 We are now moving to the next session, Marco Basetto from the Federal Reserve Bank of Minneapolis will present. His research focuses on game theory, macroeconomics and the design and consistency of macroeconomic policies. And he has made an attempt, as you will see, to look at the complexity of a monetary union with 40 budget constraints like the European Monetary Union. And so I'm really curious and fascinated about this presentation. So Marco, the floor is yours. Thank you. So I was trying to get rid of the side panel on Pedia. Here we go. Very good. It's a great pleasure to have a chance to present my work here. And I want to mention that I'd be happy to entertain questions as they go longer. So, Klaus, if you see a question that fits well with the floor, feel free to interrupt me and then we can have a longer discussion afterwards. But of course, I'm not going to monitor the chat myself while I talk. Okay, this is joint work with Gerardo Carraccio, who's a PhD student at UCL. He's currently on the market. He's a fantastic guy. If you are on the market for a junior market, you should definitely take a look at his file. Let me see if I can move this because it's not moving for me at the moment. Very good. Is it moving for you? Can you see my second slide, Klaus? Yes, very good. Excellent. Thank you. So, what's the motivation for this project? There's a lot of research that has been devoted to the issue that monetary and fiscal policy are connected by a common budget constraint. I tend to think about one of the main contributions as being Sergeant Wallace's present monetary arithmetic that was recognized by the Nobel Prize Committee. But, you know, when every time I present, Sergeant sort of thinks that, you know, this issue goes back centuries and that's true. Certainly, you know, even with the creation of back of England originally centuries ago and even before I was well understood that central banks played an important fiscal role. More recently, the fiscal theory of the price level has devoted a lot of attention to the question of how fiscal policy can, in some dimensions, help in determining the price level. So, making sure that we're not subject to the whims of expectations and that there is an anchor that will make sure that prices remain stable in some way, but at the same time, potentially creating problems for monetary policy. And let me remind you, by the way, that everything that I'm saying today reflects my own views and not those of the Minneapolis Fed or the Federal Service System. So the fiscal theory and unpleasant monetary arithmetic emphasize the single budget constraint. There's been more recent work trying to split the budget constraint of monetary and fiscal authorities and looking at potential implications from thinking about how this common budget constraint arises in an environment in which the two partially have a little bit of independence, maybe in their own budget constraint. And this was notably started by Sims. I have some earlier work myself with Todd Messer, and here Paolo Benigno has work and here this heading that I picked, New Style Central Banking, is something that Ricardo Riz came up with. He has a good taste for a catchy name, so I'm going to take it from him. Now the question for me, so the next step is, so the New Style Central Banking was about, say, the Fed and the U.S. Treasury. There are two agencies, to some extent, you can argue legally the Fed is not a government agency, although the board is, but that's certainly heavily regulated by the government. So there are kind of two government sanctioned entities and they're interacting. Now things become a lot more exciting in some dimension when you start thinking about the Eurozone, and that's why Gerardo and I started thinking about this project. In the Eurozone, it's not just one central bank and one government interacting with each other, but there are actually 40 players. There are the 19 national tragedies, plus the European Union, the European Union is a relatively small fiscal player becoming bigger by the day, with some of the issues that Evi talked about. Then that's the European Central Bank, but also something that may be a little bit underappreciated is that there are also 19 national central banks and we'll talk about today a little bit about what they do or they don't do. So what's chosen centrally and what's chosen at the level of the national central bank, but what's going to be very important is that each of them will have its own, does have its own budget constraint. So, you know, there isn't a single central bank with a single budget constraint in the European Union, the way there is, say, for instance, in the case of the Fed. Now, technically, the regional feds have also their own budget, but everything is shared at the end of the day, so that's fairly irrelevant. There are some minor complications, but, you know, everybody understands that if the New York Fed made losses, say, on made in lane, that was one of the things that was on their books, they would be covered by the system as a whole. It's less obvious in the case of the ECB. And so the key question for me is going to be twofold. One is kind of more technical. How does senior age, so the profits from running a central bank flow from the monetary authority to the budget of each country, because ultimately these funds end up at the Treasury. The profits are admitted to the Treasury and the question is how does that happen. But the second, maybe more important question is who's paying if a member country defaults on its debt. So, through various programs, the Euro system, so the ECB along with the national central banks have been buying a big amount of government debt. That government debt may be subject to default, hopefully not, but you know, Greece did default. And the question is who's going to bear the cost of that, which taxpayer will have to prop up money to account for the fact that there's going to be less profits coming from the central bank or maybe even losses. So, in my talk, I'm going to focus on two programs, the public sector purchase program and the pandemic emergency purchase program. Why these two, because these are the programs where the risk is not supposed to be shared for many of the other programs. The risk is shared explicitly at the level of the Eurozone. So it's kind of closer to the common budget. That's one central bank. If you want with a common budget constraint, there are still 19 treasures in the background, but you know, they're getting just their share of profits or potential losses. For PSPP and PPP, the rules are quite different. The rules work as follows. 20% of the size of these programs is undertaken directly by the European Central Bank at the European level and for 10% they're buying supranational bonds for 10% they're buying national bonds. These are common. So the risk on this is common shared according to the capital key. That's not what I'm going to be focusing on. I'm going to focus on the remaining 80%. For the remaining 80%, each national central bank is buying their own tragedies bonds. So Bank of Italy is buying Italian bonds. The Bundesbank is buying German bonds. Bank of Portugal is buying Portuguese bonds and so on and so forth. The national bank is buying Finnish bonds. The risk of this 80% is not supposed to be shared. It's supposed to remain with the national central bank that bought the bonds. So if Italy defaults, and you know, I apologize if I'm picking big countries, but you know, I did mention Portugal and Finland. I'm going to try to use Italy and Germany as my main example, partly because I'm originally from Italy. So if Italy defaults, it's Bank of Italy that should remain on the hook for those losses that are going to rise on its books due to the fact that now those assets are worthless. So the question here is going to be to what extent that holds in a world in which really at the end of the day there's just one consolidated budget constraint. So let me go through a little bit in practice because this is going to be somewhat important on how this works in practice. So there are two possibilities. So I'm going to use Bank of Italy as my workhorse example. And so there are two options. They can buy these bonds from they're going to buy them from the private sector. This is going to come typically from a bank or you know, it could come from somebody else who has a relationship with a bank. And the question is where do these, you know, when Bank of Italy buys, where are these the counter payment for those bonds going on and up. If they're buying them from an Italian bank, then what happens is Bank of Italy is getting acquired in the bond and it's issuing reserves. Of course, Euro reserves are in euros. So that has to be a mechanism that makes sure that Euro is a Euro, no matter whether it's Italy or Bank of Italy or Bundesbank. But technically from the budgetary perspective, these reserves are the liabilities of Bank of Italy. And so that's where they sit on the balance sheet. Things are a little bit different. If Bank of Italy buys the same bond from a German bank, then Bank of Italy acquires the bond. That's not different. But because it's a German bank, it's a Bundesbank that is issuing the reserves that are going to show up on the payment for this asset. And then of course, there has to be some mechanism for equilibrating the fact that now Bank of Italy has an asset and the Bundesbank has a liability. That comes in the form of the target two balance. So what happens is Bank of Italy is going to acquire a target two liability that matches the asset that they just bought, the bond. And the Bundesbank will get a target two asset that matches the liability that they acquired, which is the reserves that they issued to the banks. All these target two flaws are netted out at the ECB system. So technically those are liabilities and assets against the ECB. As you will see in my model, I will short circuit that. It's somewhat important for netting purposes, but it's not important for the economics of what I'm going to talk about. And so one of the questions here is how is this going to interact with the target two system. So the target two system pays interest, although not currently, but in principle it pays interest at the margin of refinancing operation rate, which is the top of the corridor set by the ECB. And then Bank of Italy is going to have to pay this interest. But if the interest on government that is above that rate, which currently is then that profit is profit for Bank of Italy. If it makes a loss either because the interest rate is below or because it suffers from a haircut that would show up on its balance sheet that way. Positions I'm talking about are big. So that's going to be important. We are currently thinking about how to design experiments to match this magnitude. I'm going to show you an experiment at the end and you see that the effect there is much smaller. So that's a question of how we account for properly for this magnitude. So currently the Euro system is holding almost 40% of GDP of Italian debt. And the target two liability of Bank of Italy against the ECB is 30% of GDP. So it's pretty big. So this target two liability has two key characteristics from our perspective. So this is variable rate. I told you it varies with the corridor set by the ECB. But it's that of infinite maturity. Unless a breakup of the Eurozone happened and people have studied this. That's a big literature on target two balances thinking exactly about what happens with a breakup. But that's not going to be on the table for me. I'm not going to be entertaining a breakup of the Eurozone. I'm going to be studying what happens if the Euro remains as it's currently supposed to. So in that case, this debt never becomes due. We remain on the books of that in the case of Bank of Italy, credit in the case of the Bundesbank. We remain on the books there. It may grow and grow. And that's the other aspect that is interesting. There's not a limit that that can go to 600% in principle to 600% of GDP. And there's no automatic mechanism for why that should ever be repaid in one way or another. Of course, it's still accruing interest. Now, before quantitative easing, this is something that the literature has addressed. That's a camp that says this makes total sense. It's just a counterpart for the reserves that are issued. So yes, if the reserves in Germany blow up, then that's this matching asset that helps the Bundesbank pay the interest that they would have to pay on all these reserves. And that's the target two system is at the heart of making sure that a Euro is a Euro, no matter who's issuing it. Because the moment you put a cap, then there's a question, how does what happens if Bank of Italy hits a cap and cannot issue more euros to its banks? Or vice versa, accept payment. That's the more relevant one and transfer those funds to a German bank. Things become more complicated now because this risk that is coming from buying all these assets is not supposed to be shared. When it was shared, the asset side was kind of boring, and then it was just a matter of giving up properly the payments on the reserves. Now it's different. So now it's time for me to get to the model. So I'm going to have two countries by two because, you know, this is, we want to keep it simple. We don't need 40 countries to make a theoretical point. So the countries are going to be called A and B. Think of A as the country that never defaults and be the country that might or might not default. They each have their own country and sorry, of course, their own treasury and their own national central bank, but the national central banks of country A and B are joined in a currency union. What does that mean? I'll show you in a second. It means that a number of decisions will be undertaken at the level of the currency union. So in particular, the obvious one is it's not each individual central bank that decides how many bonds of their own country they're going to buy. So Bank of Italy cannot go and buy 300% of GDP of Italian debt, not that there is that much, but whatever, you know, 100% of GDP of Italian debt unilaterally. No, that's chosen centrally how much that program is buying. I'm going to abstract from the European Union. I told you it's kind of a small player. I'm also abstracting from the CB because I want to focus specifically on this issue of distribution between national central banks. And the only role there for a central system is making some decisions. I'm going to be explicit about that. So now I'm going to, what I'm going to do is I'm going to look at budget constraints for all these players. I'm going to focus on who's making decisions and looking at various scenarios about how those are going to play out. So here's to begin with the flow budget constraint for a super simplified, of course, to keep things understandable for a national tragedy. So a national tragedy comes into the period with some debt, BIT minus one, which is due. If they repay it in full, that's great. They might default. So it is an indicator function that says that it's equal to one. If they default, then there's some haircut delta that they're going to impose on the creditors. How do they repay this debt? Well, they can raise taxes. I think this really is the primary surplus. I'm not having explicitly government spending in there. So this is the amount by which taxes exceed the spending. Of course it can be negative. It can be a primary deficit. SIT are remittances that are going to come from the national central bank. So these are the profits that are being distributed from the national central bank to the tragedy of country I. And of course they can borrow more. Here government debt is all one period debt. That's because I'm focusing on default risk. My earlier paper with that method was about maturity risk, mismatching maturities between the assets and the liabilities of the central bank. And so there, of course, long term debt is important, but not for the considerations I'm going to make today. So for simplicity, I'm going to have one period debt. As I mentioned, country A never defaults. So it will always be zero country B may default. And of course the interest rate you can see is indexed by I by country I so that can be A or B. And that's, you know, we reflect the risk of default. That was the fiscal authorities. Let me show you first the budget constraint of the Euro system as a whole on the monetary side. And then I'm going to break it up between the two countries. What does the Euro system as a whole do? The main liabilities are going to be currency circulation, MT, so they can finance themselves by issuing more currency, or they can finance themselves by issuing more bank reserves, and those are going to be the XT. Bank reserves may pay interest. It could be positive. It could also be negative as it currently is. With these funds, they're buying bonds of country A, they're buying bonds of country B, and they're potentially buying private assets. You know, think about before all these programs were realized, one of the main assets on the books of the Euro system were loans to banks. And those are still to some extent, although with a magnitude of reserves, that's a little less prevalent now, but that was one of the main factors before. So we're going to have it. I'm going to abstract from default risk on those so you can see those pay interest at the same rate as the risk-free bonds. That also means I'm abstracting from the liquidity reasons for this version at least of the paper that may drive a wedge between, do drive a wedge between the interest rate on German bonds and on private assets. And then of course, the Euro system pays dividends to the two countries, country A and country B. So this is period by period. You can sum up all these periods and compute a present value budget constraint from now to the infinite future for the Euro system. And this is what it looks like. It says the Euro system is currently sitting on some assets, bonds issued by country A, bonds issued by country B, the private assets. It has some liabilities on its books, currency and bank reserves. And to the extent that assets are more or less than liabilities, what's on the books? Well, they're going to pay dividends. Think of them as dividends. These are sign language transfers that they're paying to country A and B, not in this period, but in all future periods as well. So that's a present value. B0S is kind of stochastic discount factor. That just means it's a way of taking into proper account the discounting. And they're also making profits to the extent that government bonds pay positive interest rate and currency pays zero interest rate. That's one source of profits now and for the indefinite future. Of course, you want to discount all those profits. They're part of the assets of running a central bank. Same for bank reserves to the extent that bank reserves pay an interest that is below the interest on the assets that the Euro system has on its books. So that's another source of potentially another source of profits. Now there's a term highlighted in red. This term is what happens captures the following. So what happens if the Euro system rolls over? It's that for the indefinite future and never never asks the treasure is to pay it back. That's something that we usually don't think about. And that's a good reason. If I'm a private individual right then the European system. I don't want to roll over my assets indefinitely. Eventually, you know, I want to eat some some of it. That's no point in holding assets and never ever ever touching them again. Normally, if we think about a central bank, it's a little less clear because kind of the central bank and the treasure are united in a single budget constraint. So does it really matter if they do? And the answer is if there's one treasure in one national central bank, the answer is no. And I'm going to show it to you right now. Because if you consolidate the fiscal authorities of the Eurozone, that's a matching term here. So it doesn't matter. Think of it this way. It doesn't matter if the ECB remits its profits period by period to the Treasury, or it just accumulates ever expanding bonds and never asks the Treasury to pay them back. It's the same thing. It's resources that the Treasury does not have to come up with with taxes. And so normally when we do one national central bank and one national treasury, we neglect this term because it's kind of without loss of generality to think that these profits are paid out. Things are different now because there are two countries in my model. So the question is which country's debt is blowing up over time matters? If the ECB holds only one type of those assets, that's the Treasury that gets to expand its balance sheet. And so that's part of what's going to happen. So now it's time to break up that Euro system into the national central banks. So the only difference here is thinking about who's choosing what and what does break up means. So we have currency issued by the Euro system as a whole. That's chosen. So MIT is chosen by the Euro system. How much currency there is out there? The only thing is it's a liability that is shared. So it's shared according to the capital key. For bank reserves, likewise, the amount of bank reserves is another thing that is chosen at the ECB level. Now exactly the split is not chosen at the ECB level. That's going to be driven by the demand for bank deposits that could be asymmetric between Germany, Italy, Finland, Portugal, Ireland, and so on and so forth. But the overall amount XIT is XT, the sum of the XITs is chosen centrally. Now that's this tau IT that precisely captures the fact that if there's more demand of bank deposits, that's going to mean the country party of that is going to be a target to asset. And so tau IT is the target to balance for the Bundesbank that will make sure that it will somehow restore symmetry to the extent that the differences in XITs are driven by differences in. In the demand for deposits. We're going to be interested in the counter party, which is what happens if there are differences in the assets of these national central banks. And so let's take a look at the assets. The assets are the bonds that are bought, so B bar IT, that's the bonds of country I bought by the national central bank of country I. And those are potentially subject to default. That's going to be the key question I'm going to be after. Now the amount B bar I is under control of the ECB system as a whole. The size of the PSPP and PPP is chosen centrally. What's not chosen centrally is the default. So that's one of the things that is under control of each individual country. The other aspect is how these profits are distributed out. The timing of those profit payments and potentially the amount of those profit payments is under control of the individual central banks. I'll come to that in a second. So those are the two elements, the default and the seniority remittances that are under control of the national central banks. That's where everything is going to play out. Now, where's the valve? So suppose the haircut delta is imposed. Where is what's going to happen? And that's the question I'm going to come after now. So what I did here is just I rolled over again that period by period budget constraint forward into a single present value budget constraint. Recognizing that if there's a default, say by the Italian tragedy today, that has an effect on the assets of Bank of Italy and that could mean that various things are going to happen either now or in the future. It doesn't have a loss, doesn't have to be covered immediately, but it has to be something that makes up for the shortfall. So let me first tell you what is supposed to happen according to the if the risk is not to be shared. So the risk of this loss has to remain with the Italian taxpayer. Well, the answer is simple. What should happen is that the present value of transfers from Bank of Italy to the Italian tragedy should go down. So the red highlights this adjustment. So there are fewer assets because there's been this haircut. So now Bank of Italy realize the loss and their holdings of Italian debt. If they're going to remit less profits to the Italian tragedy in a matching amount, that's the end of the story. So that means, you know, the Italian tragedy has imposed this loss on some private creditors and that presumably, well, that definitely will not have to be paid by taxes. But to the extent that he imposed losses on Bank of Italy, that will still come from taxes because it just means that Bank of Italy will remit fewer profits and therefore that will have to be covered with with taxes. This is going to work fine if the losses are not too big. Things become more complicated. If you get to the point at which you need these transfers to actually go negative, which means you need a recapitalization from the Italian tragedy to Bank of Italy, because the assets, the remaining assets including future profits are not enough to cover for this loss. If that happens, then you would need a recapitalization. The question is, would that recapitalization be forthcoming? So currently, the rules for how these profits are distributed from Bank of Italy are that 60% of its profits annually are emitted to the tragedy, minus a little bit of that goes to the banks that are the technical owners of Bank of Italy, but the residual goes to the tragedy. The 40% are kept as reserves. So Bank of Italy has some initial cushion that you can use, which is a non-trivial cushion, as I'm going to mention. But once that's done, it's not clear. There's no automatic provision for recapitalization. So then, if somehow these profits don't adjust, these remittances don't adjust enough, and you can think these rules could be changed. These are set at the national level. They're not set at the central level. So what if those rules change? For some reason, what are the other two escape valves? One is the target to liability. So suppose Bank of Italy just does not change its remittances, or it doesn't change them enough to account for the losses that they make. One of the things where this wood shop is in an ever-expanding target to balance. So those could explode over time. And then what happens if those explode? Well, these target to liabilities are zero sum game between Bank of Italy and the rest of the euro system. So somebody else is having exploding assets, and they will have to cut their own transfers to their own treasury. So to be concrete, let's say the Bundesbank would have to cut transfers to the German treasury. What if they don't cut either? Well, budget constraints are budget constraints. So something has to give. At that point at the central level, the central bank policy would have to change. What's the most obvious change? Increase money printing. But of course, if you start touching money printing, then there is an inflation implication. So one of the implications, if you're not willing to tolerate this ever-exploding imbalance, is having higher inflation. And inflation is also a shared risk. Now, that was the case in which bank reserves and assets pay the same interest rate. So there's no profit to be made. So you see this is the present value of profits from having issuing bank reserves and from having target to liabilities. These are the same. This term is not there. If that term is there, so if you actually the bank reserves or target to liabilities pay interest, there is below the risk-free rate of the economy. Then what happens is, if you have a target to liabilities, you're actually earning extra senior edge profits. You're appropriating a greater share of the senior edge profits that accrue to the system as a whole. And that's the alternative. So it may be that if the interest rate on bank reserves, so the policy rate of DCB sufficiently low, there's no reason why target to liabilities should explode over time. What's going to happen instead is that Bank of Italy is going to earn bigger senior edge profits on their liabilities. So they're going to essentially rebate less money to these Bundesbank and the Bundesbank correspondingly will earn a smaller share of these profits accrue to the system as a whole. And that's another backdoor way in which the common budget constraint comes in and leads to mutualized risk. So I think I should wrap up. I'm going to briefly mention the quantitative work is a work in progress. We do have an example. It's a very, very, very benign example. And I'll show you to make that point and show you what happens to the target to balance. It goes down. So this is a default by Italy that causes the target to balance to go down just by a factor of 3%. I showed you that we're sitting at minus 30%. This is minus three. Just with minus three, you get a spillover of 9% of the fiscal cost of the default being borne by the taxpayers of countries other than Italy. And so, you know, as you design more stressful scenarios, the risk sharing is going to be correspondingly bigger. So the point of the talk here is that assessing these risk sharing principles is complicated. It depends on a lot of details. And the key question here remains about thinking possibly about coordinating and limiting remittance policies at the central level. If you have a stronger risk sharing. And so let me leave it at that and open the floor to questions. Thank you. Thank you very much, Marco. Wonderful, wonderful discussion. And of course, it's a fascinating topic which goes to the heart of monetary union. I know that there is a question from Leo Fontaten, who always says that we need more papers who really disentangle and look at money, not just a single country, single treasury, single singles and the bank, but which look into the details of monetary union. And Leo asks the question, what is your view on the underlying normative question? How to design QE in the euro area? Was the euro system right to design QE in a way which attempts to avoid loss sharing in the first place? Okay, so normative, my job is an economist. I'm on the positive side. I mean, I do have normative papers from a more abstract perspective because here the key question is a political question. Do we want more risk sharing? Well, I mean, we can talk about in principle, if you have an abstract risk, people like insurance. So that sharing risk is great. But then of course, we all realize that the moment we do that, there are also moral hazard questions. And so it's whether the moral hazard is big enough or small enough is what this is about. We know that the European, we have rules at the European level. We talked about them in the panel at the beginning of this conference. We know that they are imperfectly, certainly imperfectly enforced, sometimes for good reasons, sometimes for reason are more complicated. And so do we believe that those are enough to get rid of the question of moral hazard? Presumably, when DCB set up this arrangement of limiting risk sharing, they were concerned that these rules were not strong enough. It's not my call. It's somebody, it's the politicians call at that point to evaluate whether these rules are not sufficient and also how they believe they will be enforced. Thank you, Marco. That of your reply, of course, goes to the heart of the political policy issue. As you say with your title, 40 budget constraints for 18 National Center, 19 National Center Bank, 19 Treasury. That of course shows that whatever happens in this respect when it comes to, you mentioned, virtualization, or who is losing, who is gaining, we are in this gray zone between the monetary policy center bank balance sheet, which is of course governed by the decisions of the center bank governing council, and the kind of fiscal implications that can have and which can be heterogeneous across countries. So certainly that is the issue. In this respect, there are some academics who say, if there's no this sharing in case of default, your examples of one country default, then if the center bank has more assets, then if we start, say, there is 100 in a country, and there is a fundamental risk of default, say, if this happens, if the bad state occurs, there would be a 40% debt relief needed. So that would mean 100 goes down to 60, 40% haircut. Now, if you have the same fundamentals, but the center bank has bought, say, 20% of the bonds, if a 40 haircut is needed or debt relief is needed to restore that sustainability in the bad state of the world, then with 20 on the budget of the National Center Bank, the haircut would have to be 50 on the private sector. So to have the same net deduction in debt or the same net support for the future taxpayer. So that people arguing, as Daniel draws, if that is the case, then purchases by the center bank in such a monetary union concept, in such a monetary union situation, if fundamentals were given, should increase the default premium which the market imposes on those bonds which are still outstanding, because they have to fear that in case of a bad state of the world happening and the default infrastructure happening, the private debt has to be hit harder in order to achieve the same debt relief for the taxpayer of the county council. What is your reply to this? I think that's correct to the extent that the risk is not shared. So think about one individual country to the extent that the bonds that end up in the hands of the individual central bank can be defaulted on. Well, the one thing, let me qualify that. I was a little bit too rash. There's always an escape valve and actually thinking about the individual country is a great example. So typically what happens in those cases is inflation is the escape valve. So if we do observe all the countries that ran large inflations, they ran large inflations because the central bank was buying lots of government bonds, in particular hyperinflations. That's a mechanism. Sometimes they bought private bonds too. So for instance, in the famous German hyperinflation, the rice bank was buying government paper if I remember correctly, discounting it at 15% while inflation was so medium percent. I don't remember the exact number. So as you can imagine, everybody went and discounted that. So although government bonds other than some collector items were in the hands of the rice bank, there was a limited amount of discounting of private stuff at 30%. Of course, that also was discounted as much as possible. So one way out is it doesn't have to be an actual haircut on the private bond holders. It could be inflation. How does that play out in the euro system? That's different because in the euro system, inflation requires a choice that is made centrally. And so at that point, you sort of have three mechanisms. One is either these remittances from Bank of Italy to the Italian Treasury really go down. And then yes, there is potentially an increase in the risk of what remains on the books of the private sector. Or somehow the European institutions agree on larger risk sharing. That's another possibility. Or the third is the larger risk sharing happens because that's a little bit of extra inflation. And that is a third possibility. Which one is going to play out? I don't have a crystal ball on just telling you the three options. No, no, no. Thank you very much. This is of course very, very good reply. And nobody has a crystal ball, but therefore everybody in the euro system, especially, is working to avoid the bad scenario where this default could ever be a realistic risk, a realistic risk. There are other very good questions. One from Gennad Mule, who says it's a great paper and he asked for the clarification. Is there also a transfer across countries in the monetary union implicit in the EU in the absence of the foreign banks? Just because the banks differ in how they transfer profits to their treasury? So you can argue whether it is or it isn't. But it's true. So here's one thing that is true. To the extent that Italy does no default and Germany does not default. And now Bank of Italy is buying Italian bonds that pay a higher interest rate than German bonds. Bank of Italy is earning extra profits and those profits are not shared. So on the upside, I think it's pretty likely that I don't see why it's not Bank of Italy is going to remit those extra profits to the Italian Treasury. And so, yes, in that sense, you can think that has distribution implications. Bank of Italy is making more profits than the German Bundesbank. But of course, you can also at the same time say, look, we are in the state in which the Italian debt is being repaid. Those interest payments are being made by the Italian Treasury to the Bank of Italy. They come back to Bank of Italy to the Italian Treasury. Is that a transfer from the German taxpayer to the Italian taxpayer? You can argue, no. Certainly, if the risk is isolated. So if in the law scenario, it's the Italian taxpayer that is on the hook, that's exactly the way it should be working. But yes, that is saying that the upside is not being shared. If the downside is being shared, that's kind of an extra worry that you may have. Jermij, you're muted, Klaus. No idea. Thank you for the answer. And I would just add that going back to the question from Leo Fontanen, that the ECB having decided that most of the risk is not shared, of course, has in that sense, it has been helpful to the Italian Treasury because what you mentioned, the higher interest rates on Italian bonds are not shared within the system, but they accrue only to the bulk of it to the Italian Treasury. And I would say from a principle point of view, conceptual, to the extent that you would argue that this risk premier are to some extent not reflecting fundamentals, but some kind of distortions or some kind of self-fulfilling too high risk premier. It is also fully justified that they go back to, justified from a normative point of view that they go back to the Italian Treasury. The next question which I have here is from, I read them out both, we still have five, six minutes, one is from Chacopo, can you rationalize the role of official assistants, for example, ESM knowns in your analytical framework, were thinking, for example, about the Greek debt crisis. And if Pappar has a question, the paper is very good, she says, as it is, I have learned a lot, but I was wondering how your model would fit the Greek crisis as well. So here, if you take your model and look back at the Greek crisis, what would you say? It's a very interesting thing, and I mentioned one thing, but I guess you know it certainly, that in the case of before, in order to inform the audience, I have been there, so before the Greek debt structuring, there was the S&P program of the ECB. The S&P program was not risk-shared. Once the Greek program was finalized in May 2010, one week later or so, the market was very volatile, and so the ECB decided to buy Greek bonds, although Greek was in the program, and Portuguese and Irish bonds. And two years later, there was this PSI, the debt structuring of Greece, but at that point in time, the ECB, the EU system, and of course the finance ministers agreed that there would be no default on the Greek bonds which the ECB had bought in order to stabilize the market. So that is of course, in the history, a different situation than the one which we currently have and which you were discussing. So much to the background, so now how do you see it? So I think, personally, this is strictly my view, but I like the idea of relying more on institutions like DSM and explicitly making more transparent where risks are or are not shared. I find the current arrangement a bit opaque. Now, sometimes opacity is deliberate. I'm not going to take a stance on whether that's the case here or not, but if the risk is to be shared, certainly I would like that to happen in an institution that is designed with explicitly that purpose and where people can debate about the merits and costs and benefits, we talked about those. In the case of Greece, as you mentioned, we don't really know the scenario that I am analyzing didn't play out because the ECB was made whole. And so we would not know how this would have played out if the ECB had incurred losses. The other thing, so that part, I didn't go back and do that in time, but you were mentioning that that risk was also not supposed to be shared at that time? You're muted again, Klaus. It was no, no. When the S&P was designed, there was no mentioning, as far as I know, but colleagues can correct me, I think there was no mentioning that the S&P risk would not be shared. But the point was, it was not that the S&P purchases were on the sender bank balance sheet, so also the Bank of Greece and under the contract, I think some of them. That was not the case. They were on the ECB balance sheet. And that became, as a certain surprise, not only to the market, but also to the IMF, that the ECB said, or that in the end, the ECB was made whole, which means, of course, everybody was made whole, the whole system. Nobody, not the Greeks and the Bank had to suffer any losses and also not the ECB itself. But there was, it's interesting from a political economy perspective, Marco, there was later a development where the profits which the ECB made on the profits, on the bonds, because they were purchased, not at zero interest rates. That's a significant interest rate. And the Greeks honored this debt. The Greeks had to pay this debt because there was no haircut. And there was an agreement that the profits which the little system ECB made because Greece was honoring this debt, these profits were given back and not by the ECB because that would have been money for financing. But the pressure is of all member states agreed that they collectively would give back the profits to Greece. So in a sense, that was, I think, a nice institutional decision just to honor the fact that Greece would pay back its debt to the ECB and then the pressure is said, we don't want the profit from the senior from this, we give you back the senior exchange. That was, say, kind of, and this giving back was done in a way that it was linked to progress in the program or to the funds and so on. And that was assessed by the Unicode and the ECB and so on. So it's quite an interesting twist. So asking the question, Greece gives you even more. Can you provide a second paper to apply your model? Any further, say, I'm looking here for the question. No, there are no further questions. So a final reaction by you, Marco, to what we just discussed or more generally? No, I want to thank you for organizing this great session and it's been a pleasure. I think we kind of learned the details again, looking at Greece, the details of how these things play out matter a lot and so it's worth thinking about. Okay, thanks a lot. I would want to thank you, Marco, and you, Efi, and Röhn for the wonderful discussion we had this afternoon. I would like to thank the organizers. I see here on the screen just Jacopo but also the other colleagues who were working with Maria and Yonano and Demos, I mean. And Leo, I think I have not forgotten anybody. Thanks a lot everybody for organizing this and thank you for Marco, Efi, and Röhn for the wonderful discussion. I wish everybody a good evening and afternoon. Jacopo, do you want to make any further remarks? I just want to thank also on the side of the organizer, all the participants. I think it was a great afternoon and thanks in particular to the speakers and the discussant. And just to mention that I hope we hope to see you all tomorrow at 2 p.m. We resumed with the keynote lecture by Ricardo Reis and have a good evening. Thank you. Thank you very much. Thanks, Klaus, for an excellent chairman. Thank you very much. Bye-bye. Thank you. Bye-bye. Bye.