 So we have a few minutes left for questions. One question is from Chris from Nivenhausen. Whether the policy supports exceeded the corporate liquidity shortfalls, meaning if you sort of add up all your numbers, whether you can see that firms actually were able to increase their cash balance? Yes, definitely this is about being that most money went to storm firms. In those things, again, we don't observe actual, you know, 2021 cash balances, but our model predicts that definitely what happens. Now, to check in real life this happened or not, we need real-life data. And the only work so far I know is the work from BIS, and co-authors using French data by looking at the actually what happened to cash balances matching firms to the policy pickup, the amount of policy they receive. And they find actually very similar results to us. Again, only four friends, but they find that, you know, a lot of money went to these firms, you know, kind of, you know, could have survived without the money, but also they didn't find its own location. So there's definitely a base in that sense, but maybe, you know, yes, storm firms got this, they increased their cash balances, but maybe we didn't create so much of a notification. So that's kind of the good part of this message. Okay. And then, so you tell mostly positive story for the advanced economies because of fiscal support. Does it mean there are no long-term scars? Or let's say that's remaining high or some reallocation needs that would call for further accommodation of monetary policy? Yes, that is definitely correct. There can be further scars. And that's exactly what I wanted to communicate with my last point that we should watch monetary policy and financial markets, right? Because there is a lot of debt, not just government debt, but also corporate debt. This corporate debt is higher actually in advanced economies. A lot of large firms just went to the market in advanced economies. They just, they just borrowed, they increased their corporate debt for sure. In that sense, for the future, there can be scarring and there can be, you know, downside scenarios if we are not careful with monetary policy and if we are not careful with general working of the financial markets. That's exactly it. Okay, and then there's a conjecture here since of course you cannot incorporate bankruptcy costs in your model. Is it fair to say that your output loss is sort of a lower bound so that these policy support measures may be really well for improving, especially in these developed countries where sometimes bankruptcy costs can be very, very high, especially in Europe? Yes, no, exactly, you're right. And we cannot incorporate that and that's exactly how to interpret it. But I mean, if you really start getting details of these bankruptcy stuff, I mean, it's amazing. So I really, at this point, recommend you don't look at these type of figures that show you all this much bankruptcy and exit during the Lehman collapse and European crisis and look, nothing happened now because this is very dangerous, these type of messages. When you look at these countries originating in bankruptcy, called just in normal times, forget to call it normal times, and now what happened to call it? It is a mess. I mean, they're just Germany. Germany actually basically, when you have a liquidity shortfall, you go and file for bankruptcy. You don't even wait to install it. You look at it, you file. Germany, stop that. You just cannot, even if you have a liquidity, you cannot file. So many, many things like that. So it is going to take us some time to sort this out. We also interpret it as a lower one, but I mean, once you start bringing this type of heterogeneity and what's really happened artificially during COVID on top of that type of bankruptcy and heterogeneity, things can change, of course. And last question. Can your model also show sort of the, well, when you do the prediction of your model, can you show sort of the evolution of bankruptcies because at least in Europe, bankruptcies actually came down during COVID quite substantially. Now they're starting to pick up just a little bit. So when you sort of do the sort of linear projections of your model, can you sort of mimic that data? Yes, we can do that, but on purpose we are not doing that because then you are going to say that, okay, this model is the only thing that will give you this. But of course, as I just told you, there are all these things. But on average, maybe you can take that seriously and we have a full offset, actually a negative number in advanced countries. Remember, I put minus 0.4 and that's pretty much going to, on average, match the experience of advanced economies. Correct. Okay, well, that's more interesting questions, but I have to let you go because Thorsten is waiting. Great paper. Thanks so much, Shepnam. Thank you very much.