 So, good afternoon. Back from coffee. Thank you for being on time. Sorry for the somewhat shorter coffee break, but we have still a very, very interesting session ahead of us, and maybe just before, just to say thanks for the Schumpeter comparison, just for – I hope you're all aware that after his being Minister of Finance, he became CEO of Biedermann Bank, and Biedermann Bank went broke soon afterwards. So, and then he went to the US to teach, to make up for the debt which he had accumulated by running the bank against the wall. But, thank you very much. I will not follow his footsteps that much. Okay, now it's already a tradition that in the last session we focus, which is usually the last part, and it was also in the chart, which was shown by Claudio in the beginning, it was the last box of statistical production, namely the question of dissemination, communication, and it was already alluded to also a little bit sharing. So, this is always a tradition that we do this in the last session, because it is an absolute crucial part and parcel of the whole statistical production process. But, as always, if you are at the end of the chain, sometimes you're crowded out, that as you know, will not happen today with the timing here, but it happens often that you're crowded out due to lack of resources, etc. And so, but as we always say, we see statistics also as part of accountability, of the accountability of the institution. I'll come back to that in a second. So, it is absolutely crucial and if you had, back then in 1996, there was this famous statement by Alexandre Lampfalusi, the first EMI president, so nothing is more important for monetary policy than good statistics. Usually, we stop here, but if you read the rest, it says it is important for the analysis, for the decision-making, it's important to explain your decisions, it's important that you see how, what are the effects of your decisions later on, and it's important at the end of the day to be credible to the people and to the world, which we heard six, seven times before, trust. So, it is, this is absolutely crucial here, and so I think that is really important, this trust question, and there was just last week an interview of two pages again by somebody who was mentioned or advised today, namely Andy Haldane from the Bank of England. I left it in the office, like all my notes for this session, I left it in the office, but he said basically, and there are data, there's data, the euro barometer from March of this year said that 46% of the Europeans don't trust ECB, and that used to be 24% 10 years ago, and it's not very different from, for other institutions in Europe, so there is obviously here, you could say Frankfurt, we have a problem, and what he also says, and that's I think important, it leads now to the discussion that people cannot trust things which they don't understand, and so it's crucial to talk to the people, to go to the people, and the Bank of England, now those town halls and sits down with the people in all parts of the UK, and at least every board member goes out at least once a month or twice a month to a town hall. Okay, you can now imagine our board doing this that I don't want to say anything more on this, but I think this is important, so that brings us to communication, and we are also trying to do a few things, Werner alluded to a few things just before, if you look over there, you see the posters which we have done for 20 years of statistics, trying to pick out some of the large amount of information we have now, you saw in the beginning, we did a little video, actually today we publish a new what we call Insights, so every two months or so we take one specific statistic and explain it a little bit, so today we publish something about corporate debt and corporate debt across the different euro area countries, so we're also trying in baby steps, but again I'm back to the crowding out story, we definitely don't spend as much as we should be doing on communication, so the easier way is then simply to invite professionals and listen to them and hopefully learn from them, and so I have the big pleasure to announce to you already here a wonderful lineup of three speakers and the discussant will look at different aspects of the question of communication, but also this sharing of information, which is also I think very important, it was also mentioned several times today that the sharing of data, the access to data, the access of data, not just by the public in general, by academia and others is absolutely an important problem and important challenge and important issue, so we will have Andrea Mechler, she is member of the governing board of the Swiss National Bank and she has been, she's back to Frankfurt, she has had here some time ago, she was here at the ESRB, which you heard already today several times of the ESRB, she was deputy head of the secretariat of the ESRB and is currently on the board of the Swiss National Bank and she will let us know, there were a few things which the Swiss National Bank had to communicate about and really hear about data sharing communication from the perspective of the Swiss National Bank. Afterwards we will have Alan Smith, he's a data visualization editor from the financial time, so he will tell, that's his job, tell us something about data visualization and the enormous potential today of data visualization, you know, we used to say, I don't know if they still say, that a picture says more than a thousand words, so there is a lot of potential to make data graspable by different audiences, by using the pictures which these audiences can refer to and also Eurostat I know is very, very active in this area, so we'll hear from a professional here and then we will have last but not least Nicolas Sveron, senior fellow at the Bruegel Institute in Brussels and at Patterson Institute for International Economics in Washington DC, he's a very prolific writer on banking issues, financial market issues, so I cannot imagine that anybody has, somebody in the room has not yet read some of his papers on banking union and on other financial market issues and he is so obviously a very keen user of data and also somebody who obviously would like to have more data but I'm sure he will let us know about this in his usual very, very good rhetoric and then we will have, as discussing, we will have Brian Blackstone, his bureau chief of the Wall Street Journal in Zurich, he has been also before here in Frankfurt and ECB Watcher, and I have to apologize that yesterday Ms Lautenschläge and her speech and the three newspapers she mentioned, that was the Financial Times, Le Monde and El País but she forgot the Wall Street Journal, so I do up for this now. So without any further ado, Andrea, the floor is yours. Dear Oral, ladies and gentlemen, it's a great pleasure for me to be here. I feel we've had a number of very insightful panels today. I want to take a little slightly different perspective. We've heard about what new data is needed for monetary policy purposes from a financial stability perspective. We've heard about what does in what way data is changing. I'd like to take a more pure central bank policy perspective. How do we communicate monetary policy and what is the role knowing the complexity that underlies today's world? Whether market structure, economic structure, financial market, what is the role of data in order to support our communication? Eventually what we need is what Oral said, the most important thing for a central bank is to remain credible and trustful. So let me give you a little bit of a background. Ten years after the global financial crisis, monetary policy in most advanced countries is still in unconventional territory. Switzerland is no exception. But Switzerland, there are a couple of differences. One of them is we have a different monetary policy stance that some of our large neighbors, in particular our monetary policy, continues to be on the expansionary path. We also use different tools we have, and this is what we have here. Let me go directly. We have two monetary policy instruments that we're using right now. It's the negative interest rate that we impose on banks deposits held at the central bank, but also our willingness to intervene in the FX market when necessary. Another thing, because of these differences, and as you know we have a safe haven currency, which makes us much more vulnerable to excessive appreciation of our currency with direct impact on inflation. So that means our communication has had and continues to be slightly different depending on what we need to communicate. But one thing is clear, and that's been clear for all of the central banks, especially the ones very active in using unconventional policies, is communication with the public has become much more important and much more interactive than before the crisis. And I'm going to do here divide that communication into parts. One, there is the need, as Oral said, the need to explain your policies, how they work, why they work, and how effective they are. The other part, is the part even before, is the need to understand what is actually going on. And the main thing I'm going to, if there's one thing, I know it's late in the day, the one thing I'd like you to take away from ours is the fact that for us it's important to communicate in simple terms, in an unambiguous way, but in order to do that we need to understand the issue at depth. And it's, and that's where we use the most data, and where we're very analytically focused. And it's that difference, how do you get the right balance on these two things. So I'm going to use a market that hasn't been discussed much today. Usually we think about inflation or financial stability. I'm going to use the example here on the FX market, which is a very important market for us and is slightly different than for instance the ECB or the Fed. So I'm going to start with the second part. It's first in order to communicate in simple terms, we believe it's incredibly important to understand well the complexity that is behind there. And in fact our communication, as you know we have a safe haven currency, our communication on the Swiss franc has been and continues to be crucial. And right now it's based on two statements. Our Swiss, the Swiss franc continues to be highly valued. And the second is that the FX markets continues to be fragile. And this is really a very important, two very important parts of our communication. Now those are two very simple ways to do it and it's very important that the financial markets understand what we mean. But first before going to the communication, how do we think about it and how do we analyze it. And that's for us at the central bank incredibly important to really do deep dives into the data. I only have 15 minutes so you can imagine I can only give you a glimpse. But here's one of the glimpse. So when you think about a central bank thinking of the exchange rate you think it's all about the value. It's the level of the exchange rate. It's not. And in fact this is what I'm showing here. So here what you see is the level of the exchange rate over 45 minutes. That's already a pretty long time. This happens to be on on February 8th. I don't know if you remember that was the day when suddenly to suddenly big fears that the the the the wage data in the U.S. suddenly was much better than expected. So the big fears that suddenly inflation might actually pick up faster than expected. So the stock exchange the S&P 500s had a big dip. There was a big risk of environment and the Swiss franc got also affected. So this is just to give you one glimpse 45 minutes. It's a pretty long time. But what that it's not enough to look at the level. In fact the level is not that interesting. But here's just one other dimension. It's the what we call the order book. So we look at how much there is on the order book. So on the ordering how much buying and selling orders. Not even the actual transaction. The actual transaction you may not see them very well but there are a couple of dots. The dots not the lines are actually the actual transaction. But there's many more orders on the buy side on the sell side. And this for us is very important and it's very interesting. And I'm just going to just to give you a little bit of sense how we think about it. So here you see an order book. The blue is the the the blue side is the buying euro Swiss. The red side is the selling euro Swiss. What you see is before whatever happened. So if you look at here from 4 p.m. to 4 15 we had a very deep market. Any two sided market the buy side the sell side were both quite quite deep. Suddenly there was a there was a noise in the market and what we see the level change. The level change but the order books continued to be quite filled. Suddenly you see a change. The order books becomes much much smaller on both sides. You do see something that's very specific to a safe haven currency when there is uncertainty. Suddenly this transaction that remained the actual transaction only one sided. If you look at it on when when you see the Swiss franc falling the euro Swiss franc it's only on the you only see selling euro Swiss so i.e. buying the Swiss franc. But suddenly itself corrects quite rapidly 10 minutes into it and then you and then the the level goes up but also the order book on both sides fattens up again. What when we look at it this was actually a fairly good environment in the sense that at no time do we see that the selling and the buying interest remained in the market. So the market making remained both sided all the time. So this is just to give you a sense that for us the level only gives you one perspective but there are many other perspectives in order to understand what is in the market. Of course eventually you need to go much further. You need to understand who's behind it. What were the reasons why did certain positions change. But the point is really here how important it is to go much further down and I want to go here give you a little bit of a sense understanding the depth of in this case the FX market is key for us to foster trust and credibility. Unless we understand or try to understand as much as we can the complexity and I'll come back also the innovation that's happening in this market it's very difficult for us to stand and make fairly simple statement like the Swiss franc is highly valued and the FX market continues to be in a fragile situation. So here what you see so what I just showed you was one glimpse about FX markets that goes behind just looking at the volume of the price. So one it is important to have a very good near time monitoring of the FX market. That's important to implement monetary policy. This was true when in cases when we need to intervene. Also for instance when you do asset management when we manage our reserves to make sure we can be as market neutral as possible. But that is not enough. What is important is if you're going to have near time monitoring is to link it to more medium-term monitoring of structural trends. In the end for central bank it's not the tick by tick minute by minute five minutes by five minutes important movements that are important. What is really important is to be able to see when there is a class of investors that starts to change the positioning. For instance for us as a central bank it's important to know who are behind is it corporates that are trying that are starting to change the hedging behavior. Is it speculative investors that are trying to take advantage of certain changes in pricing in the market. Because that makes us look at for instance in this case the role of the Swiss franc and how the Swiss francs functions in a very different light. And then of course there is communicating but here it's not communicating necessarily only with the public. I'll get there. For us it's also important to be able to communicate at high level with market participants. In order to do that you need to really be in the market. You need to understand the complexity. You need to keep up with the changes like Sabine Lautenschläger said very nicely yesterday how important it is if you want to be if you want to remain in confidence you have to be able to speak the same language with all the people with the algo traders with the high whatever makes the fast-paced market. So that is something that's incredibly important for us to even start to be credible with the market participants that deal in those markets. Now what is one of the challenges that we have about data and that's what I have here is that how first of all it's easy to say we're confident. We also have to have a good sense how confident are we that we're actually seeing what we want to see or seeing enough of what we're trying to understand. In here there are many changes and I think the two important one one we've seen it in the previous panel is technology has changed and a very important one is the one that I'm showing here is the share of algorithmic trading in the FX market. So you can see here is the first data point in 2004 almost non-existent and nowadays you can see that by 2016 almost over 75% of all trades in FX markets are done through algorithmic trading. So that's completely so it makes it very difficult because it changes who's behind it to drivers how fast things happen and it really requires a very different approach to understand what it means and even to be sure do we pick up all these trades that may be done on different platforms. So new technology is a big issue the other one is new actors and here I have another graph that just shows you for instance again if you take the same period 2004 you can see that before most of the FX markets was done or a large part over 50% was done by financial institution to which we had access in this case reporting dealers nowadays much less than 50% in fact it's just about 40% is done by institutions to which we have access the rest we may not even have access to and it's done on different platforms it's done on different venues so again for us it's incredibly important to know the unknowns whether to know the unknowns or to at least have a sense of how much of the unknown is really unknown and how much do we know is out there but we can't follow it and how much do we actually are we able to monitor but this first central bank is incredibly important work for us to do in order to eventually implement monetary policy in the best way possible. So this is just a couple of glimpses on how for us the whole data the technological change and the new technology the new actors the new regulation in the FX market is something we really have to have to be able to keep up with the development and the markets. Now I'd like to switch over to what it means for our communication so communication so communication and I leave some of the nicest chart I'm sure they'll come up for afterwards from the financial times but for us communication I think I'd like to come back for us communication has several it is about explaining what we're doing and make in for this it needs to be fairly simple and it is about sharing regularly our insights so because this is too small for you to see but the graph and you don't need to see the detail a very important tools that we use is what is called a conditional inflation forecast and we simply use it in order to give our rationale of how we see inflation to behave over the next three years but clearly there's nothing mechanical about these developments right what is much more important is what are the drivers what is the global outlook how much of it is due to swiss factors swiss driven national factors that drive inflation and how much is due from other factors so in the end we can we have a number of tools our our press conference a certain number of of charts that we show regularly but in the end it is really about explaining how we think about monetary policy here we have a couple of things what is important for us I would say three things the way we think about communicating one it it's important to be transparent and what do we mean by transparency for instance we have a large balance sheet we have it's all marked to market we are very open I don't know if if I have well no but what is important is to be I think I have one but it's too small I'll just show everything I have here what is very important is we have regular data on our balance sheet we have 800 billion worth of fx reserve that's over 110 of GDP that we have accumulated in fx intervention we're very transparent it's all marked to market every quarter you see the profits and the losses and whether you have profits or losses they often go in double digit billion it's a very important communication that we have with the public we also provide on a weekly basis the level of a total site deposit that's usually a fairly good proxy it's not a perfect proxy it doesn't give you all of the information of our intervention so again we feel it's very important to be able to explain it in simple terms but you also have to to be able to provide the tools so that it can be followed the other thing is we believe in one voice we're a small open economy where the exchange rate plays plays an incredibly important role at least over the last 10 years with where the swiss frank has been under excessive appreciating pressure so having a one voice from our board members is very important so this is why you see we have different people with three governing members on the board and we all three have a single voice but the voice is a unique voice because it has to be unambiguous the way we think about the implementation of our monetary policy and the other thing is of course is how broad the type of how we communicate with these the channels of communication have to be very varied and they have been changing even for conservative central bank the way we like to think of ourselves sometimes where we end up going not only having press conference like every other central bank but going into social media having even a youtube video yes we do in order to make sure we can communicate we can communicate to a variety of audience but always what I find is the trick is how do you keep the communication very simple of course you have to do an incredible amount of work in the background you have to publish some of this work not all but you have to go through the data you need to have the best export in order to be able in the end to remain very simple so this is a little bit so this is my last slide again small open economy where the exchange rate plays an incredibly important role so if I think about just that part the ones I've been focusing on the FX market when you think about it we have two major statements that we have been continuing to make and we will change them when it's necessary to change them we have an exchange rate which is still highly valued we have an in the FX market situation that remains very fragile so in the end we all know it's incredibly complex FX market are going through a lot of changes do many drivers do external drivers do domestic drivers do to technological drivers do the actors that change some of the changes that we see maybe you know can be driven by corporate development sometimes it's by speculative investors all of this but in the end we have to provide a clear picture so then we use very simple very simple design the graphs like for instance this one which is we this is one of the way we've shown the highly valued of course you can imagine it has many models in order to decide at what point and how long we believe the swiss franc is highly valued many models that we run but you can't show models you can't convince someone so for instance this is a simple historical graph that shows where the swiss franc on a trade-weighted basis what the he I'm just showing the nominal exchange rate was in 2000 in 2000 no in 2005 and how strongly it has appreciated and where it is now and for instance for the fragility of the market you know we could show really an incredible amount of different graphs one that we have been showing is to show that the swiss franc continue really to have this safe haven feature in the sense that here it's risk reversal so it's on the using the option market and the only point we're making is that even though right now the swiss franc has been has depreciated from the most from the from its highest level what we see is the markets continue to to position itself against the swiss franc so when when investors want to ensure against a fluctuation in the swiss franc they're willing to pay still much more money against a sharp appreciation of the swiss franc than a sharp depreciation so we see that the market is still very much one-sided even though it's an option that hasn't been exercised at this point as soon as you have the slightest fragility the market will shift and we know in which way it will shift and this is one of the way we have communicated the fragility of the market but I think this is an important complexity is how do you master the complexity and in the end are able to explain your monetary policy in very simple unambiguous and yet transparent ways I will stop here thank you very much thank you very much Andrea for explaining us the challenges of a small open economy and I wouldn't say it's a small institution you mentioned the balance sheet has a very big balance sheet later on in the questions and answers you tell us how you communicate with with the algorithms and the machines and 75 percent of the trading happens through machines so I'm not sure whether they read your tweets and your watch your youtube videos I should have said before I please apologize that Andrea is the first woman to occupy the board position in the SMB so we have here a first Mariana the first woman chairing Eurostat we had the first woman on the Bundesbank board just a minute ago now the first woman on the SMB board so also for diversity purposes I think we have a good statistics here today now I move over to Alan Smith Alan Smith as I said is data visualization editor at the Financial Times in London but it may be also very interestingly here for this crowd here that he was previously head of digital content at the ONS at the Office of National Statistics I don't know if you didn't want to move to a new bury or but he ended then up at the Financial Times and he's always his task is always finding ways to bringing statistics to a wider audience so with that it's the perfect person to tell us something about the strategy for data visualization so Alan the floor is yours thank you Aral it's so good to be here actually and to see a few familiar faces from my old life in official statistics and I have to say it's not an accident that I went from the world of official statistics to the world of journalism because I was very interested to see Werner's note about the importance of syndicating content to media because that's for me I realized a while ago that if you have the words statistics.gov in your URL that's already two of the public's least favorite words in the URL and so the media becomes an incredibly important part of the communication strategy for engaging the wider public and so it was lovely for the FT to invite me to be their first ever data visualization editor now the very early on in my time at the FT I decided well we need to articulate our strategy for data visualization and what are we trying to do with it and I kind of I wouldn't say I don't use the word strategy in its highest form I'm really talking about strategy in terms of the patterns in the decisions that you make about presenting and communicating data and one of the things that we'd certainly learned at the statistics office was that you should aim for impact so I thought well I'm going to try and aim for some impact at the FT and see what where that takes us and so what does impact look like so early on at the FT I started to work on a story that was looking at the changes in the US income distribution over time and so we had some data here from Pew Research looking at 1971 household income in 2014 dollar terms so poor households on the left richer households on the right it's just a histogram and I thought it would be quite good fun to actually animate this chart to show how the US income distribution had changed over time so the line appears to allow us to make the comparison somebody was saying earlier this morning about the difficulties of having reliable high income estimates and that's where this chart shouts that actually it's a relatively neutral chart there is no message in the chart that says it's good or it's bad but what it's saying is look at how the shape is changing so it's relatively ambiguous the interesting thing was when we published the chart in the story my main argument was to say that that chart's incredibly important to our story what we were talking about is how the income distribution had flattened and so the traditional middle class because in the US classes based on income had changed what I didn't realize was how much this graphic would live a life of its own outside of the story and the first thing that happened was I mean I was new to journalism I didn't know this sort of thing could happen but the LA Times wrote their own article about my chart I didn't know that you could do that anyway and because as you can see from the photo this was at the time and it seems like a very long time ago that Bernie Sanders and Hillary Clinton was the main political story but this was at the time of the democratic nomination the race for it and so income distribution was a part of the debate so much so that Bernie ended up taking the chart and putting it on his Facebook page which if you can look at the engagement levels on Facebook just took this chart into this amazing space where it was just at the center of the debate about whether or not the changing shape of income distribution in the US was a good thing or not and this chart continues to haunt me uh because this was a little while ago but just a couple of months ago the North Korean foreign minister tweeted it at Donald Trump it keeps coming back right like and so graphics definitely can create impact okay um and one of the things that's interesting about this particular chart is it kind of adheres to one of the other elements of our strategy which is make sure you remember why charts exist right charts exist not to break up a big page of text or to to just lighten you know to entertain you you know because your concentration might be sapping the first charts were introduced at the back end of the 18th century by a character called William Playfair and I love using Playfair at statistics conferences because he said some fairly fundamental things about statistics which I need to tell you about firstly that I'm afraid the rumors are true right like that nothing is more dry and tedious than statistics unless you use your imagination and his imagination because at the time in this period in the enlightenment that's the only way that data was presented in these big empirical uh tables um and his his answer to that was this the first the first graphics um even before excel right like this is this is what you know and the most important thing is they were designed to represent financial data that was their sole purpose he actually called it applying geometry to items of finance that's how he how he designed it and so what he said was why is this a good thing he said this mode unites proportion progression and quantity in one act of memory that's a really profound thing to say and that actually is a couple of centuries ahead of modern cognitive psychology but what he's saying is compared to tables of data charts really let you see the patterns and relationships in data much more clearly he didn't ever say that was a great way to break up a dull page unfortunately the f t didn't really catch on to the potential of this so this is over a hundred years after playfest first chart this is the front page of the first f t and no charts right and also notice uh lots of data there's always been data in the f t always been data on the front page but you can see a pattern start to emerge here this is 1888 1901 1931 1955 that's when you suddenly see this amazing decision to put a chart on the on the front page right um now the problem is when you realize organizationally that you've maybe been under using graphics the reaction might be to overuse them and so this is the f t in in 1999 and i always like giving my charts names right like the income distribution is the shape shifting income distribution this is the testicular pie charts right like this is and going back to what play fare was saying is this memorable yes right you will struggle to forget this but you won't actually remember the important thing right like which is what story are we trying to tell here right like it's been lost in the attempt and and this the kind of this this sort of approach to visualizing data kind of imagines that regular people will find data boring right like and that for me insults both the data that you collect and also the audience you're presenting it to you've actually got to treat both of those things with a bit more respect so what i've been trying to do at the f t is focus much more on graphic acy uh graphic acy is a term that describes our ability to interpret and understand visual images um so i would set this alongside literacy numeracy data literacy these are the essential skills for modern citizenship okay um and the problem with graphic acy is it has fallen into the cracks of the academic curricula okay who knows how to read a pie chart just put your hand up right come on right pie charts you can read pie charts where did you learn to read pie charts school right like so how many different types of charts did you learn about at school maybe three pie charts line charts and bar charts they were all invented by play fair over 200 years ago they're still good chart types i don't want to uh criticize them but they are not the only way of presenting data and there is nothing intuitive about a pie chart you have to learn it but it happened so long ago that you've forgotten that you had to learn it so what we've done at the f t is we've created something called a visual vocabulary and this is our attempt to explain that there is a grammar to the presentation of data in exactly the same way that there is grammar to the written word and that there are different methods presenting different types of data depending on what you are trying to show um so just to zoom into a portion of this it talks about what is the relationship in the data that you're actually trying to show what what's the message not what kind of chart would you like would you like because that's what excel asks you and that's why it gets it wrong so many times um we've translated it into Japanese Chinese because we think this is a really universal visual language that's emerging for for working with data um and if you want your own copy you can get it from that url it's not behind the f t paywall you can download it as a poster and do whatever you like with it it's um it's there but what i thought i'd do is show you a couple of charts that we've created using this framework for designing with data rather than a round data which is where the underpants graphic got it wrong okay so this is a fairly simple chart um the dow jones index when it hit 20 000 and it's uh the chart of the the dow jones from its inception in 1896 now i am really sorry i didn't know that brian was going to be the discussant when i created this slide because the reason why i think this is quite a bad chart uh as you can see from the title um because it's pretty good at hiding one of the most significant events um in the dow jones history which is all street crash and it was exactly the same chart that the the journal put on its front page now just in simple kind of chart design terms the big improvement you can make to this chart is to use a log scale right like there's a log scale on an index series 1000 and 2000 is the same difference as 10 000 and 20 000 it's just the index doubling in value and notice not only does it bring the wall street crash into perspective but it also makes the financial crisis much more balanced as a result that the most recent events seem less dramatic and and so on but even this chart is problematic because with a time series of over 100 years we know that every single wobble in this line is a big event right like it's it's likely to be a recession and some other events that are associated with it and we don't have the space to do it here so on our visual vocabulary we say look you can use a line chart to show change over time but it's not the only way one of the other mechanisms we can use is a vertical timeline which allows us to switch the axis over and turn it into a vertical timeline where every so this is stretched out big so as you scroll through this graphic you get every recession every event annotation the serving US president that's the wall street crash wait for it the financial crisis comes in there it is and a nice punchline that trump is elected president right at the end there and now we tested this with readers because they said can we can we rely on readers to know that the chart is this way up the response was phenomenal they loved it because it allowed them to see more detail and depth to the series than they would in just the original version which is fine for an overview but not for the real stories behind the index similarly i love this one partly because there's a frankfort connection here this was what happened when somebody one of our banking correspondence came to me and said we need a map you know i said why do you need a map and they said well we've got some cities we've got some data about cities and i said well what do you want to show and she said i want to show which european cities are lining up to take over from london as the EU's financial center after brexit and i said okay well maybe you can do that on a map and so this was a map that shows that data and and she said well also i'd like to see which banks are most federated across europe already so which which banks have multiple presences across the cities so let's just have a look at it i mean i think you can probably see from the map that frankfort looks stronger than lisbon for example right so we can we can look at that and see that what about jp morgan that's really difficult because you have to do what play fair told you not to do 200 years ago which is lots of remembering right so the visual vocabulary says right if you want to compare two things like that put them into a grid because actually the distance between doubling in paris doesn't really matter nobody's going to use the map to plan their holiday it doesn't need to be a map because the spatial relationship is not important so in the grid we have cities across the top banks down the side and they're they're all in order so you can see the strongest cities the strongest banks the weakest cities the weakest bank and then you can start to realize that actually it's a false thing sometimes to have this distinction between text and graphics good graphics need good text and good text very often is improved by having a graphical angle to it as well and the way that you bring those things together is incredibly important so this is another type of chart that you probably didn't learn in school but i think it's probably quite a good chart to learn how to read it's called a marimekko chart and it again allows you to see two very different things at once on the top across the horizontal axis is the proportion of each country's gdp that was the size of their banking bailout and the little shaded area is how much of that had been recovered okay so with the proportionally based on the country gdp what was the size of the bailout the vertical axis is the size of the economy right how big is that economy right and so again we end up with a very tall graphic that we've ended up bringing the text on to and so as you go through this graphic the story kind of writes itself right the US was the only country to recover more at the stage certainly at the stage that we did this graphic and that created that kind of memorable aspect for us so we've actually bolted this poster to the wall in the FT newsroom so when people want to come and talk to us about graphical treatments for stories this provides us with a framework to change the discussion all right to base it around what are we actually trying to show okay and in turn our readers have ended up becoming much more informed with graphics that carry more insight and more relationship um in doing so we've had to take on the very difficult argument which is that some people have said to me well that's all very well but a graphic is terrible if i can't understand it in five seconds you know it has to be simple and understandable in five seconds now there are lots of times and occasions where it is important to understand a graphic in five seconds probably shorter than that like your car dashboard you should be able to read that within half a second because you have to make an instant decision but some graphics need to be read here's a here's a five second chart it's the votes in the 2012 french presidential election you know i can scan it very quickly and see that oland got the most bayrou got the fewest that's a simple chart i can understand it here's another simple chart of the 2017 french election first round i can see that macron won just a few more than le pen and amon was down there at the bottom end of that chart now those charts are fine if all that you are interested in is knowing about those two separate events um what if you actually had data that allowed you to get more insight out of it how about the data that says what happened to oland's voters the ones who voted for oland in 2012 where did they vote in 2017 well that's where you do need to create a more complicated graphic right it should still be clear the two opposite ends of this chart are exactly the same as the charts we've just been looking at the votes in the first round in 2012 and 2017 but in this case we can now see what happened to the voter flows so you can see that half of oland's voters went to macron roughly but the other half split between melanchon and amon and contrast that with how loyal le pen's voters were that most of them stayed loyal to le pen across the those elections so this is a graphic that i would not expect you to read in five seconds because i think it's worth spending a minute or two with it and if you spend a minute or two with it it will reward you with more insight than the originals and that then leads me to the final point i really want to make which is about visual rhetoric it is wrong to think of charts fundamentally as an exercise in saying here are some numbers that is not the end point right numbers are a vital ingredient but charts really are visual rhetoric visual arguments and so if you start with a chart like this which is just here are some numbers here's a vaccination rate data there you go i'll put that in my report that's not particularly helpful to anyone to turn it into a visually communicative piece the researcher should be brave enough to dare to interpret the data they're communicating and tell us what it really means why should i care about it so just very quickly let's see how much we can make this chart over well that rate looks fairly high overall there was a slight dip and a slight increase but actually if you zoom in there was a very steep relative decline followed by a recovery did it just apply to mmr just to measles mumps and rubella vaccination well yes because the other vaccinations at the same time didn't have the same trend and in fact there's a target for vaccinations to prevent epidemics so we should put that on the chart as well and then we probably want to explain to readers well why why did the rate change so much and so we actually write remember i said writing is important write on the graphic what happened an article suggesting a link between mmr and autism which was eventually retracted and then you might say well so what just because the rate dropped what happened well more kids got measles is what happened and there they are that's the kind of impact of that and then actually mmr immunization in england is a terrible title for a piece of visual rhetoric like this we should be actually making it clear what is it saying and when you put the two side by side that's the difference that you should be looking for when you're using charts to communicate data and this is the important thing not every chart can carry that transformation so then question yourself is it worth doing that chart at all be selective pick the ones that have the impact and so the most successful charts that we've done recently they're relatively simple but they've been worked over hard to carry that kind of message like this sort of chart here incredibly powerful for us in terms of turning these charts into visual arguments just to finish on kind of learning more because it was one of the things that i worried about a lot originally when i was at the statistics office the ft have been stupid enough to allow me to write my own monthly column where we post examples of graphics and how we're designing and making our graphics we also did a reader event in january which we might run again later in the year to to learn more about this but i think one of the interesting things is you've got a lot to learn from the people around you right like and one of the things i was in dc in march and the world bank have got some amazing graphics if you look up their latest sustainable goals report it's absolutely phenomenal and it implies most of the techniques i've just talked about there in the context of major bank reporting thank you very much thank you thank you ellen for this fascinating view and also for the positive message there is there is life after official statistics also now i turn to nicolas nicolas veron is uh as i mentioned she is working for broigle in brussels and petersen institute in washington she actually is a co-founder of broogle in brussels so that links him to this house because Jean-Claude trichet is his big boss so to speak as broogle he's been working also has been working as financial policy expert for the european commission for the european parliament for the european court of auditors for the imf and the world bank and lately bloomberg has named him as one of the 50 global most influential people referring specifically for his early engagement or advocacy for the project of european banking union which as you know now at least stands on two two of the three feats which are foreseen two are already developed the third one is still a little bit limping but uh nicolas the floor is yours thank you orl and thanks for having me today um it's a special privilege for somebody who is not at all a statistician to be invited to speak at the statistical conference and i'm i'm grateful for that and i also am more generally grateful for orl shuberts leadership at the ecb in his capacity he's an independent statistician so he's completely indifferent indifferent to praise uh but nevertheless i think this has to be saluted in the current circumstances so thank you orl so title of the conference is what next um and um we have heard today that a lot of great stuff is in the pipeline we heard about anacredit to give only one example which is very directly linked to the banking union project and indeed the ecb has been producing statistics since it's this new mission was hoisted on it in 2012 thanks to orl and his teams and all the all the staff here and with partners such as for example the banking supervisory statistics which have been produced since 2017 and lots of other great stuff so at um orl's invitation i am going to speak about what i would see as the next steps not only beyond what is already available but also beyond what i understand to be already in the pipe and as you mentioned i care a lot about banking union i think it's a significant new development and therefore the natural next step to consider is bank level financial and regulatory data uh in terms of public sharing and communication so public disclosure by uh the eu institutions including the ecb the importance of granular data has been mentioned several times today starting with the videos this morning uh mrs marquez they was there talking about exactly that uh so uh so so in a way that's what my presentation is about and i apologize that there is no charts or graphical rhetoric in there um but uh but it's uh it's a presentation about how to make the banking data that uh this uh house and uh EU institutions more generally provide to the public more granular my perspective obviously is one of a practitioner of policy research at brugel since 25 as a researcher Peterson Institute since uh 2009 um 2005 sorry orl said that i also have to mention something i disclosed earlier which is my board membership in the uh board of the GTCC global trade repository which is relevant in this discussion because it's very much about data and also investment in new fund uh private equity funds that's not relevant but it's a disclosure um and also relevant here uh i've been together with Ted Truman in uh washington and a few others very much involved in the advocacy for uh andreas georgieux uh evas novotny mentions that this morning trust in statistics is nothing if statisticians can be persecuted by their political employers at the national level uh i think it's really important that all the statistical community continues to do what it has done so far which is uh display a very significant level of concern and mobilization in this course it's not an anecdote uh as mr. Novotny said i think very rightly this morning uh we can discuss plenty of technical projects is there is no integrity in the system if somebody who has integrity is punished and lastly convicted in a near evocable way by the greek supreme court to uh for jail uh there's little chance there will be good statistics so uh so i think getting justice for mr. georgieux this is the latest hashtag is actually a high level objective for everybody involved here i think uh continued efforts not just by the ecb but also by euro state and other institutions are very important so that was an aside uh the um an important one uh i i will start from the data gaps initiative uh language that's data gaps initiative two for those of you who follow this a couple of years ago and there is a very important recommendation here number uh two dot 20 promotion of data sharing so this is uh obviously relevant to the title of this session and what it says is the interagency group um calls on um agent a g uh no it calls the a yeah sorry so the grammar of this is not straightforward yeah so recommendation is for the iag and g20 economies to promote and encourage the exchange of data to improve the quality of data and availability to for policy use i highlighted this this is my highlighting so g20 economies are also encouraged to increase the sharing and accessibility of granular data if needed by revisiting existing confidentiality constraints and i also highlight that so this is very relevant to banking data and more generally to what is called in the jargon uh supervisory transparency so supervisory transparency can be about the transparency of supervisors about themselves so for example disclosing how the ssm works uh what's its budget performance staffing uh methodology i actually would give the ssm very high marks for this i've been reading the annual reports every year they're actually very substantial before the annual reports they were quarterly reports on the setup of the ssm before the ssm was operational uh i'm not sure how many readers they're aware for those quarterly reports but i can say i was one of them and they were incredibly informative and substantial and useful for policy observers like me so i think the ssm can really be applauded for that i'm not going to talk here it's not even in the in the in the slide about supervision about transparency about the the job of supervision itself so for example if you take the critical Shrep process of the ssm the supervisory review and evaluation program i think the Shrep um it's for external observers it's a bit of a black box for the banks themselves it has some features of a black box i think that's good and i wouldn't suggest that the Shrep uh that all aspects of the Shrep methodology and certainly not the Shrep proceedings should be disclosed to the public i think it is in the nature of good banking supervision that there is an element of confidentiality uh vis-à-vis the banks and vis-à-vis the public in terms of what they do if only for financial stability reasons you if a bank is near bankrupt and you have to act you perhaps don't want to tell the public first but also for reasons of good operation of supervision more generally because if you give the public and the banks too much information about how you do it you create scope for gaming the system regulatory arbitrage that's a lot of the debate about stress test methodology in the US for those of you who follow this and i salute the chair of the SSM who has joined us um what i will talk about however uh is um public disclosure of bank level financial and regulatory data so that's the theme of the rest of my time and there has been some of this uh there has been a bank specific disclosure at the time of the 2014 comprehensive assessment at the time of every year uh the EBA the discloses uh bank specific data for a number of large banks how they're selected is a bit of a mystery the number fluctuates over the years but the very largest banks are always in and that includes stress test uh on a fairly regular basis and in addition i already mentioned the ECB banking supervisory statistics available since 2017 um but aggregated on a country level so it's a little paradox here as an aside which is that banking union is all about erasing national borders in the eurozone banking system but the very value added of this statistical series is to show banking systems on a country by country basis there is a little cognitive tension here but it's very useful data i think um sometimes when i make this argument that the supervisory system should disclose more bank specific data i'm told well but this is a substitute for pillar three requirements and for those of you who know the banking jargon pillar three is this pillar not of the banking union but of the basal framework so pillar one is the mandatory regulation pillar two is what the supervisor does on a discretionary basis but pillar three is the market discipline so it's requirements for the banks to disclose to the entire market uh some stuff about themselves including regulatory uh information and financial information of course and information about risks so that the market can discipline the banks and both the credit markets and the equity markets can send the banks the right messages if they're not doing the right things uh i don't think that's an argument at all because i think supervisory transparency and pillar three uh requirements are complement there are no substitutes actually when you look at jurisdictions where there is the best pillar three disclosure they generally have the highest supervisory transparency and conversely so the question is about public accessibility of bank specific data if possible free from vendor fees and a lot of the bank specific data not all of it but a lot of the bank specific data that i have in mind here is that as that is available for price if you want to pay you know whoever is your preferred financial data provider i think there is a very strong case for the for reasons that i will come back to in a moment that this should be made public publicly accessible for the banks for all the banks in the system and by definitions that cannot be done by the banks themselves in a comparable format so that has to be done by the public authorities i have no strong views on whether this is more of a job for the ECB or the EBA anyway if it's done by the EBA it will probably be data generated to a large extent by the ECB for of course those banks that are in the eurozone but but so in terms of who has the responsibility for the disclosures that's an important discussion but not ones that i will elaborate on for me it's really a question of the european level of supervision which of course is a slightly different considerations if you think eurozone or EU 27 831 in the EEA but but but i will limit myself to the eurozone today because we're at the ECB so there's also another issue which is hugely important which is the legal limitations and i'm very aware that there are legal limitations to trans to what i call here supervisory and i'm not the only one what i call here supervisory transparency because there are countries not all member states i think where bank level data is subject to legal protection in terms of confidentiality competitive advantage whatever that's not the perspective i take here because of my perspective is just what is the right public policy in this matter for europe and if then there is a choice of the right public policy with the right trade of smade then it's a matter of course for the system to deliver the right regulation and legislation but i'm taking a step back here and i'm looking upstream from that question of what's possible in the current under the current legislative and regulatory framework now you probably heard a very significant vote of confidence for the ssm that happened a couple of months ago it's not yet fully implemented but it's on its way which is the move from sweden to finland by nordea the largest bank in the nordic european region and some people said oh nordea is moving because supervision in sweden is tighter than in finland under the ssm and so that's supervisory arbitrage that's self-evidently not true it's not true here but i say the same when i'm in sweden sweden and the riggs banks are members of google so i can be even-handed here and what nordea said is actually the opposite they said not that supervision was laxer in sweden and in the eurozone but they said all our competitors are supervised by the ssm and therefore that's a standard that's a global standard from our perspective if we want to be assessed properly by investors we need to be under ssm supervision because that's a yardstick and and that's actually i believe for lack of a better hypothesis the reason why they did it so i take them at their words there are other theories but i think this is the most plausible theory of what motivated the change so by which i mean the ssm is the global standard it has to be the global standard in supervisory transparency on a long term vision i'm not saying this should already have been done i'm saying this needs to be considered right now the global standard for supervisory transparency is the u.s so u.s has something called the financial um federal financial uh institution examination council i think ffeis iec uh and uh the centralize on their web website something which is called call reports and call reports are reports where uh so this is directly accessible on the web this is just a screenshot uh it's too small print for you to see what is being reported but it's key financial data key regulatory data key risk data so they have data on assets profits and the like they have assets on regulatory ratio capital risk-based leverage whatever and they have data on risk for example npl and uh critical exposures and by the way this report is not directly accessible for the uh on the web for uh credit unions which are the smallest banks in the u.s but uh i'm a client of a credit union so i selected mine which just happens to be the bank fund federal credit union in washington dc known to some of you here and uh it's not on the web but i received it by email automatically after less than an hour so it's automatically generated and it's accessible to uh anybody you don't have to prove your client so it's the same report it's a they call it a call report it's a very complete report and as you can see or you cannot see because the print is too small it's on a quarterly basis and it has plenty of useful information now that's in the u.s if you look at the global level uh there are plenty of vendors as i said which disclose a lot of information and you can even say it's out there into public space so this is top ranking by the uh one of those lists which has a bit of a notoriety uh in in the professions bankers annual top 1000 banks and they disclose some information which you know shouldn't be secret uh as i said financial regulatory information but it's out there in the public domain so so the question then is why is it that it's not out there on the web supervisory websites uh of europe what's in there uh in europe is just a list of supervised entities um uh published by the ecb so as you can see this has also supervised entities including the less significant institutions ls size but the but the uh information about financial um data is very formulaic it's just a bracket for the largest banks of the size of their balance sheet for reasons of why they are considered uh significant institutions and for less significant institutions there is no information at all um none none whatsoever there is just the name of the bank and the list of the entities but no uh quantitative information uh the eba of course has much richer information on the large institutions for which it publishes information at this point as i said it's the samples that varies over the years it has started to stabilize a bit but it's still unpredictable and therefore you don't get uh excellent series uh and also uh even that information uh is very selective for example i haven't been able to find total assets which is a pretty basic number uh on that uh data set so what are the policy implications i think having bank level data in the public domain is important for market discipline uh as i said market discipline comes with pilars three uh disclosures by the banks themselves but uh this is a complementary uh framework to pilars three disclosures and increasingly you get market discipline on a euro area based uh white basis i quote uh nordia again here uh but this is also market discipline is clearly and that's in the basal framework an element of the financial stability framework i will also say that with better market discipline and better transparency we should we should be able to get better evaluations for european banks so that's a collective action problem no bank has incentives maybe to disclose more than its peers but if we have a high level of disclosure for all as what i showed you in the u.s um then there should be a better uh valuation uh behavior of the marketplace for all banks and that should strengthen not weaken strengthen european banks in the international competitive uh playing field and level it by the way visa visa u.s my contention here is that u.s banks have an advantage not a disadvantage uh in the international playing field because of the high level of supervisory disclosure by the different supervisory authorities brought together in the ffic um it's particularly important i would mention for the less significant institutions which have none of their data published by the eba at this point and so the lsi framework i think is a big uh important uh concern going forward because my assessment which is very judgmental at this point is that the si frame supervisory framework the framework for the supervision of significant institutions is much more tried and tested than the lsi framework and therefore if you want to avoid accidents or incidents it's probably good to rely also on more market discipline in the lsi space um uh as not a substitute of course for a good supervision supervision but here again a compliment and importantly transparency can help policy development transparency would enormously help awareness by the policymakers of gaps in the current framework for example very few policymakers in some member states not all knows that a number of uh your own banks including some significant institutions don't use the same uh accounting standards as uh most of the large banks which is international financial reporting standards so some significant institutions and many lsi's don't use IFRS for their reporting they just don't produce the numbers there's nothing the ssm can do about it because this is a matter of legislation and regulation uh it's crazy when you think about it and needless to say in the u.s every bank no matter how tiny or credit union has to report under us gap uh so same with auditing uh and i think more generally uh transparency would support further steps towards a more complete banking union as you know all of our political leaders say they want to complete the banking union but at this point they're not acting on it or else said there is only two pillars and i would say even one and a half the one complete pillar being the ssm of course um and uh and to get the progress towards deposit insurance i think we need much more transparency about what is in the system what risks are in the system what behavior is in the system and that would dispel some false narratives that exist and that currently act as obstacles to reform so let me conclude i'm coming back full circle to uh data gaps initiative two and recommendation two dot twenty um we need good data we need it available for policy use we need accessibility we need if needed we're visiting existing confidentiality constraints and i think that applies perhaps even better to a bank level data than to other areas currently the gold standard is the u.s i dream of a world with which i think is perfectly attainable whereas the gold standard would be the eurozone and EU thank you very much thank you very much uh nicolas and for your for a wish list and also coming back and stressing very much the g20 uh data gaps initiative which is very close to our heart and also especially the 220 recommendation which i think is uh very very important in the meantime i it's my pleasure to welcome daniel nuy the chair of the supervisory board who joined us and she will address us in in 20 25 minutes but before that we still have a discussant brian blackstone uh from the wall street journal looking at his CV i would say he's always at the right time at the right place i'm not just referring to today's conference so that's obvious but uh he was in at the federal reserve and liman broke he was in frankfurt between 2009 and 2016 so from tlr to uh whatever it takes he had this all here live and currently for two and a half years he's been in zürich i just say 120 so no 120 happened i hope it's not an early warning indicator i'm a leading indicator so brian the floor is yours thank you uh you know i think it's uh kind of fitting that i'm the last discussant on the last panel because usually when there's a conference with a lot of policymakers and a lot of statistics at the end when the lights go down and everyone goes home someone has to write an article about everything and that's usually the journalist so fortunately i don't have to i don't have to do that because i get to speak about it instead so i'll in a sense kind of read you my article uh at least about this panel um it's you know i also think that it's a it's a good panel here because it in a lot of ways it's the it's the it's the universe of people who sometimes are involved in the in the articles that i actually write i'll i would uh go to a press conference or follow a speech by andrea or mario drage before or ben bernanke before that i would talk to experts like nicolas uh to get perspective for my article i would write it i would propose some lame graphics and then someone like nick would say that's not or not someone like alan would say that's not good enough we need to we need to do better and it would be back to the drawing board um i think that it uh you know one one conclusion i have on uh on a lot of the central bank communications especially when it comes to uh how statistics are employed is that it's is that it's really difficult i found that when i was uh covering the european central bank uh during the at the height of the greek crisis it's it's difficult to make simple declarative statements about what you're going to do whether it was we won't make any exceptions for greek debt collateral or we will not buy sovereign bonds or no member of the eurozone could could have a default it's a uh it can be difficult to make those kinds of statements uh the snb is is an example with the minimum exchange rate from 2011 until the beginning of 2015 i always actually had a very concrete problem with it because editors would say how can i write about a currency that's too strong and write about the minimum exchange rate so i had to find a fancy way to say that it actually uh find a way to say it was a ceiling on the franc's value rather than a floor on the euro's value but when the snb uh dropped that uh obviously there was a lot of uh fallout in the financial markets and these are these are difficult communications to have with the public and it's difficult to um sometimes to explain the underlying data that go behind it um and obviously not to uh turn it into a a press conference with the snb even though i didn't go to the last press conference so i get to ask all the questions now but some of andrea's points um you know again kind of highlight i think uh the difficulties when uh europe a central bank that's sensitive to the exchange rate the ecb can have communications about or the fed or the bank of england about the inflation uh outlook they can um they have a lot of forward guidance they have their own forecasts and you can put the economic data that comes out whether it's an inflation report or an employment report in the context of the forward guidance and you don't get too many market movements maybe as we would have uh 20 years ago when there wasn't as much transparency when there wasn't as much forward guidance when you're a central bank that's and i i've learned this in the last two and a half years when i'm in switzerland when you're a central bank that's very sensitive to the exchange rate that becomes a lot more complicated um andrea mentioned the uh the the the the swiss franc the the assessment that the swiss franc is is highly valued and she knows i ask i ask this every time there's a an snb press conference um what does a highly valued currency mean back in the old days that weren't that old uh central banks and finance officials wanted a highly valued currency robert rubin said a strong dollars in the united states interest jon claude trichet said a strong euro was in europe's interest uh back back in the day um so it's difficult to communicate what are the um what are the underlying data that would show that a uh a currency is too overvalued or whether it's just highly valued which might just be a a fact of life in a in a country and um and as andrea said that the the snb does actually provide quite a bit of information um on its uh on its uh weekly site deposits number that actually do allow you to kind of work back and say has the snb intervened or hasn't it intervened i think that the expressing the willingness to intervene is is an interesting dilemma for central banks for the snb because um you know there have been times when the snb has actually publicly said we intervene in the financial in the in the currency markets uh the day after brexit but there's a balance because if you share too much about when you're going to intervene or when you did then uh then silence might be misinterpreted by the financial markets and signal something that you didn't mean to to intend so i guess what i'm trying to say there is that there's there's a lot of challenges that are exacerbated when a central bank is very sensitive to the exchange rate um as we're seeing uh as we're seeing in switzerland now and i think that's an added dilemma than in some ways the um uh the ecb and other central banks don't have right now um and one other just sharing a little bit of uh you know of the swiss perspective um another kind of area where um you know the communications and the mixture of how to communicate data i think were were pretty relevant was in the in the recent um sovereign money referendum in switzerland which had a lot of interest uh overseas my editors were interested in it uh the ft people were interested in it and this got to a very basic concept of uh what is money uh and who creates it and how much is there and it was a little bit humbling for me because when i was working on this and having covered central banks for an awfully long time i didn't really know how money was created so when i had the graphics editor uh asked me um could we do a little a little visual chart showing how money is created now and how it would be created under the initiative i had to go back and redo my article because i realized i hadn't even explained it because i still wasn't sure um but it was a it was an it was a dilemma i think also for the swiss national bank because uh and this is something that central banks are having to do more and more it's not just a swiss issue of central banks getting involved in in in political debates and so in this case it was a i think by the snb standards a quite vigorous uh opposition if you if you read some of andrea's speeches thomas jordan's speeches uh it was a position that the snb was not used to being in and it was very interesting for me to follow that and to then have to go to the data and to find uh how is money created by the banks where is the data from the swiss national bank that shows this and how do i package it together so that i'm not just showing one lame line or bar chart that doesn't really uh add much uh for the reader so that was kind of a very concrete example i think of this intersection of um of of of data and policymaking and the reporters who have to um make sense of it all and also uh have it published on various platforms where people are reading the stories in a fraction of the time maybe that they used to when they would open it up in the in the newspaper um one other uh so i think one thing that would be interesting would just be what some of the lessons learned from central banks have been when it comes to some of these communications problems that they've had because in some ways the simpler the communication is and the more confidence it can instill in the short term the more difficult it is to unwind it and the more the potential for financial market disruptions are if a central bank has to take a very simple declaration and say that no longer holds anymore whether it's the snb whether it's the european central bank or whether it's the federal reserve during the uh during the leman crisis um on nicholas's points you know i think it was clear that it um you know that he's pushing the ecb to raise its game on banking statistics and i think the challenge here is the there's so there's such a richness of data out there when it comes to banking in real time the question is how do you distill it and convey it so that it so that it tells people something that they need to know because you could come up with i remember just during uh during the uh during the debt crisis this had this this wasn't necessarily banking statistics but there would be different times when there'd be kind of a flavor of the month when it came to economic statistics i remember uh at one time it was the private sector loan rates within different eurozone countries because it would show that a a small business in germany was paying a lot lower interest rate than a small business in spain or italy uh you know a few years ago you know target two was all the rage and then it comes it's still all their age exactly these and these things would these things would come and go um but it's clear that when it comes to the the the sources of data that we can get from the banks uh that can maybe provide real-time um warning signals um it could be quite useful that the challenge is there's so much out there that how do you how do you convey it to peoples in a way that they um you know that they can that they can understand it and they can relate to it uh on on allen's comments i was i was glad that he did include um one page from the wall street journal so he made up for uh sabine loutin schlegers uh comments last night um but uh you know i and it and it you know and it it it reminds me a lot of the interactions that i have with uh with with my own uh graphics team because it's a i am one of those people that will send uh five data points in an excel spreadsheet and say why don't you just put a bar chart there uh because the way we used to think about it would be you know i do the reporting for the story i do the writing for the story i'd file it it would be edited maybe it would be be getting close to being published and then they say oh wait what are we gonna do on graphics and i'd go to imf.org look on the look on the statistics database and and try to find some data that would that would help me or i'd go to the ecb balance sheet data or euro stat uh or now the snb's website and now it's obviously something that if you don't have a good vision for what the for what the graphics are going to be in an article and how it's going to work both in print and uh and mobile then um then your then your story is not going to get the um you know not not going to get the attention that uh uh that it should have my i think the balance there is having visuals like the one on the on the vaccination rates that are that are so detailed and uh and so um uh and provide so much information that is the reporter i might say well why would anybody even want to read my article after they've read the graphic so that's a uh that's always a challenge for us to anyways those are my comments thank you very much okay thank you very much brian also for this very very open and honest words about the world of journalism so i suggest we collect uh quickly a few questions and then let everybody andrea wants to react immediately no i was i thought brian had a very nice framing and some of the lessons learned um and maybe i could just say a few words on the lesson learned uh maybe someone else and then is that okay because it's it's actually quite um short but what what i was thinking also with as i was preparing for my um i think and and and it and it is something that among central banks and we ask quite a bit about going forward because we're still in a normalization and what does it mean i think there are a couple of lesson learned and and if you think of the world 10 years ago one is the policy tools need to be simple you need to be able to explain them in a simple way and i think that that i think going forward it wouldn't surprise me that the way we even think about monetary policy the tools need to be simple we have to answer the questions and i think every press conference shows it how important it is to answer the question you have to be focused in what you want to say if i think about the initiative on the on the on the full money you know you had to figure out what are the two things you want to say for instance in our case was the referendum whatever the objective it says it will achieve will not be achieved and uh and it will have an impact on our ability to implement monetary policy that's it you have to be able to put them in a nutshell and the last thing is but in the end we are policymakers in the end we are a central bank we have to do what we have to do no matter what financial markets are going to react or not react in the end there is a reality we have a mandate and we have to fulfill that mandate so that would be just my quick lesson thank you thank you very much andria so now we give a hand Selmot