 Hello and welcome to the seventh meeting of our virtual seminar series on central banking and digital currencies Today's host is the wix band and I will now turn things over to our moderator today David Bestin Thank you. So today, we're really happy to have Andy Levin here to present the paper Golden Papers and the paper fetters It's a joint paper with Michael Bordeaux and the Arunimah Shina And he will present and he really doesn't need much of an introduction, but For the sake of it. He's a student of John Taylor. He has numerous of publications in in the monetary economics primarily and on top of as you all know having a very Successful academic career. He's also been extremely successful in in the policy circles with the long-standing Career at the Federal Reserve, so it would be really interesting to hear and this perspective on these issues And he will have 25 minutes to present then there will be a discussion by Todd Keister And he will have 10 minutes to to discuss this paper. Todd is professor at Rutgers And we're really happy to have this this Constellation going so with that said Andy the floor is yours 25 minutes Well, thank you so much. It's really great a really great honor for me to be able to speak to all of you and present this really new work with Michael Bordeaux and Arunimah Shina both of them fortunately are here and They're gonna be active in the Q&A But but before I dive in I just also wanted to say to reach out and say a Shout-out to Gaudi Egerton who was ridgling gonna be the discussant today He's a extraordinary remarkable scholar as many of you know with just amazing combination of familiar with monetary history and with DSGE modeling and with policy. So what we'll look forward to having Gaudi. Hopefully Get his comments sometime down the road wish him well But we'll also really really appreciate Todd Filling in on short notice As discussed it today. So with that, why don't I go ahead and dive in here? So I'm gonna share my screen and hopefully this will go smoothly Okay, so as I mentioned, this is joint work with with Mike Bordeaux and Arunimah Sinha and Just read the red ink at the bottom that these are just our own views So There's a lot of history about the gold standard Mike is really an expert on this work on it since I think For maybe 50 years or something maybe more than that What's interesting about the gold standard is that during the time the gold standard was in effect a lot of distinguished scholars noted at shortcomings Year after year a decade after decade and these are names that would be familiar to many of you Jevons and Marshall and Bixel Irving Fisher Who not only pointed out the shortcomings, but also recommended? improvements to provide a more elastic currency that would provide More systematic price stability of the general price level The distributional effects of the gold standard were also a heated issue Year after year a decade after decade the most famous being William Chang's Brian who was the US presidential candidate in 1896 Referred to the ordinary people who were being crucified on a cross of gold It was a very vivid analogy that's kind of stuck in among the great campaign speeches of all time Despite all that the gold standard prevailed Of course, there were exceptions countries went off of it temporarily, but but generally prevailed as the monetary standard until the Great Depression At which point every country went off of it and no country's ever gone back to it and in the modern era There's very few modern economists who would advocate it now So I think there's a real kind of question here. Why did it take so long? Given, you know, these analyses of key distinguish economists like Jevons and Marshall and Bixel and Fisher Why wasn't it abolished earlier? And one of the reasons we want to flag that is because we see A lot of parallels with the current constraint in our monetary system that arises from the effect of lower bound on normal interest rates And that effective lower bound arises from a similar kind of coincidence historical coincidences technical factors, idiosyncratic factors that weren't really very relevant for a long time It's starting in the 1990s. The Bank of Japan became constrained by the ELB And then later in the 2000s many other central banks, especially in the wake of the global financial crisis The ELB became a serious constraint And there's been a huge research literature. In fact, Gowdy was one of the pioneers in that literature and worked with Michael Woodford Others at the VOJ, Satoma Watanabe was a pioneering research in the early 2000s And many many papers since then about how to mitigate the effect of lower bound Using some combination of monetary policy under commitment like forward guidance Or balance sheet tools like quantitative easing But we even surprised looking through this literature and again people may know of papers that we should be aware of But I would say at the moment It's remarkable to me how little attention has been given by our profession To the possibility of eliminating the effect of lower bound. Can we just get rid of it? And can we do that before we have a major crisis like a great depression? Can we learn the the lessons from the gold standard? Okay, so what do we do in this paper? And again, it's a brand new paper literally We just sent it to Todd about a week ago if I recall And it still needs to be polished further. So any comments people have Missing references we have if they're typos or any other kinds of improvements We really appreciate your comments and feedback But there's three parts of the paper The first part is historical analysis where we're identifying the parallels between The gold standard and the ELB and the gold standard of this, you know famous work seminal work by A very icon green Where he referred to it as golden fetters So we decided what called ELB paper fetters to kind of emphasize that analogy and that's in the title of our paper So that's the first section the second section is To formulate a new DSGE model that we can use to analyze The disparate effects of montage policy at the ELB And in order to do that, we need two key elements Coming from two different strands of the literature. One is the bounded rationality That limits the effectiveness of forward guidance So that the ELB is a real constraint that you can't make way with magic wand and get rid of it And the other is to have a model with heterogeneous agents different types of households In this case, we're going to have a model where There's asset holders and then there are constrained households that basically just consume their current income And then we can look at the effects of the disparate effects of montage policy at the ELB Again drawing on the analogy with the gold standard where we know that there were very important distributional effects of the gold standard and we want to look at those kinds of effects of the ELB And we believe this is novel but again if someone knows about work that's looked at this before we really appreciate your letting us know about it and then the final part of the paper, which is brief, but we think it's important is That it's not just desirable to get rid of the ELB We think it's actually feasible to do it. And in fact, there's an opportunity right now Over these next several years as many central banks are considering digital cash central bank digital currencies That we see an opportunity here to eliminate the effect of lower bound And that that could be done without abolishing paper cash Nor imposing any fees or restrictions on ordinary households and small businesses So I'll come to that at the end of this talk, but let's jump into the historical analysis And I'll try to be brief and then people can ask questions here. I think what you see on this slide Is remarkable kind of historical things I didn't know until Mike Bordo alerted me To the fact that Isaac Newton because I think of him as the apple falling from the tree That he was also the master of the mint for a while in Britain Um that he was really responsible for establishing gold as the de facto unit of account in Britain Prior to that time the silver coins were kind of the unit of account for normal day-to-day transactions and gold was mostly used for international trade But because he made what seemed like a minor shift in the relative prices of gold and silver that Basically pushed silver out of circulation And gold became the de facto unit of account and a hundred years later the parliament adopted That into law so it became disorient and not just de facto Similar kind of story in the United States. Originally silver mostly circulated, but the southern states in the United States Were powerful in the u.s senate and in the 1830s they pushed for an adjustment to the gold silver price ratio So it didn't require Isaac Newton just required some u.s senators And gold became the de facto unit of account in the United States And finally seven years later more or less the u.s congress established that into law In the 1870s germany unified they adopted the gold standard. They defeated france in the prussian Franco prussian war a few years after that france adopted the gold standard And again like is an expert on some of those details. So i'll leave a q&a for him on that part Um If you look at the origins of the elb You see similar kinds of sort of technical factors historical coincidences and so on That of course, we know that corporate bonds and securities and and treasury securities pay interest So then you might wonder well, how come paper cash never paid interest and the answers are basically technical factors That in order to do day to day transactions consumers and businesses want small denomination notes or even coins In those coins and small notes pass hand to hand without any ledger entries So unlike a corporate bond or a treasury security where The the the debtor knows who the who's the owner of that Um, uh security Um, and and oftentimes these paper bills would circulate over an extended period of time There were exceptions for the bank of england where in the beginning of the bank of england They had notes that just circulated in london and would come back to the bank In a few weeks and those actually did pay interest for the first three years of the bank of england and then they discontinued it um, there are other exceptions in the civil war in the united states Um, where there were paper bills issued that where they tried to attach coupons to work very well The main point to emphasize here is that the fact that paper cash was non-adjust bearing didn't matter under the gold standard And the reason for that is because paper notes were claims on gold And gold had a real rate of return that was competitive with other assets Um, in particular the real rate of return on gold was roughly similar to the growth rate of real gp for capital And since paper was a claim on that paper cash was a claim on it. You didn't have to pay separate coupons necessarily on paper cash Only at times when a country or the u.s. Confederacy departed from the gold standard Of course, the bill came a separate issue, but under the gold standard generally There was no need for it What became apparent in the modern era of fiat money is lack of interest on paper money is becoming a problem First of all, it was a problem during high inflation Because then there was a big opportunity cost of holding paper cash And more recently, as I said that elb has become a problem Because that's been a downward constraint on the nominal interest rate But again, the fact that we have an elb is kind of a historical coincidence much like the gold standard Okay, um the gold standard was a severe constraint on mod j pulse. I already mentioned that marshal and vixen and others had emphasized long periods of deflation because The supplier goal wasn't keeping up with demand so the real price of gold had to change But even more importantly the gold standard was a constraint on central banks acting as linear loss resort And that was really what led to the banking collapse in the early 1930s and to the great depression And I just want to emphasize here that the consequences go beyond the economy and the financial system There's an important book by strawman 2019. It's in our references Where he discusses I think the title is the deluge Um how the banking crisis in germany was a key element in the collapse of the weimar republic and the nazi takeover Now for me personally just in full disclosure My great grandparents all perished in the holocaust And so for me personally trying to make sure something like that never happens again And at least that has never linked to any failure of central banks or the monetary system Adjust kind of a core motivation for me in working with mike and aranima on this paper Okay, now there is a question some there's been some discussion. Well, how severe of a constraint is the elb? And I think the simple answer to that is to just look at practical experience over the last decade In europe and in japan even in the united states Um, we had a very slow sluggish recovery took more or less 10 years for employment in the johnmark in the united states to Return to levels that were consistent with full employment And inflation during that period was falling below the fed's target and similar kinds of issues and in other in other economies Um, finally part of our history section here is the disparate effects of the of the gold standard And the the the disparate effects of the elb and just to this is not in the paper yet, but um The new graphs that we've been putting together just to show you here in the united states The um, the blue line shows you the prime age labor participation right now. This is workers um 25 to 54 years old, so they're called prime age prime working years And what you see here is the labor market was very unsatisfactory in the united states in the wake of the great recession From 2009 to 2013 the prime age labor persuasion was dropping Flattened out and finally started recovering During the second half of the decade. So very long slow painful recovery of the labor market Meanwhile because of lower for longer, you see the qe 1 qe 2 qe 3 the first rate hike finally comes seven years after the the onset of the um of the of the of the financial crisis um And during that period asset prices are booming in the united states the wilshire 5000 goes from Which is a broad stock price index goes from 8000 to 32 000 so you can see here pretty obviously from this picture how The lower for longer works well for the asset holders and not so great for ordinary workers This is another set of data from the congressional budget office Showing essentially the same thing with the income distribution. This is pre-tax income So before taxes and transfers And what you see here, this is not even showing the lowest quintile the lowest quintile Didn't improve it all over this period But the middle quintiles are the blue and that basically their income average income went from 72 000 more or less Till around 80 000 so call it a 10 percent increase if you want Meanwhile, the highest quintile the top 20 percent of the income distribution went from around 250 000 up till around 320 000 Okay, so, you know huge increase in their income, which is mostly um asset income reflecting the boom in stock prices and other asset prices during this period Okay, so now i'm going to turn to the model analysis And i'll try to be brief. There's a lot of detail here. Our name is really the expert on this But we start with a benchmark model that was formulated by Fernandez Villeverde and Rubio Ramirez It's actually in diner. So it's easy for us to start with it Lots of basic features that are familiar to you And the two things we add as I mentioned before one is that we add what are called hand-to-mouth Agents literally its households that don't have any assets who can't borrow They don't have any savings. So they just consume their current labor income And that's following a recent work of bill B and co-authors A whole bunch of papers actually written on this that are imported So we're really building on what they've done And then the other piece is to add in Myopia, which is a way of resolving the four guidance puzzle The boundary rationality that referred to here and we're taking the particular approach of gebex his aER paper About a year ago I already might have also looked at other approaches here, but but in this paper we're following gebex And we run simulations that model where we have a persistent preference shock And we're looking at two simple rules one is a tailor style rule Which you can think of roughly as something similar to optimal policy under discretion And then we look at an alternative rule Following Kylie and Roberts that's much closer to an optimal policy under commitment Where you have A promise in effect that you're willing to allow overshooting of the inflation rate Above the target as opposed to the tailor rule where you kind of converge Monotocly to the target the Kylie Roberts rule you get the overshooting And that can significantly improve the outcomes relative to the policy under discretion Okay, so let me show you the simulations quickly and again, there's details here But just to give you a quick overview This is a simulation under a moderate sized shock where you don't hit the ELB You can see the nominal interest rate drops to around half a percent or one percent in that range And where we're just trying to show you in some sense how this combined model, which is the blue line compares to the benchmark of kinsley model the model with The heterogeneous agents and the model with myopia and the simple answer is that the Myopia makes a big difference. This is an important point that kebaks and others have made McCain alcamoor steinsen and Angelitos and Leon. There's a number of papers on this The effects of outer rationality and a market financial market constraints That you've got smaller drops in inflation here and you don't need as big of a response of the interest rate But generally speaking these aggregate implications are fairly similar across these four variants Where you start to see some importance here even for this moderate shock is that in this combined model There's a big difference in the consumption behavior of the asset holders compared to the constrained agents And the reason for that is because the asset holders benefit Right away from the lower real interest rates that push off the asset prices and so their consumption Recovers pretty fast the constrained agents they hand them out. They're just consuming their current labor income and so Eventually they recover too, but it takes You know, maybe twice as long Okay, then we turn to a much larger shock Where you do hit the elb in this case? um It's around three to four years period where you're at the effect of lower bound In the green line here is the tailor rule and the pink is the Kylie Roberts rule and there's some slight adjustments here. So when I say Kylie Roberts and Taylor, it's not the original Taylor rule It's not the original Kylie Roberts, but but something similar to those And as you can see they Kylie Roberts does in fact do significantly better inflation stays close to target Consumption comes back a little bit quicker And in fact the lift off from the zero bound happens faster under Kylie Roberts because You get the faster recovery and faster inflation rate You just have to be willing to tolerate some moderate overshooting of inflation But Kylie Roberts works great Now here's what happens to the individual agents within the model under the Taylor rule Um, and here it is under the Kylie Roberts rule what you're still seeing the same pattern Is that the asset holders? Benefit a lot more from the monetary policy This experience is pretty painful for the um The hint-a-mouth constrained households And that's because the labor market takes time to recover They're not benefiting from the low religious rates that push up the asset prices Now this is just an example here. We're playing to do, you know more simulations with alternative parameters and Variants and so forth, but just to give you an idea here that that generally speaking Um, the ELB is a bigger problem For ordinary workers who don't have a lot of assets Um, it's not as much of a problem for the for the large asset holders. Like we saw in the data They get pretty quick benefits from from the lower for longer policy Now, um, I'm just anticipating here things that Todd may not tend but just Um to be upfront that we can see lots of room for improvement in our model lots of issues that we know about that We haven't gotten to address yet or even look into Because this is a novel model um Assumptions out lump sum transfers Assumptions about the labor market Assumptions about where's the myopia apply And you know serious treatment of fiscal policy. Those are all things to Look into In future work Some of which might have to wait until future papers. It's probably have limits on how long we can make this one Okay, the final question. I promise to say is that we think it's feasible to get rid of the ELB Using digital cash And the principles i'm showing you here at design principles are ones that michael bordel I've already been talking about for five years So i'm going to be very brief here just to say we think it makes sense to have public private partnerships You can have digital cash wallets that are provided by financial institutions. It could be a bank or some other payment provider But the funds in the wallet would be held and reserve at the central bank. So it's perfectly safe There's a centralized ledger. You can make instant payments for free It's efficient. It's secure and convenient and because it's a public fiber partnership It'll protect the privacy of individual transactions Unlike something where the central bank is directly doing that directly with the public But also facility appropriate law enforcement as opposed to something that's completely decentralized where it can be anonymous and Has a risk of money laundering and other sort of problems We think that the the digital cash should be legal tender But we also feel strongly that consumers of business should remain free to use other types of payments Including paper cash and paper cash may start disappearing quickly in the next few years, but people should be free to use that if they want We think digital cash should bear the same rate of return as other risk-free assets And monetary policy could start using the interest rate on digital cash as its primary monetary tool Okay, so how do you get rid of the ELB? Well, there's two relatively simple elements of this Um, in fact, these two elements don't even need to be present during normal times The central bank just has to be prepared to Put these into place. Um, if there's severe financial distress Or there's a severe average shock to the economy Then these two fees would need to be implemented to get rid of the ELB The first is there has to be fees on large balances of digital cash above some high threshold And the reason for that is to make sure that asset holders don't sell off all their private assets and try to ship into digital cash It creates a disincentive Um, we think it's a much better idea than just putting a strict limit or a quota That effectively would ration the digital cash the amount of digital cash would be Indogenous in the system, but by imposing these um, large balance fees With discourage asset holders from from the fire sales We see it as very analogous to a safe deposit box fee that banks already charge The only difference here is they would only apply to very large amounts of digital cash Or ordinary households and small businesses wouldn't be affected by this The other component is you need to apply it a transfer fee for transfers between digital cash and paper cash Again for only for large transfers above some threshold So that this would be totally irrelevant for normal families and small businesses But what it would really do is discourage Big financial institutions or wealthy investors from shifting out of financial assets into paper cash In that sense, it's like an ATM fee except that instead of the current ATM fees which You know apply to a 40 dollar or 40 euro withdrawal from a machine These cash transfer fees would only apply to huge transfers like a million dollars or something Okay, so conclusions Again, the historical analysis We think there's a clear message from the gold standard era that we should get rid of the ELB promptly Let's not wait We can't wait for another crisis where the ELB becomes a really severe constraint That really hurts the economy and potentially has severe political consequences We just cannot allow that Our model analysis emphasizes the disparities of the effects of the ELB and we think that deserves more attention Certainly more Work needs to be done in this direction And finally the practical analysis just underscores the opportunity we have right now with digital cash To get rid of the ELB without imposing any new restrictions or fees on ordinary consumers or small businesses So again many thanks to Todd for discussing this on short notice and we welcome comments from everyone else too I'll stop there All right, thank you Andy for that presentation was excellent and perfectly on time So without further ado, the floor is Todd's for 10 minutes. Thank you Todd you might be muted here. We don't hear you Todd are you there? It would be he switching the microphone Oh, hi, sir How about now? No, finally. Okay. Believe it or not. We did test this, you know, right before we went live and it was working Okay, so apologies for that So I was starting with a with a disclaimer that I'm not an expert in monetary history nor in dsg models with or without bounded rationality But part of the fun of being a discussant is you have a lot more freedom than an author to make speculative comments that may or may not be correct and I plan on taking full advantage of that freedom today So this is an interesting paper, you know, as you saw it covers a lot of ground So I would summarize the main arguments in the paper in with these four broad statements First the physical attributes of the payments instrument be it gold or be it paper money can constrain monetary policy And here there's the the analogy between the period of the gold standard and the effective lower bound Second monetary policy has distributional effects And that means these constraints on policy that come from the payments instruments may fall disproportionately on some groups Third efforts to ease monetary policy further at the effective lower bound So unconventional policy may exacerbate these distributional effects And as a result the effective lower bound might be more costly than you would otherwise think it is And then finally that a new regime based around central bank digital currency could eliminate the effective lower bound without eliminating paper currency Okay, so the paper had these three parts. There's a part about Historical parallels. There's a part about the model and there's a part about cbdc And so I'm going to organize my discussion the same way So let me first start by making some comments about the historical parallels So so in the paper a lot of the discussion of the distributional effects of the gold standard Focuses on this period between the civil war and 1900 in the u.s. The period of the so-called free silver movement Which is a period of deflation at least perceptions the credit is tight and that this is benefiting creditors and unpopular with borrowers So, you know, if I were to think of the elements of a model that might capture my Imperfect understanding in this period, you know, my starting point would be something in the spirit of this old paper by sargent and wallace Um, this is the paper on the real bills doctrine versus the quantity theory at a different monetary history debate But but with important parallels and you could update the model They had an overlapping generations model. You can think of something in a new monitors framework based on logos, right? There's some borrowers think of these as being the farmers. They need inputs to produce Lenders have access to these resources, but the lenders may also want a payments instrument So banks are going to intermediate banks lend funds to borrowers The borrowers use this funds to buy the inputs from the lenders the lenders hold the bank deposits Maybe they use them to transact And then there's also this money and exogenous supply Think of this maybe as being gold and it's growing at some given rate, right? Which is determined by by gold discoveries All right, so that's a real rough sketch But but hopefully from that you can see kind of roughly what would have to follow Suppose we think about stationary equilibria in which both money and bank deposits have the same real return All right, and that's effectively what sargent and wallace did So so then just think if the exogenous growth rate of the money supply is low That means the real return on money is going to be high in the stationary equilibrium And that means that the deposit rate is going to be high banks face a high cost of funds That's going to mean lending rates are high. So it's credits tight Is that good or bad? It depends on who you ask, right? It's good for the lenders and and it's bad for the borrowers in this model And if and if you change the monetary regime, maybe you bring in silver And so the money growth rate is higher than it's just the reverse The rate of return on money and deposits is lower. That's bad for lenders But the rate of return on borrowing is lower and that's good for borrowers So the key point, you know, the reason I wanted to talk about this framework is when I think about that period And when I think about a framework like this, it kind of says well monetary policy faces this fundamental Distributional tension, right? And if you ask should the u.s. Have allowed free minting of silver Well, I don't know it doing so, you know, seems like it would have helped borrowers But it would have hurt lenders and I don't know that there would have that the that people think there would have been clear large macroeconomic gains at least in the steady state And I want to contrast that with the 1930s where again, you know, given my imperfect understanding of the history But my impression is the issue is very different that the main story in the 1930s Is that there were large macro gains from abandoning the gold standard It would increase inflation and loosen monetary conditions like before But now it would reverse this debt deflation problem that had arisen if done early enough It could have avoided maybe a lot of bank failures and so on Okay, and so and then thinking about that so that seems like a different issue and you'd want a different model to think about Presumably there are also distributional effects at work right moving away from gold would help debtors and I guess at least initially It's going to hurt creditors But my understanding is that those are generally thought to be secondary That is the argument for leaving the gold standard in the 1930s was not the need to help debtors at the expense of creditors But rather the need to promote an economic recovery even if it hurts creditors in the short run Okay So with that in mind question which historical episode better corresponds to the to the current period That is do we think about the problems associated with the elb As being that it alters interest rates or asset prices in a way that makes some people better off and others worse off But maybe you know in terms of its effect on total output. Maybe it doesn't have a huge effect You know once the unconventional policies are used Or to you know conversely do we think that it does have significant macroeconomic costs and there are these distributional issues But that maybe the distributional issues. Maybe they are maybe they aren't secondary So I thought the message of the paper could be clearer on this point So so again in the paper a lot of the focus is on the free silver era Which got me thinking about winners and losers from monetary policy But then the title golden fetters I understood to be about the 1930s and the macroe issues So now some quick comments on the model First, you know thinking about heterogeneity So and this is going to be related to the comments. I just made what type of heterogeneity matters today So again, you know in the 19th century period I'm thinking of its borrowers versus lenders and you know and when we think about the poor farmers But the poor farmers, you know, I don't think it was being hand to mouth It was important I think that they were borrowing to buy land to buy equipment to buy seeds and therefore were in debt And were hurt by movements and interest rates So the model has a different focus right at savers versus hand to mouth or owners of capital versus just workers And I was wondering as I was reading to what extent is that choice for technical reasons That is it's a rich model. There's a lot going on Hand to mouth consumers at least they're easy in the sense that their decision problem is easy And or to what extent is that really the relevant type of heterogeneity? We'd ideally want to look at meaning that this issue is very different from at least how I think of the free silver era is not about borrowers and lenders It's it's really about who holds assets and who doesn't and therefore the disparate effects are coming not from high versus low interest rates per se But you know at least directly from from the changes in asset prices second so You know in the model in the model the presence of the effective lower bound clearly is going to lower welfare and it's going to affect Andy showed in the graphs that affects a distribution So so I'm I'm kind of wondering what are the relative sizes of these effects So so if we measure sort of the welfare loss due to to the So if we measure the welfare loss due to the effective lower bound on average in say consumption equivalent terms And if we somehow measure the distributional effects Which of these is bigger or to put it differently to what extent is the effect of removing the ELB largely Distributional or to what extent is it largely about the macro benefits with some distributional impacts? and then then you know the answer that question I I would think of is telling us you know Taking at least the model's view of the effective lower bound which of the two historical parallels I talked about or sort of the more appropriate or the closer comparison Second second comment just about intuition and Andy touched on this on the model is very rich There's a lot going on and and I'd like to understand the underlying mechanisms better So when we think you know one clear pattern that comes out is the consumption of the hand-to-mouth consumers recovers more slowly Following a negative shock. Is that all about asset prices? I mean clearly that's part of it or are there other things going on? Um, I would just I would just like to understand that better Okay, some comments on on central bank digital currency. So You know, it's widely understood that the effective lower bound we say ELB rather than ZLB because the effective lower bound is below zero How low I don't know but certainly below zero But short-term interest rates in the u.s. Have remained positive. Why? So, you know, if we go back 10 years or so in the aftermath of the financial crisis I think the answer was there were a variety of institutional factors that were important Money market mutual funds can't effectively pay negative rates and there was concern they would shut down At least initially at the time treasury auctions couldn't accept negative bids So I'll call your attention to this uh, this old new york fed blog post insightful post by a then new york fed economist That that talked about these points So I don't know to what extent that's changed in the last 10 years But but I want to say at least it's not immediately clear in the u.s. We have a zero lower bound It seems and it's not clear that that zero lower bound is actually related to paper currency Perhaps the fetters are not of paper, but they're institutional and regulatory And and if that's true, how would that fact affect the plan? How would the plan here? addressed it, you know that issue Okay, two more comments one Is central bank digital currency necessary? So so the paper Proposes removing the paper fetters without removing paper money And and he talked through the idea you set the interest rate on cbdc negative and then you impose these fees on transfers to prevent People from essentially arbitration that with paper money But do we really need cbc to do this right? Why not just set the interest rate on excess reserve negative and then bank deposit rates are negative And then you have fees to move between bank deposits and paper money So so, you know agarwal and kimble had had proposals along these lines So I just want to ask what what's different here. Can we remove the cbdc from this proposal? And and would it be largely the same or or not? And if not, what would we lose? I understand there are other reasons you might want to have cbdc and I agree with that But what would we lose in terms of this narrow question? The ability to circum to mitigate or remove the effective lower bound if we took cbdc out of the proposal And last comment, you know the model is used to evaluate the effects of Removing the effective lower bound And then question will the proposal here which involves introducing cbdc and the scheme will that lead to the same outcome? You know, I think the answer is no right if people are using cbdc presumably they're holding less of something else What else? What is that something else? And how is the cbdc introduced? How is the central bank balance sheet going to change? Particularly given that their proposal hears calls for cbdc to earn a market rate of interest It's going to be attractive It seems designed to lead to a different outcome than simply removing the effective lower bound So how would affect how would it affect for example the hand-to-mouth consumers? So given the model, I think it's hard to say so for analyzing the effect of introducing a new payment instrument like cbdc Well, it seems desirable for the model to have in it payment instruments And there's a literature that's developed along these lines to what extent could these effects be combined with your already quite complicated model I don't know, but I think it might be worth thinking about Okay, so I'll end here and and look forward to the discussion All right. Thank you very much Todd for that excellent discussion Should we give Andy a couple of minutes to respond and then we take questions from from the floor Well, thank you Tom. That was really amazing Yeah, how deeply you understood the paper just a few days Um, especially because even our graphs were still evolving as of last night Or any when I were preparing the ones they showed even the slides that we think might be a little easier to understand But Todd got all the key points So just let me just address two or three things that he raised First of all, as you said Todd the model is already pretty complex We thought at one point about like incorporating a gold standard and trying to represent that era And we just realized it was going to take us pretty far afield and not only that but there is a nice paper by Crissel motto Estonia that looks at the Great Depression Um, there's some other recent work. Um, are any remembers? Who who who it was? But you know other people who've looked who is it already man? Eric Sims, yeah, Eric Sims, right has a beautiful paper about the uh, a dsge model Um of what would a gold stear look like if you put it into practice now? um So we could think about it again made me make that would make sense for for a separate paper likewise, we thought a lot about Incorporating a banking system and payments And and some of the other issues you mentioned Todd and we were just worried because dsge models can quickly become A black box even as it is I'd like telling you that like just combining Myopia with the hand to mouth agents is not trivial Aren't even I can probably write through your four papers about all the different You know things that arise in that situation and where you put the myopia really matters too um um Okay, so You did ask why why does consumption from the hand mouth recover more slowly? I think that's an important question. We actually understand well from the real business like a literature The asset holders own capital That's the point right they own the capital in the model and so early on They can cut back on physical capital investment and use you know essentially Use that for for to to smooth their consumption Where the hand to mouth agents don't have that luxury they don't own the assets If their labor income goes down because they're working fewer hours then they have to consume less and so part of the The consumption smoothing that you're seeing for the asset holders is is definitely coming through that mechanism but then that's amplified by the The drop in religious rates from the lower for longer policy um You raised a question about you know, um Why is the us never gone below zero and I just want to emphasize here um early on In fact, uh, there was a bank of Finland conference um in early 2009 And there's all the papers the federal reserve staff wrote Um at that period or now in the public domain Um, there was thought that maybe you know money market funds would just disappear for a while and come back once the crisis was over I worry that the extent to which The federal reserve has never gone below zero reflects a very strong influence of banks At the fair the the the commercial banks own the regional federal reserve banks. They literally own them. They have quite two-thirds of the directors um The the banking system is very influential in the United States And banks were worried about cutting rates below zero and so that was the major reason it's never happened So you said institutional regulatory maybe but also probably political and institutional factors as well And then the last question you raised was do we need digital cash to get rid of the yield B and I think We could brainstorm a little bit more if If a central bank ruled out the possibility of adopting digital cash for other reasons It would be great to think hard about are there other ways to get rid of the yield B I think our point in this paper is for many central banks that are seriously considering the design of digital cash right now anyway Is to take advantage of this opportunity to get rid of the yield B We think the motivation for doing that is critical We want to make sure there's never another great depression that caused by monetary constraints Um, and now is the chance in these next few years to make sure something like that never happens There could be another great depression for other reasons But let's make sure it doesn't happen because of an arbitrary constraint from a monetary system. Yeah Can I jump in Hello. Yeah. Okay. Just a history question. Yeah Todd you're right. I mean there there is a difference between the 1890s and And the 1930s and and you're right. It's the 1890s It's a it's a distributional issue And in 1930 primarily macro so but and we used in a sense that the golden fetters analogy is from the 30s And it it is a constraint. So that's why we we had both of those things, but you're right It's it's really a distribution story and that's why we emphasize that All right. Thank you. So we have one question from the panel panel side Raphael Hey, thanks very much for for this this great presentation and discussion. I just um Coming really with a policy hat on I I sort of I I sit in, you know, all the various work groups. We collect data on where central banks are going and sort of Actually there the trend is really It goes more in the other direction that sort of two years ago we were discussing possibilities for monetary policy and and that has come really Lesson lesson to focus where it's really sort of the the microeconomic advantages of maybe Having, you know, a better handle on on the market structure of the payment system How to to sort of frame privacy frameworks in a payment system And and and and and these considerations have have come much more to the fore And and the central banks have have sort of agreed on certain principles and actually the the riggs bank is is is part of of that discussion and Where to to you know as part of sort of the notion to do no harm There is an adamant emphasis on sort of that cash will not be phased out and that the systemic implications Would tend to be limited So and and obviously right with with these measures then you will as as a corollary also have very limited effects on the effective lower bound Now I just wanted to know sort of If if if you think about sort of where central banks are coming and that is sort of the political economy How could you make it sort of if you still say no no we we want to sort of maximize the sort of The margins of the effect of the margins we have in mind and But on the end sort of we know what's sort of politically feasible What would be sort of the The measures the central bank could achieve that well, I think that um but I mean as michael bordell, I have been giving our presentations on digital cash over five years. We've gotten lots of feedback One part of that definitely was the extent to which many central banks have been approaching payments for many many years as essentially microeconomic issue And that it's urgent actually to talk, you know for having more conversations between the monetary department and the monetary policy department and the payments department The the I think it's also urgent to design digital payments That don't create new restrictions and new fees and new burdens on normal consumers and small businesses I feel strongly about that and I think the design we put forward in this paper You know satisfies that that that objective um I think it's critical that we have a monetary system that works well for everyone And I think the problem I showed it to you as clearly as I could over the last decade is that That people at the top end of the income brackets We're doing really well during most of that decade But when you talk to ordinary families in the united states, they didn't think so And you can see why because because those are they're not exactly hand to mouth But there's a lot of american families that didn't have a lot of savings that didn't own You know significant bundles of assets Um, and for them it was a very very long painful slow recovery And I think her question here is if it could get rid of the ELB um, maybe monetary policy would be able to promote a more You know more rapid economic recoveries and the the political benefits of that would be would be very significant Okay, so we have uh, antuan Yeah, so I think my question is uh somewhat along the line of what rafael asked so your paper relies a lot on sort of historical analogies And one thing you could look at is there are a bunch of countries that have gone slightly negative and and see what how they like it and so um I had a minor paper by champiadontein who who's an academic and and and was under governing board of the swiss national bank uh, and and he made a point that he doesn't think deeply negative rates are politically feasible and and In switzerland negative rates are incredibly unpopular. So so I don't know if So it's so it sounds a little bit like Your story would have to be well if we if we go negative a little bit people will hate it But then if we go negative a lot people will love it and and and I find that uh, it's an interesting argument So I'd love to hear what how you think of it Actually, you know what i'm i apologize because I can see now that the presentation didn't cover this We really need one more slide um, uh And probably we can expand the discussion of the paper too. So let me try to be very precise here We don't think that you need negative rates for 90 of the population Okay, for all ordinary households and almost all small businesses that have that they have Bank accounts with with a thousand dollars or a couple thousand dollars a few thousand dollars or a small business that has a $50,000 bank account You can just put a lower bound of zero on all of those And in fact that is essentially what the ecb and the bank of japan and some other central banks have done for practical purposes They've tried to insulate ordinary families and small businesses the reason for having negative rates Is risk-free assets like treasury securities and triple a corporate securities Need to be able to drop in nominal terms well below zero And the reason for that is because you want the risky rates when there's a spread between the The risk-free rate and the risky rates When that's spread widens you want to keep the risky rates from going way up And the way to do that is to allow the risk-free nominal rate to go well below zero The only people who care about that are the asset holders who own millions of billions of dollars in assets Okay, and you want to drive their risk-free returns below zero So they're continuing to be willing to hold the commercial paper and and the corporate securities And and other kinds of things At moderate positive rates, then you have a much quicker smoother recovery of the financial system and the real economy So I totally hear you. We we're not calling For steep negative rates. In fact, maybe not for negative rates at all for most of the population But the central bank should have the ability to cut rates further below zero On that margin And what I understand from reading papers gouty eggerson's written some of them Is that even in europe even in the swiss national bank and certainly in japan The fact that there's a lot of paper cash is what creates the frictions as These some of these rates are going below zero Starts to mitigate even the benefits of cutting rates further and that's why central banks haven't gone any further than they did Okay, so we have three more questions. Uh, so try to be please try to be brief russell Yeah, but maybe boy can ask first. So I I would might tell in the end of the line. So, uh, please go ahead John Now the sun works Yeah, no, it works Just a short observation here and just in line with what and they said that sweden had negative interest rates For about five years and we could see that they were not passed on to households, but only to Bigger companies, etc. So that was Market uh decision by banks, I would say and that was also without the cbdc. That's all. Thank you Thank you, uh, dick Yes, hey, thank you very much Just to make sure I understood and did you suggest that the interest rate on cbdc should be equal to the risk-free rate Or equal to something like a deposit rate in the current system And the second and and why probably some treatment type argument that you have in mind There is there some some other story that you have in mind and and the second thing is I'm um, I'm also a bit confused about these kind of market segmentation like the ecb suggests We could particularly implement and that you also seem to have in mind that some people pay a certain interest rate Or receive and the others have a different than some people have fees and some others don't have fees If you think about this in the international context, um Exchanging one cbdc into another cbdc from one currency into another the marginal investor who would be Exposed to some interest parity conditions. These would be the guys I I suppose With a high volume and therefore the one subject to the negative interest rate presumably and subject to fees or subject to no fees I could you clarify that please Thank you Okay, great. I mean, this is great. Dirk. I hope you and I can talk offline sometime Because you're asking deep questions Just just briefly I think that the fees that we're talking about may be totally unnecessary during normal times That that everything could just be instant and free Okay, the logic of these fees is that in the middle of a financial crisis You want to make sure that people don't Have a fire sale and try to shift all their assets into digital cash and that's why You'd like to push interest rates below zero and so instead of That it may be possible to do this like with a safe deposit box fee. That's the analogy that we've come up with here That during a financial emergency Asset holders can move into digital cash if they want. There's no quota Which some people have suggested we think that would be a mistake what we think is better is a pricing mechanism where the central bank can discourage fire sales By imposing a larger safe deposit box fee and there's externalities to doing that So there's a very strong public motive for it. Again a lot more analysis to be done on this is just kernels of ideas I think that the the reason for the transfer fee is to discourage people from withdrawing out of digital securities into paper cash So I my instant answer which could be mistaken, but my instant answer is You don't necessarily need any fees for moving from one Say from a euro digital euro to a digital dollar or a digital swiss franc or a digital e-crona Okay, those could all be free instant all the time even in the middle of a financial crisis even in the middle of a significant downturn So that all the benefits of the the free international transactions that we much, you know more You know secure and efficient In a digital currency world Okay, the only fees we're recommending here is If you want to take your money out of the digital system and put it into paper cash And if you want to again, we're not going to prohibit people from holding paper cash We're just trying to discourage a huge financial institution from building a warehouse in the swiss alps and you know, and storing billions or billions of of paper bills there All right, is it russum Yeah, I'm I'm thinking of who who are really constrained by the ELB So maybe talk mentioned some of it So, uh, there's a paper by erics grandson saying, okay, uh, we look at the Fed So we are not that much constrained. So, um It's one view. So and another view is a We talk about okay, the band may be constrained by the zero deposit rate It is true But we also see in the core reportator the fee income of the bear increase a lot So although they cannot avoid CEO lower bound on deposit rate, but they can charge fee income So there's another way to go around it. So, um, that make me want to work hard way a little bit But uh, since we have a hack model heterogeneous agent model One reason people talk about the cbdc is now the central bank can trade directly to the open market operation to people So maybe they can trade directly with the hand to mouth model and inject liquidity directly into their own pocket And maybe there is some another first order, uh, consequence of doing cbdc So, uh, with cash, uh, the central bank cannot do it But since now there may be some everyone have an account at the fair realize, uh, the cbdc It may be become feasible. I'm just wondering, uh, What do you think maybe that might be are doable in the model and may have some bigger consequences? Okay, so Remember we're advocating a public private partnership here. We're not advocating the central bank to start opening accounts directly with the public But even under this public private partnership If if you have a low income person who's receiving unemployment insurance or other benefits They can register their digital cash wallet with um The treasury department or the commerce department whoever is paying those benefits or even the state unemployment office And then those payments get deposited directly into their digital cash wallet Okay, it might be much more efficient What we had the problem we had last year during the pandemic was um, the government was sending out benefits to people in paper checks That took weeks to deliver and if someone had moved, um, or or was even just staying with with friends and relatives Then there's an additional delay before they finally get it And then they have to find someone who was willing to cash the paper check And so it it really was difficult for a lot of families that were in need at that time in small businesses Um, so we think setting up this kind of public private partnership with digital cash wallets would help But it's probably beyond the scope of the law The other thing I just want to emphasize here is Eric's wants into the esteemed colleague of mine, you know, I work together literally offices next door to each other for many years at the fed But if you look back at Eric's analysis, he's looking at the extent to which Some fed for guidance announcements move policy rates by five basis points The fed's conclusion from that was well, we can just make bigger announcements We can do a bigger QE And that may be true, but the fact of the matter is when we look at what happened over this last decade It was a very very long slow recovery for of the of the job market and long creative shortfalls of inflation So it certainly looks like the fed was constrained and likewise for the ecb likewise for the bank of japan Um persistent shortfalls of inflation from their target But even though they were trying to use other tools as actively as possible, so I I this idea that the elb is not a real constraint. I just I have trouble getting it But again, Russell, maybe you and I could talk offline Help me see it more clearly. I really appreciate your feedback All right, thank you. So we have five minutes overdue But before we close I want to offer the floor to michael or aranima if you have anything you want to add so Okay. Yeah, these these are really these are really great comments We can we can we can deal with these historical issues. Okay, because I think as I said Earlier, I think that both aspects gold and fetters are relevant Okay, both the macro effect of the great depression because as andy says we don't want that again and the distribution effects So I think that we can do that going forward um, and the other Comment is about a negative rates and I give this paper all over the place when I get the same comment all the time I also get the comment about privacy, but I think that I think I agree with andy on the case for cbdc That it can be it can be welfare improving not just in the sense. We're talking about here, but it can make It's an efficiency argument And it's a transformation That we've seen the kinds these kinds of transformations we've seen before in history The origins of banknotes and central banks Okay, paper money. These were major transformations that had social savings. And that's how we see is cbdc Thanks Thank you for the comments and Todd that was a really excellent discussion And so I just wanted to just circle back to the Russell's point about The you know who the ELB matters for there is work, uh, you know Jordi golly has work that talks about the relevance of the ELB and so on I think what we are trying to focus on here is um, a consistent ELB episode and how that's going to have Distributional effects across these two type of agents or you know other types of agents And that we think is some That that has been perhaps not been focused on as much in the literature But your points about you know, how would these be getting and helping the Hand to maps and so on are are very well taken. So thank you Okay, so with that we thank the speaker as well as the discussant for excellent contributions and we end the session Thank you. Yeah, thank you again to andy Todd and David and All of you right for the presentation and thought-provoking discussions We hope you will be able to join us again next month When the feed effect will host our next session So next time we'll have a Marcus Boone Maya talking about Platforms and tokens Charlie Kahn will discuss and Daniel Sanchez will moderate the session So until then I hope you have a pleasant day and a nice weekend And stop