to their target, now running for 25 years or more. Moreover, if you think about the shocks we had in 2010, in terms of a deep recession, wide swings and inflation because of commodity prices, and many others, you would have potentially guessed that look, inflation is going to go out of control and you could see people you could you can easily find people pleasant then saying that inflation is about to shoot to plus infinity or about to have deflation. Is the amazing performance of central banks and keeping inflation literally tight around two.

Reis: I think it's quite remarkable that we give nowadays the Fed and the ECB a very hard time. Because in the last three years, they've deviated from… percent by about 0.5%. If we're giving them a hard time on that, then I think they're doing in terms of monetary policy, we're doing a fantastic job relative, given that deviations the past like in the 70s, where inflation would go to 10 or 11%. Being off by 0.5 doesn't seem so bad. So in that account, again, in controlling output, certainly not good, controlling inflation extremely good. I would say I would just behave in last few years. And then more generally, in the my third and final point, David, is that, again, let's be somewhat realistic in terms of what one can achieve or not given how little knowledge we have, given how little funding we have in economics. Given how few decades we've been studying economics in a serious way.

Reis: To expect that we can keep an economy healthy at all times seems like a great demand when, in comparison with like I do in my article will say, Edison or many of the fields, which we've been studying for much longer with budgets that are 100,000 times bigger. And yeah, every so often we also get hiccups. We also get viruses that come in spread and kill quite a few people and others. And that doesn't lead us to give up or think that epidemiology or medicines to crisis leads us to just make more progress and keep on seeing how things are better. But that doesn't prevent crisis from happening. Likewise, I think if you take that perspective for me, I don't think we make some of the mistakes that we make in the 70s. That doesn't mean that new shocks will appear new mistakes are made. The same I think applies to health we don't die of what our parents or grandparents died, we cure some of them. But that doesn't stop new viruses and new diseases from appearing and mortality from taking its toll.

Beckworth: I think that's those are great points and fair. criticisms of some of the points I've made in the past. I mean, yeah, we've gotten past a great inflation. Some of banks have done a great job keeping inflation anchored the counterfactual as well, we could have had a great depression. So why complain? And I'll confess, I've been one of those folks who've been critical of fed not hitting its inflation target. I'd go even beyond the 2012 the Fed increases its target.

Beckworth: But I think as you well know, there's been many studies that show it was implicitly doing it well before then. And since, the end of the crisis has been kind of persistently below and your points well taken, if it's 20 basis points target, Count your blessings, right? I mean, not the worst thing in the world. I guess, where, let me just push back on this regard wouldn't have been nice to have had a role, more robust recovery. I mean, it's yes, it's great that we didn't have a great depression, but-

Reis: Absolutely, yeah, like I said, I think I have no way but I want to say that macro policy or the macro economy, let alone macroeconomist have done very well, I think there's been a flop. I, of course, would have been great. It would be great if the average growth rate was much higher. But I was counteracting that some failures and also some successes.

Beckworth: Yes. And stepping back again, look at my children, they live in a much better world than the one the 1970s that the previous generation went through. So again, we were very blessed in this day and age. But I since I have you-

Reis: Well, the more than that, more than just macroeconomists are not responsible for your children living better than you do I mean. Macro economists and macroeconomic theory and policy only as a small contribution to both the frustrations and the joys that you have on the wellbeing of-

Beckworth: This is true, even if they gave perfect prescriptions. There's both technological developments and policy blunders that can contribute. Let me ask this I have you I don't very often get to talk to the editor of the top journal. So one of the critiques I've had of policy is that QE no QE was radical to give the Fed and ECB credit for doing QE. But my reading of like, Woodford and Egerton 2003, and Koopmans’ 98 paper and some of the literature that followed that is that QE is going to be irrelevant unless you commit to some kind of reflation unless you have a permanent increase in the monetary base. But unless you have like a price level target or something like that, QE, by definition is going to be temporary. If it's going to make some difference. There's lots of empirical evidence that leads supports that, but it's not going to create this rapid, rapid recovery. Is that a fair reading of the literature on these papers?

The Debate Surrounding QE

Reis: So I think QE is a very exciting topic for research right now. So let me try to summarize at least what I think I know about it, without say that that's the state of the let's reverse this is why I think I know. I think the benchmark is that QE is neutral. It has no effect whatsoever. And again, I think going to air so Mike Woodford show that very nice, three paper and more recently Pierpaolo, Benigno and Salvatore listicle. And even I think, more thorough showing of that result.

Reis: When you show that essentially, if you're buying one government liability government bonds by issuing another public liability reserves, you really aren't doing much aside from shuffling the assets from the overall government that are being held by agents. And so there's a very strong case for this to be actually completely neutral. Now, therefore, this has led to a lot of attempts of theories that show this is not neutral. She wants some people refer to the neutrality few is a little bit like Modigliani Miller neutrality, and it's the result is not the case that it necessarily holds, but provides them a very strong benchmark. And so you see, very often stories you hear about TV being not neutral are really not about per se. And again. And that's the beauty of the neutrality result.

Reis: Same as what a… for instance, you mentioned the permanent increase or permanent transfer. Mean, those theories already theories where you're not doing QE, what you're doing is increasing the stock of banknotes of currency. And boy, yeah, that's inflation. And we've known that since the Friedman or much before Friedman. And so that's not QE, QE is buying reserve issuing reserves to buy government bonds, and that is neutral. Now, what is the evidence? So now the evidence is shown. So let me now tell you the evidence. And then some of I think some of the theories out there may be convincing. Evidence one, QE had a clear effect on different yields spreads of different assets when it was implemented, particularly when you're going out and bought 10 year bonds, you had clear increase in the price or fall to the yields of 10 year bonds, as well as changes in a bunch of across the different financial markets. On that evidence.

Reis: Let me note also that the estimates tend to be small insofar as it takes a trillion of QE to have an effective 30 or 40 basis points, which is a lot. So that but the privilege sure, I think I shouldn't miss either zero, or that they're not zero. I'm sorry. What about the brilliance of QE? Is that any effect an awkward inflation? My reading is that it's had zero impact on inflation, at least, because one sort of looks at the same tools using inflation options, and others very little. What about an output? Output is harder. It's been done using VAR, they tend not to be very well in the fight. It's hard. I'd say the jury's still. So as far as I can see, QE have clear effect on financial markets but not large, zero effect on inflation, close to zero, and effect an alpha essentially out there to determine. Now let's try to make sense of what I think are some viewers out there.

Reis: Theory number one, relying on neutrality. QE all it was, was basically a signal that we were serious about for guidance. QE itself was neutral, but its signal that the Fed was really serious about keeping interest rates low for long. There's a very nice spin on the theory that I like very much builds a little bit on some workouts that involve Hall, especially the very nice paper by so rush back ri, and UT Austin and others. That makes the following point. When the Fed bought a lot of long term bonds by issuing overnight reserves essentially became a long short hedge fund. Well, what's the big risk of having a portfolio is that if the younker steepens, very quickly, the Fed would make big losses. Well, by gives you QE essentially what the Fed did was adopted commitment in terms of its portfolio, because if it indeed it reneged on the promise to keep interest rates low for long, you would suffer very large losses on its balance sheet.

Reis: So again, you see QE here is a commitment. It's a signal of what it would do by interest rates, but its interest rates that put down inflation and not that not 3%. So it's one line of theories that essentially, if you want the QE not as effective per se, but just as a signal, there's a commitment. Now a second line of theory says look, there's a lot of brick failures of arbitrage and bond markets. As a result QE that buy some bonds and others as they are sometimes when image filling the gap by Rob Reid with many other … especially, there's been some phenomenal work on this showing that indeed when you make purchases of bonds of different maturities, you seem to change some of those yields and those initiative why's that that as you change those yields, that transmits it to changes in investment that may have some effect even if very modern inflation perhaps an output.

Reis: In those theories, opinions and valid get it's hard often to make an quantitative work although Glasgow Korea but very good work on this and others, then you know third set of theories that look what QE does, is in replacing government bonds with reserves, which are doing is affecting each other, sorry even though governments have reserves are two public liability, they different their properties. So reserves are issued by a central bank. So it's seniors that response for them as opposed to tax revenues, two there being held only by banks. Where's bonds can be held by households or banks or any other investments up.

Reis: Three, you choose the interest rate and reserves response has to be option and so have a price determined by the market. What they mean is that when you substitute to one by the other, you are changing the risk composition of assets in the economy. For instance, reserves are safer than government bonds. This is not a point that's particularly valid for US. But valid in Europe, reserves are much safer than the Greek bonds they that the ECB was buying was a result you are increasing some of the supply of safe assets in the economy, as well as changing the distribution of them, because reserves are being held by banks, whereas the bonds will be held by lots of the people. And this change of the supply if they fast is that an old one can have an effect in financial markets as well as those three I think of the exciting theories going on it.

Reis: There's a lot of again, people trying to make sense of unique, Ben Bernanke he put it very well some years ago when he said, QE is something that works, but we don't know quite why it works. And I think there's some truth to that. And so that's kind of my summary of what we know. At this point, I think it's very important to neutrality benchmark, and not get fooled into just facilitating monetarism or some other theory under the guise of QE. But understanding why QE per se would actually matter. I think is like I told you it's either because of financial imperfections, because reserves are special, like a bond and not substitutes, or because QE is a signal. I think those are the three leading theories if you want. Yeah, approaches.

Beckworth: It's very interesting research being done. So let me step back because you framed this very nicely, like my critique of a permanent transfer that you say that well, that's not QE that's something different. That's fair enough. So why didn't the Federal Reserve do something more like that? Why spend all this political capital on like QE two, QE three and the QE one, I think you've been putting your articles it really did make a difference. QE two QE three, I think there's less evidence. I may be wrong on that.

Reis: Yeah. That's true.

Beckworth: So why not take all that political capital. Because I mean Ben Bernanke he got a lot of heat for QE two QE three, why not take that in and overshoot your inflation target adopter price level target nominal GDP level target. Any thoughts on that?

Reis: I'm a huge fan of price level target. I've written a lot about it in many years and defended that we should … So I couldn't agree more with how the adopter price level target report. Now of course, that's adopted the target that you have to go about implementing and trying to achieve it. And so QE and or interest rates would be one of the way the private key the target now, so the target just by itself, might have again, very extreme rationalizations models won't make a difference in itself. So you have to then do the actions to back that. So I agree with the price level target. I couldn't agree more.

Reis: I think it's a great idea for many reasons. But you still episodes when you asked, why not do just the transfer. You said, Well, why not just do the printing app. I guess the modern version of this and why not do the big helicopter drop of money would be really I think that version. And there's some reasons all on perhaps why not to do that, which is, which again, has to do with how desperate The situation was. I think that if you read indeed we were plunging towards deflation, if the situation was very bad, then you know, at some point can the central bank indeed stop the economy from going to a big inflation spiral, of course it can, just print banknotes and put them out there, we know how to generate inflation. The issue is that we also know from the use of monetarism as when it was applied, that while it worked as a theory as a practice, it is very, very hard to calibrate.

Reis: It's extremely hard to figure out how much M do you actually print or your transfer to just get 0.5% more inflation as opposed to five or 15% more in play? So I thought, the situation that in mind you never got that bad that was justified, that given the amount of let's say controller you'd have. A second point is that, then comes the... I think the Lucas critique, and it's one, it's somewhat more applied sense, which is, if you're going to do something completely unprecedented, like flying a helicopter and distributing cash, you are going to expect that our knowledge of the structural problems is not that good enough that there's going to be some big shifts in the economy. So I'd be very worried about doing something that radical given that I would expect that the money demand that many other functions would shift in a big way, compared with QE.

Reis: QE I mean, once you start with open market operations, which is of course what I've been doing for decades, they said, "Well, what I'm going to do is just increase reserves, buy bonds such that essentially, which was what QE one, did, we enter the free mineral, let's just pay the interest on reserves equal to the interest on bonds, do something we understand very well and which economists have been talking about for decades, which is let's achieve the Friedman rule and have the interest on reserves people's interest on bonds by having a balance sheet of 1 trillion." And then he said, "Well, now that we're here and again, I think because we want to signal or because we want to do interfere with national markets or change the competition of risk and safe assets. Let's go from 1 trillion to 5 trillion $4 trillion." That seems to me like a much safer bet, then then go into something much more radical.

Beckworth: Fair enough. Well, let's move on to the other two pieces in your article macroeconomic forecasting. I mean, economists get a bad rap. We didn't see the Great Recession coming. And you got a great analogy between medicine and economics. So share that with us.

The Analogy of Medicine and Economics

Reis: Well, then also the following, you should go to a doctor and ask him, "Doctor, will I be dead next year?" He's going to look at you and think you're crazy as hell that's really not what I do. I don't forecast them conditioning. What I do is I tell you, if you smoke, you're increasing your likelihood of being dead next year. Or if you do engage in this or that ages activity, then that can be very dangerous or not. Well, likewise, I think with economists, where we forecast is, likewise they always say look, if you adopt the bad policies then you're likely to pleasure economy into a recession or not. That's very different saying what's up to be exactly 12 months from now?

Reis: Secondly, I mean even the doctor would say even after he told well if you smoke I think you increase your chances of dying you still realize that look, but a lot of people who smoke are very healthy and live to very late. Likewise the meaning not take into account that forecasts are always probabilistic that there comes some probabilities and just point to the fact that something did or did not happen is not really I'm placing the forecast percent I think in medicine we find this very normal. I never expect, if I go to my doctor, he says, "Ricardo if you keep on eating french fries the way you do, you're going to end up with massive cholesterol six months from now," and then six months from I don't have cholesterol, I don't do the doctor was wrong, I just realized that I got lucky or not.

Reis: So that's the next one. The third continuing on this analogy, if you want is that what I do expect the doctor to is that when are two things when they get or at least in my relationship with my doctors one, that if I get seriously sick, he has waste security. Well, the economy got seriously sick in 2008. And it didn't die. The Economist, the fact that we did turn to economists and what to do in terms of whether was tarp or QE, or lots of interventions. And you know, the we did, were able to provide some quality advice to prevent a new Great Depression doesn't mean that we weren't sick. Again, when I go into the hospital, I spent the doctor give me alive, I don't expect them to get any fine and recovered within three, three hours.

Reis: And finally, then when I start to go to the doctors, that they have some understanding that he can tell me if you want the story, why isn't it some of my behaviors will lead to more or less health will lead for me to recover from a disease or not. And likewise, an economist would expect economists to focus on a strange in a sense, how is it that not the body but the economy works, and so on, and in doing so, being able to therefore nothing affected but especially understand how it works. So many ways, I guess... My point is to try to get the push back and say that economists are supposed to forecast what GDP or inflation some of the variable is exactly. But understand that all forecasts are probabilistic, you have to be conditional that are important responsibilities to prevent big shocks, and especially to understand, and I use the analogy medicine simply because I think we all leave the doctors and we, this is the way doctors live with us. And we don't throw them under the bus for.

Beckworth: There you go, economist should be treated just like medical doctors. Fair enough. Let's go to the last point in the paper and that is macroeconomic teaching. What are your thoughts there?

The State of Macroeconomic Teaching

Reis: So when I read the criticisms of macro, and this has already come up in our conversation, several times-

Beckworth: Yep.

Reis: Many of them seemed spot on in my view on the way we teach macro and that is where you don't see enough variety. You don't see enough use of data. You don't, you do seem to do much with this regard a facile frictions in equality. And all of those faces many of the things we've spoken about. And so the question is now I think a reply by many of my colleagues would be well, yes, but that's just teaching you start with just the frictionless finish line, and then you build up on that. And I think that's a valid person. That's indeed, the way we've been teaching economics for a long time or macroeconomics as a for a long time.

Reis: But I'm not completely convinced you right, therefore, in some ways, I joined the critics in the following sense. First, on when I look at the way my colleagues have started adapting the way they teach micro, what I see them is not saying, "Well, yes, we're going to have to go through the Slutsky matrix in the compensator elasticity because that's just how it is." And then you learn about how these things are interesting later. No, I seen the prosecution economics that use a lot more data to see a lot more examples use the insights from behavioral experiment economics in order to make the material likely munch especially closer to the frontier more easily. Likewise, I'd like that we do macro, the more of that we start with the frictions much earlier, as opposed to going through all the neutrality benchmarks, and then never really quite making it to the frictions until all the advanced classes.

Reis: Second, and perhaps more point to, I think it's in graduate macro, I think graduate macro, there's still too much of going over, especially in the first few years, just going over the fictionalist cases, these ideal cases, if you want the real business cycle, the SG, whatever you want to call it a little bit benchmark. And then the students especially most of them, don't write this the macro. That's the only macro they see. And they really didn't leave with this perception. That macro is this frictionless set of models, all of which we know are completely wrong. Yeah, there was the neutrality benchmark, and then ever get to see what's the exciting macro that's at the frontier, where there's a lot more data as a lot more methods and how to bring them all so the data, there's a lot more in terms of incorporating a lot of things.

Reis: And so I kind of I support a critics here, but I think that decreases, we better point to saying, "Hey, we should be writing better textbooks in graduate macro, are we thinking those attempting some different things in them?" Let's see what works. Let's see what doesn't. And I think that would be a much more fruitful debate. I mean, part of the suggestion is also that far too often debates on the state of macro and up with, well, I think we should do this, you think we should do that? And then we just go and keep on doing whatever we're doing. Whereas I think if we had like, actually a serious debate in association meetings, or some others, what should be in the core macro curriculum, I just find that to be a much more productive age to change the profession, and that would have a much bigger effect. And I see a lot of potential room for improvement. How should we teach first year graduate macro? Is something that I think we could have looked it up a very, very productive debate and when they would be very useful.

Beckworth: Interesting. One of the things you point out in this section, and it's a theme that runs to the rest of your paper, well part of your paper and that is the emphasis on empirical work. You know that you could have an entire course centered around a macroeconomic time series or empirical work. And an earlier in the paper, you want to read a quote, you echo the same spirit when you say, "My decades long frustration dealing with editors and journals that insist that one needs a model look at data, which is only true in a redundant a meaningless way and Lisa dismissal too many interesting statistics." So I'm sensing from you that you think there needs to be more empirical work done with macroeconomics?

Reis: I think that's true. I mean, I think macro was more among the different fields in economics was towards the more theoretical one. I think it's changed a lot. By the way, I think it's become a lot more empirical for the last 15 years. And I think it can even become a little bit more so. And I think there's room for that to be specially again, though, and as you notice, when you start your question, when it comes to teaching macroeconomics, and again, pick up the standard macro graduate textbook, or walk into a standard first year graduate Magna PC program, or even in the master's program, there's a remarkable absence of any empirical work or group of techniques though. I think we can do better. It will be back again, though, going back to the end of our conversation, there was a lot being done in research already, I think in teaching is where once you work on your first quote on the first one, you know, I mean, what I mean is that there's too much often.

Reis: This is one of the put downs you sometimes get in seminars, which is for sophistic, and someone says, "Well, you need a theory to think about data, you need model to system number." And I've always found that not really good point so far as you could also say, well, we need some data to even to put up a model. If you don't have that I mean, sure, an ideal world and be both but understanding that each article is trying to do one mass a contribution to our models. I think writing a paper that has no models that just has some data and documents is just sophistic, this is a great way to progress and likewise make something a theory and then have some papers that combine them. I find in, especially when you get to empirical work, I too often, as I say, microstration like to offer me very empirical papers, I kind of need to have a section with a model to make sense of them. And so often those models are completely redundant them. It's just an interesting statistic. That's it. Just stick with it. It's important to know and that's progress.

Beckworth: Well, that's a fascinating paper, we'll have it linked on our website for the podcast. I want to move on to the next part of the show to your research on what you call reserve ISM. But the idea of maintaining this large stock of bank reserves with central banks in the balance sheet, and using that to manipulate the price level to maintain price stability. And one of the articles you've done several articles on this, but the one that I read closely was one titled achieving price stability by manipulating the central bank's payment on reserves and you co-wrote it with Robert Hall. And you note in this paper that your idea follows the spirit kind of the original thinking of Irving Fisher's compensated dollar idea. So why don't you kind of get this conversation started by telling us what was Irving Fisher's compensated dollar idea? And then how do you build upon that?

Background and Building Upon Fisher’s Compensated Dollar

Reis: Sure, let me take just a little bit step back just to what reserve ISM is and then I'll turn to that. Why reserve is what I mean. I mean, is it's more a fact than a theory. So what I mean is, we just were I talked about the central banks, for better or worse engage in QE and QE consistent. QE had a lot of variety across different countries, but in all them consists of issuing reserves. And so if you look around now, there's just trillions of reserve central bank reserves outstanding being held by banks everywhere. I mean, as simple in the US, if you look across US banks and their assets, 94% of US banks was 15 held more reserves than Treasury securities.

Reis: And so for bands that now make much major asset. And I think we have to understand with these reserve ISM. So reserve ISM was the name I used one of the many, many work and that includes all the work on QE by the way, that we discussed 20 minutes ago on target and said what is the role of reserves? What role do they play? What can we do with them? Because this is new and again I find exciting work because 10 years ago there were barely any reserves, there was only a tiny amount of required reserves and excess reserves in the system. Now we are treated as an outstanding we have to understand what role do they play in the economy and thus the title reserve ISM which is really refers to a very broad line of work not just for myself only a very small part for myself, but all the work if you want QE be and others as part of the on reserve, it's understanding the role of reserves. We have a lot of fears of course.

Reis: The problem here with monitors we have a lot of theories of money while, banknotes with Queen pieces of paper, the Queen drawn on the matter, and why they're treated invalid and not. We I can comfortably less understanding of why is it that reserves issued by the central bank matter. Too many economists would say knee jerk would say, well, that's just the same as money. And it's not actually the monetary base adding up currency reserves in my view is actually a very misleading concept that we should abolish completely because reserves and currency are very different. There was the properties of what they do in the economist, they successfully reserved this, then you ask about achieving price stability and Irving Fisher. Irving Fisher had this idea long, long time ago of saying I think the sport as well imagine that we have a note a fantastic so this is one of the ideas that have that paid a floating rate equal to the real interest rate.

Reis: That is instead of having a piece of paper like a note, like an act notes that are zero interest, what if we have that the currency the unit of account better was a floating rate note playing, paying the interest rate. No, that sounds a little bit like a crazy idea, but he argue that well, if that was the case, then so far as this note, this piece of paper, this assets is escaping your return as soon as the real interest rate, then it must be that because you could just earn … another investment, then it must be that inflation is zero because you can either use this asset which is the unit of account or invest in a real good, and given that their returns have to be the same, then it must be that inflation is zero. So, we start from that those are Irving Fisher's idea. I don't think his idea was ever taken super seriously insofar as we would have thought it would lead to a complete change of what we think money is or not.

Reis: Well, let me involve do is if you want the one inside of this paper, is to realize that, look, we have reserves, they already exist. They don't require any revolution. Again, there's trillions of them out there. And importantly, because of QE because we're at the Friedman rule, because the market for liquidity station, we can choose our remunerate central banks have chosen to pay a fixed novel interested in them that is you have $1 reserves with me, I give you $1 and 10 cents if the interest rate is 10% tomorrow, that's a choice. But it doesn't have to be. So you can just read me the right there in some other way. And so our proposal, in simple terms would be the following. Imagine that for each dollar of reserves that you have to pause it with me, the central bank, I promise to give you tomorrow, one unit of real goods, let's say the real good is apple. So you get one apple for the $1. And you get that's the payment I'm going to give you tomorrow.

Reis: Now, this may sound a little crazy, but think through me is there's nothing stopping center back from doing this, after all with $1 that you deposit the central bank, your reserves, the central bank, and of course, invested in tomorrow buy the apples or whatnot. There's no issue in terms of coming up with the apples. But then what does this do? Well, the insight then comes from realizing that and this is where I think Irving Fisher had it just right reserves are the unit of account of the economy. Therefore, their real value today is one over the price level. That's what it means to be the unit of account. Your value is A over B. Well, but then what is the value of this tomorrow? Well, given that are better even what is the real return on a reserve? Well, the real turning reserve is going to be the one apple gives you today. By tomorrow, I'm sorry, divided by the apples it's worth today. How many apples today will repeat?

Reis: Well, if the limit is shooting today and tomorrow is zero, then one apple today or sorry, one apple tomorrow has to be worth one apple today, but how can it be only if the price levels equal to one? And so that in some sense was economical as very Fisher, which can be achieved with reserves if they and that's the achieving price ability by manipulating the way we pay on reserves. Moreover, once you take this perspective that we have to say, look, this is just about a ruler rudimentary observes. You can go way, way beyond every Fisher. Meaning, what if you don't want to stabilize the price level but you want to have an inflation that's 2% plus 10 minus whatever's happened to nominal GDP saver, like we were talking earlier. Well, then the same logic applies because it says that look, I'm going to remunerate you I'm going to tell you that for each dollar reserves to deposit me, I'm going to tomorrow give you a number of apples that adjusts the real interest rate and to the price level that I would like to achieve.

Reis: Again, though, the forces of no arbitrage the fact that this reserve list gives you the same return as the return of other assets in the economy of other safe fast collected reserves, is going to imply that excess demand or supply of reserves if you want this people want to get more reserves or less is going to imply that the price level has to adjust that is the value of these reserves to adjust. And in doing so you're able to achieve whatever target you have. This is an argument I so far I've made this argument in terms of just forces of arbitrage, what is that observes by setting the payment on reserves and given that they're valued today that's the price level it comes out. So you can make the second most in terms of savings or consumption. For instance, you would say, Well imagine that precisely because if I promise the payment or reserves, it's very generous, well, then everyone wants to buy reserves the value of reserves goes up, well, it means the price level goes down.

Reis: But as the prices go down, then people stop wanting to save so much. So they start consuming more. And again, that's the way it reached. It's an argument that as we show in the paper applies, whether there are sticky prices or flexible price, whether there's financial frictions, or not, especially an argument that this is why, as you will put it, it's falling very much on Irving Fisher, of noting that the unit of account, we've made it to be an asset, again with a gold standard or almost, but what sort of commodity standards you can make the internet account be some good. Because we've made it to be an asset reserves, well then if you manipulate the payment on that, the forces of arbitrage will be a way for the central bank to control the price.

Beckworth: It's a very interesting idea. I mean, going back to Irving Fisher's compensated dollar idea, I mean, correct me if I'm wrong with... What he argued was, you want to change the gold content of the dollar, right?

Reis: So it'd be a way of doing it.

Beckworth: Yeah. So you so you, you actively, change the resource contents of the dollar. And what you're doing is you're taking it to an asset perspective. So the question I have is how do you operationalize that? And your paper you think three different ways but what I took away from your paper was, you would be something very similar to TIPS.

The Reis Proposal and its Challenges

Reis: Exactly. I think the simplest way to offer ourselves without the central bank is saying one pound of reserves gives you one apple tomorrow would introduce a one part of reserves today gives you a certain fixed amount that after the festival target times whatever the price level is then in the future. So just like the way we organize tips, they'll be very simple way to that is mathematically economically equivalent.

Beckworth: Yeah, so with TIPS just for listeners don't know tips for Treasury inflation protected securities, they pay you a real yield real return.

Reis: So here is crucial data that this is not paying your real return meaning, again, remember, it is the keys that it's promising you a real payment. And that's right TIPS. So different TIPS there is that I'm saying I'm going to promise you a certain dollar amount tomorrow. But that dollar amount is equal to some promise I made times the level of the price level tomorrow. And that's the way in which the payment on TIPS as a term. How is this very different from TIPS is that TIPS of course, with given that their payment tomorrow is organized, just as in my reserves, but the price today is pinned down by an auction, whereas the price of a reserve is equal to one always because you never count. That's their real price is what a repeat. And that's why by promising a payment, but not a return, you will must equilibrate. It's the value of it today. And the only way for the value today to liberate is through the price level change.

Beckworth: It's really fascinating. One thing I really liked about it is it gets past the knowledge problem that's you face I can a tailor rule until really have the output gap and output gap requires, in real time real GDP and potential real GDP, which orphan media shows probably isn't the case. And it can lead to confusion like in the 1970s. So I like that aspect of it. My question to you is, have you had, I'm sure you've presented this to central bankers and what's been the reception? What are they? What have they said to you?

Reis: Well, I think so there's one very positive side the reception, which is I think, this papers, one of a few more that have come up and I didn't, for instance, I think you've interviewed John Cochrane and a few others.

Beckworth: Yep.

Reis:      That I think we need to rethink a little bit, how was it the central bank's control inflation? I think the lesser became a little complacent with the Taylor rule working. But the Taylor rule has many flaws as a theoretical principle. And the last few years, if anything, we have convinced central banks that we don't know exactly how to control inflation so well, and so we need to do more theory. So insofar as this is a brand new theory, different way of thinking about how is that inflation spin down, so forget about the pulse recommendation.

Reis: It's just for to some economic factors, the channels through which prices are up and down, these other banks are very happy and warm to receive this, because they want more people to think about how is it exactly what the now inflation. The Taylor rule didn't and should solve that problem because it didn't solve that problem. And so in that sense, there were seven very good. Now second point is if you meant by that is that if I had some central bank, which means the recorder we're going to do this tomorrow? Absolutely not. To be completely honest, they shouldn't I mean, this is an academic paper mind the thing about how is it that the price levels of tournament.

Reis: I think this paper contributes to our knowledge of the madness, which possible determine what's for an alternative that gets you to think gets you to think about how we do right, these balls. I think it's very far from something that the central bank should be doing tomorrow. Again, it's far too if you want original or radical or whatnot. For it to be some, I don't mean as a possible mutation. As the macro economist, I, as many of the reasons I often wear two hats. One is when I write sensible pieces for policy conferences on trying to influence policy, given what we know today. And there's some of the times when I've been your academic work where you're exploring ideas, understanding mechanisms, trying to figure out how things work, maybe down the line, yes, all central banks will do this. And they'll figure out what's better about it or wrong with it or not. But this is a academic paper in that sense. And so it should be read as an academic paper, not as a policy proposal in itself.

Beckworth: Would this proposal be able to get around the zero lower bound problem?

Reis: No. So this proposal has essentially doesn't change this zero lower bound at all. I mean, the zero or meaning it will bind just the same virtual zero lower bound. So let me answer this in two parts. First, it's still the case that if you try to promise a payment on the reserves, that implicitly means that the nominal return the reserves is less than zero, you're in trouble meaning the zero lower bound. So there's also is not a zero but there's also a lower bound on the promise came in on the reserves that derives essentially from the non-negativity of normal interest rates or whatever the effect of lower bound. So that constraint is still there.

Reis: Because remember, David, that constraint just comes from the fact that we have another asset and the economy bank knows that pay zero interest by definition. That's what raise the constraint that puts a constraint on the return on any other asset, any economy, whether its reserves, bonds, or whatnot. So that doesn't change. Which is a little bit is that of course, the policy, the optimal policy, the zero lower bound, much of it in the policy advice we have today relies on tailorable type arguments that say, look, if you just promise far ahead to do a typical once we leave the zero lower bound, you'll be able to control inflation both today as well as in the future.

Reis: Now, those arguments as exactly you put rely on enormous amount of foresight on agents workforce today, all they're seeing is zero interest rates fix and have to believe oh, yes, but I know that in two years, the Fed is going to go back to the Taylor rule of 1.5 coefficient or whatnot. You may think that that's a little bit hard to believe. This payment was ruled has that virtue that again, you can say, well, instead of having to promise this incredible knowledge of what the parameters are not I trading all the way to infinity. If you just think that look, once realities are all about, and we're going to implement this payment on reserve, and use it to implement a price level or not. That seems to be much less demanding on rationality and foresight of agents that today observe zero interest rates.

Beckworth: Yeah, so let me put some other practical challenges out in front of you for this proposal. So there seems to be a consensus, maybe consensus is too strong of a word, but there seems to be a movement of the Fed towards shrinking the balance sheet. A lot of talk about it. I think so there's interest inside the Fed doing this. I also think politically it's probably direction if you look at like excess reserves in the US, which makes it most of their reserves in the US to big parties own them or help hold them, the big banks and then foreign banks. So that's kind of a politically I think toxic recipe right there. So my sense is there's this movement towards shrinking the balance sheet, it may take a long time. But if that happens, does this constrain the ability to use this tool?

Reis: So let me ask that also in several parts. First the answer. The simple answer is no, it doesn't change the setup. This was a big battleship over the small battleship in the end is really the payment to offer on the reserves. In the weather that paint whether there's a trillion reserves per minute of reserves doesn't really make a significant difference to the argument. So let's answer number one, absolutely not it doesn't really change much. Answer number two, though, I think for monetary policy implementation, and generally for using the payment on reserves, whether the way we have it now through an interest rate or through the payment that Bob and I propose.

Reis: It does make Montreal civilization much easier and I think much leaner and I think better if we are the Friedman and the Friedman rule implies it again, I wrote this Last Jackson Hole conference my paper, essentially making this point that with approximately 1 trillion of reserves were at the funeral was a funeral need were satiated the market with liquidity, i.e. the interest rate in overnight private market is equal to the interest on reserves. That's what Milton Friedman defended the central bank should do, it has the virtue that then you can choose the interest on reserves, again, either repayment or some other method in order to control inflation, while potentially using QE or others in order to try to achieve financial stability. It has a virtue that again, you satiate the market liquidity provide this public good.

Reis: And so I defended those and many more arguments on why the Fed's balance sheet should have at least 1 trillion of reserves so that we chew the freaking rules so that we better control inflation. And whatever rule we want to set that we stopped doing this federal funds market, open market operations game of cat and mouse division, a little bit of reserves and having the scarcity of them determine the equilibrium interest rate. It seems to me a better outcome to just have the satiation and that seems required for the research, not once. So I have defended publicly that we should have a balance, we should not go back to the old balance sheet, but we should have a balance sheet that for the case of the Fed has excess reserves in the amount of about a trillion in order to be at the funeral Lenore to set monetary policy, I said, even shopping search.

Reis: Now, but should it be 1 trillion, or sort of before five or 7 trillion or $5 trillion nowadays? On that one, I think there's a hearty debate. There's a debate on different points. You made one point, which is look, banks hold reserves. I would just note relative to what you said that only banks can hold reserves. So has to be the bank, nobody else can hold. I can't hold reserves. So yeah. You said it's the big in the foreign banks. Well, that's because they're the big ones the small banks hold reserves to it's just that they're small. So it has to be the big banks and loser. Although that opens a very interesting question, my view which is led to, again, it's another branch of what I call reservists, which is why not late, let David have an account at the Fed, have him hold reserves.

Reis: That goes into the heading of this whole digital currency debate. Again, it's I think, a very interesting side of reserves. Why is it that all I can do is hold the, if I want to hold it a bonds, the Fed, I need to open a deposit at Bank of America, which then puts them it deserves, that there's a lot interesting debate and discussion on that. But let's but that's the point you made. So on the big banks or banks, I think that's not a question of whether we should have a lot of reserve or not, is whether we want to keep on not letting people deposit the Fed. That is we want to keep it such that only banks can hold reserves. A final point, though, is that are going back. Should we shrink the banks or not? I think there's two arguments there. That point in two directions, aren't when one is when the central bank holds a large balance sheet and it has low reserves. It also holds a lot of assets on the other side. If those assets were all short term risk, plus government assets, then the first store I don't think It wouldn't matter very much.

Reis: I think like said it would make a little bit of a difference because reserves and short term government bonds are not exactly the same thing. But I'm not sure this would be a particularly huge deal. However, what happens, of course, is that a large balance sheet comes with central bank holding long term assets, holding foreign currency or holding in the case of ECB, risky government assets in the form of government bonds of countries, they may default. But that implies that with a large balance sheet comes the risk that the Central Bank suffers large, potentially losses. Well, given the independence of the Central Bank, losses by the Central Bank are not don't lead to automatic recapitalizations by the treasury. As a result, they can really jeopardize the ability of the central bank to either remain dependent or to be able to keep inflation on target. Therefore, what I see as the big risk is rather that the assets that are being held against these reserves are very risky.

Reis: And that's the argument for why I end being closer to the view that we should shrink the balance sheet. Again, not the zero, but the one trailer to get the … But mostly because of the risk, there isn't the assets in the way that compromise the independence, not because of who holds them as in your point, but because of the probability that the Federal Reserve will suffer losses and suffering losses will have its independence and inflation target jeopardize. On the other side, though, David. So let me go put the other side of what's an argument for keeping it high, insofar as the Federal Reserve keeps a bunch of reserves out there and use them to buy even just short term government assets, which as I said, is approximately neutral when it comes to the income risk so far as the risk of losing money on the three monthly bill by issuing overnight reserves is a very, very small one.

Reis: Then, I think there's an argument that says that well, maybe we should have a very large balance sheet, because then we'll we will have the Fed doing essentially crowding out the money market edge, that is half the Fed just provide all the liquidity you want in terms of reserves, potentially go through digital currency. There's no point in there being a money market fund that likewise issues money market accounts and buys three month paper if the Fed is doing it, if you think that the money market fund industry is not, is brings more harm than good to the world, and so far as the externality goes into big to fail, and bailouts and others, then why not have the Fed credit out by having large balance sheet? I think that's a perfectly valid argument to make. And that really is in central discussion of what are the benefits of having a money market fund industry? So that's the way I see the debate. So to summarize, sorry, because I-

Beckworth: No, it's interesting.

Reis: …for a while, I would say that we should absolutely not go back to the old days of zero reserves. That's the one that I should strongly about, or my research tells me very strong, we should have at least one trailer, we should realize that Friedman rule, we should use the interest on reserves, hassle to rule everything, in my opinion, our reserves rule but whatnot, but still the payment reserves is the key monetary policy tool. We should forget about this federal funds market and this open market operations. Those are inefficient, incomplete. inferior ways of setting monetary policy that requires 1 trillion of reserves. Now, should we have more or less, I tend to be on the side of those with people be better to have less to be close just to the 1 trillion because of the risks on the asset side. I think there's an argument that says no, we should get bigger so that we crowd out with me perceived as inefficient fashion activity. And I'm generally open minded on that.

Beckworth: Well, this is all very fascinating to me on several fronts, the whole realizations of Friedman rule, something you studied in school, and here it is in practice. Also mean i think the general point about expanding access to the Fed's balance sheet is another way providing safe assets to everybody, right? That everyone can have a very secure asset during times of stress, but unfortunately, we have run out of time that our guest today has been Ricardo Reis, Ricardo, thank you so much for coming on the show.

Reis: Thank you.