 So my question is, is the following one, well, as you say the worst seldom happen, in fact. I could multiply examples in my, for example, I remember when I was a young director of the policy planning staff in the French Foreign Ministry, everybody was talking about thieves, terrorists using plutonium in the Hudson River and this kind of horrible scenarios which never happened. We could multiply scenarios like this. The problem seems to me is that today the degree of uncertainty is probably higher than any time I remember before because of the complexity of the world and the playing with all the possible conceivable scenarios create this degree of uncertainty and therefore the degree of anxiety. Hence, I suppose, the pessimism you mentioned at the end of your presentation at the Bali. So from a theoretical viewpoint, the link between the real economy and the monetary financial economy is through some measurement of uncertainty to formalize this kind of situation. So I don't want to push you to be pessimistic. On the contrary, I think your rational analysis is extremely precious. But nevertheless, if you integrate some measure of uncertainty in your implicit equations, what is the outcome? How do you modify your relatively optimistic description on your real side economy at the first part of your speech? I would make two points. The first one is it's not clear that this is the highest uncertainty that we have faced in the long time. I can tell you that in 2009 I was a bit confused as to where we would be in 2010 and I suspect you share my confusion. No, uncertainty happens. It's not a different type of uncertainty. Sometimes you have statistical uncertainty and then you have, again, subjective or naive uncertainty which is you know that things can happen but you can't quite get your hands around it and you don't know exactly what the effect is. I think the financial crisis was more of an example of that and this is even more in the sense we have a very hard time thinking about all the implications of the trade war. We probably are missing a number of them. So that's the first part. The second is you're completely right that I'm going to give you a nerdy answer showing that I'm still an academic but we used to think of models in which our decisions depended on our expectations of what would happen. I think we've learned and we kind of knew it but we've learned empirically that's important. It depends very much on uncertainty itself. Now how you react depends and for example if as a consumer you know that you face more uncertainty when you're old you're likely to save more for example but if there's more uncertainty about your time of death maybe you save less. However there's a case in which it's completely clear which is investment because here the question is if you think the uncertainty is going to at some stage resolve itself. So there's going to be Trump will be out or there'll be some agreement on Brexit or something like this. Then you know that if you wait until then you will actually be able to take a better decision. So you take the decision which to you is a minor decision which is why instead of spending the funds this year we're going to wait two years. And for an individual firm it's not an enormous deal you're just putting money aside but for the economy it's less investment and it can lead to a very large decreasing investment in recession. So these uncertainty effects are clearly essential I think to understanding the world today. Well thank you very much Olivier so I think we will stop here at least there is one area where the degree of uncertainty is relatively limited this is what we are going to eat now. So thank you very much again tomorrow we will have a session on the future of the euro and I am sure that we will have a very rich session in terms of uncertainty as well. So thank you very much. Bon appétit.