 Let's start the last session of this today's program. We will have Bengt Holmström from Boston joining us. Bengt is the Paul Samuelsson Professor of Economics Emeritus at MIT. He is best known for his theoretical research on contracts. He has made fundamental contributions to the theory of the firm, to corporate governance and to liquidity problems in financial crisis. In 2016 Holmström was awarded the prize in economic sciences in memory of Alfred Nobel together with Oliver Hart. Holmström has served also in many non-academic roles. I want to mention his less known role in providing independent advice to the Finnish government. That work has started in 1992 when he was with Seppo Honkapohja, the key members in an expert group that prepared economic policy strategy for Finland after a deep recession in the early 90s. His last contribution for policy was in the context of the COVID-19 pandemic when he was part of an expert group that prepared a policy strategy for the economic policy strategy under the pandemic. Holmström's results are often surprising, the research results. However they are readily understandable and they have a sort of clear intuition and they are also based on realistic assumptions. I think we can wait also this time something surprising, something interesting, something exciting from Professor Holmström. Holmström, welcome to this workshop. This is last session of the first day of this. We are delighted to have you present from Boston in this workshop. Please, the floor is yours. Thank you, Marty. Thank you for inviting me to this conference. I'm very pleased. I would have been even more pleased to be in person there but this is the reality today. I want to talk about the topic that I have studied since the great financial crisis 2007-2009. It was work that I started in those days. Some of you probably have heard some version of this or something very similar. I apologize but I still think that for this audience I checked. For this audience this may be containing new ideas. That's what I'm presenting. Basically I'm going to lay out the reasons for the financial crisis and then finally come to the end to talk a little bit about the effects it has through digitalization on emergency economies, how this perspective is affecting it. I just have to, I don't know why it's not changing the page here. I'm a little bit unfamiliar with these themes. We use tools. Can anybody there in the technical side say what I should be doing? It worked just fine but we tested it yesterday but now it's not moving. Now I have to break it up. Thank you. Let me start by the crisis itself in 2008 when it broke out. That was a shock because we had that 70 years without a crisis so nobody even expected that in a developed world like US and then subsequently in Europe that there could be a crisis. The instant reaction to the causes of the crisis were many but by and large it was the criticism fell on Wall Street. Wall Street greed and wrong incentives and securitization created complex opaque asset backed securities that nobody really understood what was in the bags of chopped up small pieces of mortgages. And then the ratings agencies got an air full and for having been careless. I often mentioned Michael Lewis, the big short some of you have seen the movie. He asked what I think is the key question and that is that how could Wall Street trade without knowing really anything or so very little. In his movie and his book he lays down the fact that nobody kept trading on Wall Street but very few people seem to know what these different very big sized trades were worth. And so the first instinct was to call for more transparency and so as to avoid this crisis. I want to present an alternative view. I think that's the question is right. Why didn't Wall Street know anything but basically going to put forward an explanation why they didn't. And that is that the money markets as these markets are that is very short on debt. They are repo markets. They are they are short overnight treasuries and treasury bills and so on. They are these very short on debts that get rolled over frequently. Huge by the way size of this market is many times larger in terms of dollars than say in the stock market. So stock market gets all the attention but these money markets are gigantic and global. And so I'm putting forward the notion that no questions asked is liquidity money markets that is they were very liquid and the only way they were liquid was that one did have to ask any questions. And the bad that this was astute observer of banks. It was a banking expert. He said already in the middle of the 19th century that every banker knows that if he has to prove his worthy of credit that is if he has to become transparent in fact his credit is gone. So another way of putting it is that ignorance ignorance that people don't know they just don't have to ask questions is almost bliss in money markets. And that's that's the opposite. So it's not as parents is not going to solve the problem in a way that people imagine and. And the problem here underlying this sort of view has to do with the fact that the stock market is very dominant in this in this in people's minds because the stock markets are exciting all the all sorts of things happen in money markets when they are working well nothing really happens. And people are very unaware of what what's going on. So so I'm going to make the thesis of one of the key thesis I want to say is that that you know stock markets and money markets are very different almost each other's opposites. And therefore I should draw lessons from stock markets to deal with the problems in the money markets. That's one of the main lessons. Let me just say that why isn't transparency important. So there is this sort of common common but false influence that is that that that somehow transparency will make us believe the same things when in fact very often it's just the opposite. So symmetric information about payoffs. That is the payoffs from the instruments. That's that's important for liquidity. That's everybody agrees on that but transparency does not imply symmetric information. In fact in fact it may be an often is the case that if we make something transparent then parties that watch that same information actually start to believe something very different. So imagine that you put that up. You're buying a car and you look at it from a wrong distance and there's a mechanic next to you that's bidding on the car also. So these are typical forms of bidding. Then if you know that the mechanic has seen the service book and looked at the hood and driven the car then you don't want to bid against him because he knows a lot more. He understands what this means. But if you look at it from this long distance which they usually do that is they sometimes they just flash the car and then you have to bid. That makes the uninformed person about cars put him on the same in the same position as the mechanic. So actually hiding information in cases of expertise, varying expertise is very important. We process information so differently because we have different experiences. So there are two ways to symmetric information. Investors know everything of relevance for instance the stock prices or something like that. That's what the stock market does in terms of getting symmetric information but symmetric for ignorance that you are going to get through over collateralized debt as I will show. That's a much more common way of actually getting to symmetric information especially money markets. So these are the main thesis. So why debt? It's a good question because we get always in trouble. All financial crisis has to do with short-term debt sort of breaking down. So why are we at all playing with debt? And the short answer is debt is cheap. It's a cheap way of lending. And so this goes back all the way to pawning which is thousands of years old. All these documents of pawning are from China in the 600s and so on. So why is pawning so efficient? And that is because it doesn't require very much information and it especially doesn't require that we agree on the value of the port. That is the item that's being poured. So handling over price if I try to sell something when I have a liquidity problem then maybe I value it more than Marty or anybody else, the pawn shop. But all we have very different views about what this is. Maybe it's my grandfather's watch or something that has emotional value. So the solution to this problem of costly selling and handling over the sales price is this very clever solution. That the pawn broker buys the pawn and he literally buys the pawn at a safe price for him. So if I think it's $500 the pawn broker may offer $50 or $100 or whatever. On one condition I accept this trade and that is a tongue given the right to buy back the pawn. So if you only offer me $100 then in a month I will want to be able to buy it at $105 or whatever the added interest rate is there. And the key point here is that this is all designed to avoid price discovery. This is the key. This is what makes debt so cheap and quick is that there is no price discovery. The repo market is the world's biggest most liquid market in a money market. And I'm talking about the US but repo markets elsewhere also. It is actually almost the same as the thousands of years old pony. It's a little different in that it may be the pawn shop that wants to park money with somebody. But otherwise it's a modern day version of pawning and the basic project is very similar. So let me say a few words looking at it a little differently. The design of the debt is designed to be informations insensitive. It's not just to overcome this problem of having to buy it or having different information or valuation of a pawn. It's also the fact that when you look at broader areas, the market for debt, especially the short-term market, then it's designed to be so that you don't have to ask any questions. And this is a picture for some it means a lot or it's simple. The red line is the payoff from the collateral is placed to support the value of debt. So there is a face value of debt, say 100. And there is collateral that is both more than 100 in order to secure the debt. And the more you have collateral, the more you go to the right and the red line goes to the right. That is the value of debt as a function of this collateral value at the time of expiration of this loan. So the red line is the market value of debt at expiration. And then the thin black line is market value of debt looking towards the end. That is if it's one month left to go, then it's below this red line because it's an option. It may go into default. And eventually the more collateral you have, the more valuable the collateral is, then eventually it stands out with the red line. So it's 100 if it's the case that there's a lot of collateral underneath it. Plenty more than the face value of debt. And that's the information in sensitive region. But if something happens and we get into closer, you know, the collateral value falls as we will see shortly, then we get into the information in sensitive region. So that's the basic idea. Let me just mention that one of these things, there's purposeful opacity. If you actually start thinking about this, you see that the bear sells diamonds by basically putting them in a bag. And you are not allowed to look into the bag. So it makes it more liquid. Credit raises are coarse and money market funds have delayed information release and so on and so forth. So let me say what the problem with debt is. I don't have that much time, so I can't elaborate on these things. These are very dark side of debt. And that is that relying on debt, on securitization, coarse ratings, mechanical rules for valuing this. The rating agents use very mechanical rules for valuing these securitized products, the ABS's and MBS's and so on. That all makes good sense in good times. But there is one concern about it. You don't remove any risk by making it more collateralized. The fundamental risk underlying is it pushes risk into the tail. I should have not said that more collateral will reduce the risk. But it will push the risk into the tail when you make a debt contract. So you make it very safe in the flat region when, as we say, the debt is in the money. But if it ever goes out of the money, that is it goes into the red region, in the information sensitive region, then the risk gets bigger. So the social trade-off in some sense is everything that enhances liquidity increases, increases also the severity of tail risk. So you can get more of it. For instance, you can shorten the duration of the contract. That's going to make it more the region, the red region or the information insensitive region factor. But if it ever goes into a crisis, then it's going to cost you a lot more. So one way of, I'm going to have to jump over this thing. One important thing if one asks what happened in the financial crisis is that it was assumed that diversifying in debt markets is a good thing. That's one borrows from the stock market the idea that fully diversification really makes this safe. So here is an example that if you have ten identical banks, each holding say one-tenth of the total portfolio of the debt portfolio of the country. So everybody has identical portfolios. They are fully diversified just the way stock markets are the stock market advices. Then you realize that if every bank has exactly the same portfolio, a slice of the same portfolio, if one banks go down or its portfolio breaks the buck so to speak, that is it isn't fully valued, then everybody does it at the same time. So the banks are perfectly correlated in this case. So this is an example how the logic from stock market has absolutely the opposite consequence. And pre-crisis people believe that even though debt levels kept increasing, they said well it's very well diversified. Actually that's a disastrous thing on the whole if it goes to a very extreme case. And this has been proven, exposed empirically that actually this was a big reason. Things were correlated too much. So I jump over this shift when the distribution shifts from information sensitive to information sensitive. Here you see one illustration of what it looks like. Let me just say that this is actual trading data from money markets. But it shows the left line is very stable and these trades are just clustered around. That is they are all made roughly at the same price with small variations. But then when the bear fund collapsed in July 2007, so it was a year before the Lehman or more than a year before the Lehman, you can see how there was total confusion in the market and the prices just spread. Different parties bought at different prices. They were not at all uniform the prices. So this is an example what happens when you end up in that tail where the money, when debt is believed to not be anymore worth 100. It's something less. But because they never was a price discovery of the tail of the actual collateral value, people are totally confused about it. Here's another picture that you see. This is 2008. You see how steeply they kept issuing more and more debt, structural products and more debt. These are different varieties. You just see that home equity loan is actually the biggest. So this talk about the subprime loans, they played basically no role in this whole story. But this whole market dried up. Nothing was issued. And it's interesting to note that in money markets you see because there is no price discovery, it's always quantities that adjust. Haircuts or in this case, we don't issue anything more and so on. So again, this is very different than in the stock market. So people were really puzzled that nothing traded for a while after the Lehman crisis. But how could we not find some price that we can trade? And the answer is the adverse election was so severe that people had such different views about prices. And they were afraid that others knew more than they did. So the summary of this view that I presented is that money markets and stock markets are two polar systems of liquidity provision. They are both liquid, but in very different ways liquid. In money markets they are urgent, trillions of dollars of people roll over every day in those days especially. Everything is information insensitive in the sense that there's no price discovery. Nobody knows anything expert like really that is of value for trading. And there's just a shared understanding and it's trust based. So it's a no questions asked environment. And that explains that's the answer to Michael Lewis' question. Why didn't anybody ask any questions? The answer was because the market was working well. And that's the sign of a market working well. If somebody asked questions in money markets, we are already in trouble just like Badgett suggested. Stock markets are very different. You can wait for trade. Money markets you have to trade. You have to roll over these debts every day. So if somebody stops trading like J.P. Morgan did in the case of Lehman. It's just a heart attack. And because you are thrown into the situation where you have to try to find out prices or start a market where there never was one price discovery market. It's information sensitive. So every piece of information matters. There's very accurate and continuous price discovery. And it drives in some sense what drives this market is that there is all the time information coming in and people have somewhat heterogeneous beliefs. So there is people that can make money on these trades. You don't make money on money markets because everybody believes the same. But when you have different beliefs then you can make for short periods of time you can make money. So this is what the stock market is and here transparency is critical. Here every even today we are talking about nanoseconds mattering. And so it's a very intense, very expensive market relative to money markets. Money markets cost very little to run. But loans are made. Trillions of loans are made in Snapchat in very short periods of time without any information to stock market in order to get started. For instance, it really requires a lot of if you want to do an IPO, you have to build a book. It takes months sometimes to do it and so on. So it's an entirely different. Let me emphasize money markets are very liquid because they are two leg. There is this pawn shop logic. So it's what we call a two leg market. And it's a primary market. Secondary market then bonds, for instance, I issue. Then you have to find the price because if you are trading bonds, that's a price discovery. You need to know exactly the price because it doesn't have that second leg. And those markets are highly liquid. And in fact most bonds are not traded at all. So this is all to say that we need to understand. So this insight from the information view that I described is that it answers the question. Why didn't people ask questions in good times? Because that's the way it's supposed to work. Why ratings were caused because that helps us have the same beliefs. If you start to provide detailed information, then experts become better than other experts. So you're hiding information. It also explains why it was so chaotic when eventually the money market funds broke the buck. The role of government in crisis, it has a very simple implication. Even at this level, you're right away when you understand it. You understand that the key is to get back to the no questions our state. Trying to buy up toxic assets as the US plan to do initially. With the $700 billion that they got from Congress, it would have been a disaster. Because you would never have known whether all the toxic assets or sufficiently many toxic assets had been bought up. So they abandoned that plan that they initially had and went into the idea of just recapitalise the core of the 19 banks, big banks, 80% of the banking system, recapitalise those banks. And there they've been transparent for a while by showing that if we pour in this, I think it was less than $100 billion that eventually they put in. They said that's enough to get us back into the no questions asked region. After that it became more or less non-transparent again. So we have stress test, regulatory considerations are. We have got higher capital ratios. We got liquidity ratios, which I think are very questionable and partly a course of the troubles we are seeing right now. So I'm not in favour. I won't go into it now. You can ask questions about it. But stress test I'm very big in favour of. But I'm not sure these stress test should be as explicitly revealing as they are. That is I think they should look like elevator tests. It should be people coming repairing something but they shouldn't put on the wall that last time we repaired it. These cables were almost broken and it could have fallen down and so on. You don't reveal that. You just said it's fixed and it's good. So then the third thing to keep in mind is that transparency. If you want to reduce liquidity in money markets, then you can put in information. So that experts come in and that actually excludes a big bunch. We saw it actually in the money market funds nowadays. They became more revealing after the dot-rank transformation. Just like this theory predicted, half of the money markets actually cease to exist as money markets. They became just trading markets of different kinds. So their sort of moniness disappeared. So let me say a few words about the big data because I know both your people who are working in the area of emerging economies. So I'm very excited about the big data. It's an entirely new form of inclusive financing or financing. And it goes against the idea of just putting collateral in. It's as I put it, information is becoming the new collateral. So it's reversing the idea that information is expensive and collateral is cheap. In the cases where we are trading on big platforms and getting data for free in some sense, then we can do some very different things. So the leader in this area I think is clear and that's China. But many other developing markets have also Africa's hat. But China is really in terms of volume and sophistication. It really leads the world. And the reason they are the leaders I have to do with having been behind. They never used checks. They didn't have PCs, for instance, the computers. They went to mobile internet almost directly for most people. They went from cash to actually paying the mobile phones directly. And so this being behind has allowed them to leave for business societies. So they are by far the most advanced in terms of payment systems. So these payment systems we saw brought from the platforms of Alibaba and WeChat. Through there Alibaba and WeChat Pay are the payment, basically the main payment systems available in China. You can't go there with cash. You can't pay with cash anywhere. Even betas use Alibaba and WeChat Pay. So the payment systems, everything gets recorded. And there's massive amounts of data and upending the traditional substitution. That information is expensive. It's very cheap. It comes for free almost. So as I said, information is the new collateral. And mobile credit is superior version of credit cards. So Alibaba, for instance, they have also a credit line. It's not that you apply for credit. So that's an interesting feature. You just get a message into your phone and asking, do you want to increase your credit line, so to speak. We have evaluated you basically and we have found that you are good credit. So here is an offer to get credit. So it's similar to the Western world, but you can't apply at all. And also everything happens electronically and without any human interaction at all. So it's called the 310 system because it takes three minutes to deal with this message. It takes one second to get the money into your bank. And there's zero human involvement. And one of the things Alibaba uses as a slogan, they know you in the sense that they have to know you in order to give you credit. But they don't know who you are. So it's based on partly at least on the pseudominium, which is a future that the best has started to be explored. The idea that we have a second identity on the net, that is say a number or social security type number or something, but very few know how the mapping from that number goes to our actual personal information. So this addresses partly the concern about privacy. The enforcement mechanism is very sophisticated and it's by the exclusion and continuous monitoring of fraud. That's one thing. But exclusion is very important because there are only two payment systems. So if you somehow get included from the trading platforms, a lot is taken away from you. So there's very little fraud in principle and people behave. And these algorithms that evaluate people based on how they purchase things and what they are doing on the platform and how they are trading with each other. All the information is in some way used. Who are you? Who are your friends? Who are you dealing with and so on? Sounds horrible, but actually it's very informative. And they just mentioned that even if you just tell that do you have an iPhone versus an Android phone, that already is informative in forecasting whether you will pay back your loan. So that has been tested in America. It's a little scary and Android people are poorer credit than iPhone and it has just to do with the prices. So these are dramatically lower costs and what I find most exciting is this has had a dramatic impact on inclusive financing in China. I mean the 200 million or 300 million people that have brought out of poverty. This has played a role. This mobile internet has played a big role. And if I may just say Finland. Nokia has done a lot in this time. It's not talked about much in Finland, but actually they were by far the leading network provider and mobile phone provider in the early days and had over 50% of market share in many of the developing countries Africa and India and even 40% in China. So this is I think an enormously valuable thing. Let me mention and I could talk about it a lot because I've been involved with law and academy in China studying this. I would just want to make mention of B-cash in Bangladesh because Bangladesh was very poor and this B-cash was partly the help of well B-cash was subsequently helped by investors like Alibaba, Softbank, Bill Gates, IFC and so on. So major investors have become excited about it. But it was established in 2011. So it's only 10 years roughly old. And it now has close to half of the Bangladeshian sort of addressable public customers and of them about half is active users. Active users means they use at least once a month these accounts. So it's not just that they can buy things and so on. They also get banking services more generally. They get the bank account. Most people absolutely the majority have no bank accounts because they lived in villages and so on. So now they have with their mobile phones, they have a bank account and they are using it and they can buy products from far away and again I say China is far away the most sophisticated and most developed market. But Bangladesh, you look at these numbers daily volume 11 million transactions and yearly transfers 40 to 50 billion inside Bangladesh. That's a lot of money for a relatively poor country. Bangladesh, I think partly thanks to this, has recently surpassed India in GDP per cattle. That's remarkable considering how poor it was just 23 years ago. So big cash is something that you should be aware of. Let me say that the key point here is that I think there are enormous opportunities ahead with this microfinance based behaviour credit scoring that I have described that you just watch what people do on these platforms but various kinds of platforms. What kind of insurance they buy or whatever they do everything is followed. The point is that this is very reliable first of all in terms of forecasting how their credit scores are and secondly it's very scalable. This is the remarkable thing. You can go the numbers in China way over half of Chinese population using mobile phones or have at least access to mobile phones. Bangladesh is coming behind as you saw and nothing like this has happened. If you look at the traditional microcredit where you have these communities that you lend to villages and they monitor each other and therefore they are more reliable it requires a lot of work to get it set up and obviously a lot of human work. It's just not very scalable and it's also very spotty to results because you need to have skills. Under this author some of you may have his idea of creating new property rights for these poor people. That's just massively difficult to get registrations to do that. I think by far the best way, the future for eliminating poverty or reducing poverty and also getting these unbanked people in the developing countries into the periphery of finance which will bring the amount of poverty is this system that I described here is this behavioral credit scoring system. Blockchain technology is going to make things a lot more secure as well. It's used a lot. There's a lot of discussion about that but I think I stop here in terms of telling. I wanted to flag this. This shows you that collateral can be reduced. Let me make one more point. Collateral is still very important. This sort of mobile finance mobile lending is not that big a share of it. It's a very big share of the poor people's purchases. It's coming actually to the US also just because it's a lot more convenient than credit cards. The last point I think I had to make is that J-PAL there is something that is part of my department at MIT and it was started in early 2000 and they have hundreds of these evaluations of these behavioral credit scoring efficiencies and I think they are pretty excited about this. Thank you very much. And now questions from the works of participators. There's really two paths. Yes, please. Two paths. There's this general thing about information sensitivity and so on and then there's this part of innovation or inclusive innovation, innovative finance. Please. Thank you very much. It was a very enlightening experience for us. I have two questions. This one is that like say in the in your diagram there is a different default zone and there is a non-default zone depending on the value of the collateral, right? So now so if you kind of in the pawn shop example suppose if I give some watch watch is kind of a tangible object it is verifiable but when actually in the actual transactions in the modern days, shadow banking and etc. So when the kind of instead of the tangible collateral it became kind of other financial assets, right? So which must become a back done another real asset, right? So isn't that kind of a doesn't that create an adverse election? Sorry, I could be wrong. I mean if I could understand you that is in that zone no non-information I mean in that default zone or what do you call kind of a information sensitive zone, right? In the collateral wouldn't that be a possibility of adverse election in exchange? That's my first question and my second question is that how do you kind of relate to mobile banking in kind of reducing the risk of transaction to enable them enable lots of poorer people make an entry so that part is kind of a little bit not clear to me. Yes, please. So yes, thank you for giving me the opportunity to clarify the picture so exactly right. There is an underlying value of collateral is on the horizontal axis and initially we are far in the right side so that it is information insensitive because the collateral value is substantially over the obligation to pay back debt that if debt is 100 you know the collateral value may be 150 or so. One key point is we don't have to have to say we know there is enough collateral that's the big sort of reduction in information but if we are asked you know what do you think the collateral is worth or are somebody else we have very different valuations of the collateral but none of that matters because variations in the collateral value and as it changes over the days and months perhaps in this case mostly days because these are short on debt nothing happens to the it's still 100 it's still well above the 100 so that's the point that's why this information is sensitive to the region it's not sensitive to changes or shocks to the value of collateral or even the people's valuations of collateral so we don't have we are sort of in disagreement if you ask us about the value of the collateral but it doesn't matter for the value of the debt or the repayment probability and therefore the value of the debt so that's in the information in sensitive information in sensitive region then collateral in rare cases but as we saw it has happened the collateral value start dropping and that's because that causes us to move to the left in the picture of debt and we are coming then the thin black line eventually will disconnect from the red line so it's no longer 100 but you know the let's call it the market value though there's no market value on it conceptually it's less it's not worth any more 100 in the minds of people and that's a moment that usually I do it smoothly but usually you take a jump something unexpected happens such as the beer stones default or somebody breaks the buck elsewhere in the money market fund or something those are very shocking events and then you are right so getting to the adverse election then suddenly it matters that some people know different we all know different things so this massive we are going from a state of no questions asked we are all on the same page we are thinking about the same way in terms of the value of the debt we are suddenly thrown into chaos because we are we are we now realize that it's not 100 but what is it and we have very different views and there's adverse election I'm not going to touch it because somebody knows more than I am I'm not going to trade with anybody and that's actually what happened in the United States so they had no idea of what the value was suddenly so did that answer your first questions on the mobile banking I think I will have to just default that question to some other discussion but thank you because I understood that is from the horizontal line whereas information is insensitive there actually doesn't really matter what is the value because already the loan is covered right so but it is that actually you answered my query kind of very well because when it becomes like the black zone you know the black art zone that is information sensitive zone there could be probably the market could be the market participant might invest in some information to information acquisition or something like that could happen probably in that zone probably some more action could be happening where the information sensitive zone I'm talking about yeah that's but I'm telling you just the empirical fact it is chaos most of the time if it happens in a little place you know it takes I mean if you look at what happened in the 2007-2009 the crisis really actually started 2007 already so it was a year sort of simmering but it was mostly in the over the counter market and therefore nobody saw it so it was still hidden but those in the market they actually saw that something was happening and they got increasing numbers and bear stones the firm had to be rescued by the Fed in March 2008 and so on so there were kind of shocks some of them rather significant but as we saw the government came in and rescued so people people felt comforted by that but so it doesn't have to completely collapsed but they will substantially reduced in value initially in the OTC market the collapse paradoxily came in the safest supposed the safest market which was the tri-party market because their adjustments to hear cuts and such that took place in the OTC market they didn't take place in the same way in the the tri-party market because that was overseen that was more transparent and it was overseen by these banks JP Morgan and Boney as it's called Bank of New York and so it was less abrupt but then the Lehman collapse was a big enough event to get everybody thrown into caves and after that nobody really knew at all what the values would be because they couldn't see what consequences it could have and it was immediately the money market that was contagion so they basically the Fed had to come in and back up everything it's a very chaotic time I lived through it because I was actually involved with the New York Fed at the time and the first days they didn't know whether the whole world would collapse and this is a concern by the way today I think the same situation it's more transparent in some ways through the stress test and so on but these discontinuities are in some sense always there because once the system starts collapsing everything feeds on everything else and it tends to be very chaotic so I hope nothing will happen but I worry that central banks have used up their sort of gunpowder to defend against this crisis in several steps already so they don't have as much left certainly some countries won't have much left to protect their currencies and prevent crisis Can you explain this sort of I know that you have explained this what happened in the sovereign debt market in Europe how was the information sensitivity removed from the market during the euro crisis it may be yeah that's a fun case to see this sort of logic information logic at work the EU situation in Euro crisis was it 2010 2012 so that spread from the US to Europe that you saw that the spreads went up between the sovereign bonds had differential spreads Greece Greece was the main object of concern at the time and the spreads kept growing and the response by the EU was to tell people that now we have 300 billion set aside to defend against this which by the way is much more than the US had they had only used 100 billion but then they said well actually we have just increased it to 500 billion and eventually people started calculating that if we take all our sort of all our weapons in use then we will have actually a whole trillion nothing of this really affected the spreads this sort of explicit mention of how much more they had it didn't do much good and one reason being that when we collapsed there is almost no amount of money enough to cover everything so a trillion is a lot of money but that's not enough to save all the European potential bankruptcy then eventually they did stress test also and they did but they did stress test they took the first half of what the US did and they sort of explained what the bank they studied the banks and and they came up with a ridiculous number like you know oh we need two billions to save this system or I don't remember what the number was but I made it just very small and so people asked well what do you mean did you test the Greece and the answer was no Greece we couldn't test because we don't know what is in the Greece Greek banks and everybody said oh my god that's the only thing we are worried about is really Greece and so then they opened up their books and showed all the banks including the Greek banks what they knew about it and opened the books so to speak and of course that was in this theory that I have propounded that's the last thing you should be doing when you approach a crisis opening the books because then you are really opening the door to adverse election and so things got just much worse by that by that moment and it was eventually druggy that resolved this whole thing by saying you know he only said that we will do whatever it takes and you better believe us and the spreads came down and this was a very a fake statement the assumption in the minds of people was it took all the experts experts can't put the sentence whatever it takes into a spreadsheet and start calculating so it brought sort of it took away the experts expertise to some extent and also it assured people that probably they have been talking with Germany and they were but it was the opaqueness of the statement that we are behind everything was ultimately what resolved that crisis unfortunately Europe hasn't used the time to really fix their banking system and it's been 10 years since druggy said that and now we are again in sort of uncertain waters thank you are there other questions yeah Kunal please first time it's Kunal Sen from Univider I wanted to focus on the second part of your presentation I completely agree with you absolutely right the microfinance the Grubbin Bank the back experience has simply not been replicable other parts of the developing world exactly as you said because of scalable problems and I totally agree also that this new approach to microfinance the big data approach might well be the way forward but I wanted to ask you that what can one learn from a Chinese experience with VP and Alipay and the Bangladesh experience with big gas for other low income countries what can governments do to try and do what exactly happened in China and also happens in Bangladesh that I think is important policy and I think I think I think what you learn is that this work on the whole pretty well I mean they are not working perfectly and that's why this J-PAL this institute that's this randomized control trials that have become so popular they are trying to study exactly what the effects are from these systems but but they work well enough Indonesia is going there Indonesia has just started started moving there India is already having its own I'm not that familiar with India but India is having its own version of big gas and I'm sure so learning what they can learn from it is basically see Alipay has shared with Bangladesh its information with big gas its information about they have invested there and they have sort of described what they are doing I'm not private to knowing what other parties have shared information but it's not like you can sort of these are automated systems these are computer systems I mean they are as I said very scalable also in the sense of transportable whereas we know that microfinance for instance of the old variety that Mohammed Yunus was using and that was very valuable and still plays a valuable role but it's just small scale because it takes so it's very spotty some communities it works well some communities it doesn't work well it go to another nation it doesn't work at all well so you don't know how it works whereas it seems like this this computerized systems which are based not on interviewing people or anything like that you just look at what how people are behaving on this and by the way my colleagues Abidit Panagy and Esther Duflo who won the Nobel Prize in 2012 19 for this work on the JPL and you know these randomized control trials they they they are enthusiastic also about this and agree that you know it's sort of much more robust than much more transportable than and they one of the point I don't know if you have read poor economics is that poor people are very astute we westerners and you know people in rich countries think the poor people are just lying in the streets and they can't calculate anything they are actually if you go a little bit about that that total misery you see that people who are really economists in this world or behaving like economically sophisticated people are the poor people you see it for instance that they are paying you know for mobile phones even everybody has mobile phones in America even the homeless here there's nobody anymore without the mobile phone because it's so valuable and with that base you know as the tool of connecting is in your pocket all the time that makes an enormous difference and the Bangladeshians incidentally learn to read people didn't understand who are they going to call and how are they going to call when they can't even read and write it turns out that it worked in the reverse when they got these phones originally in Bangladesh which also was involved this career family was involved in that that was an enormous impulse to learning to read so there are other effects coming from this that have been very beneficial for in my understand I'm speaking to people who know a lot more about poverty than I do so I'm a little hesitant but I have followed this particular this mobile phone arm both as when I was on the Nokia board and then then later with the law and academy in China thank you all right if there's no other questions thank you very much for this very informative presentation and discussion and thank you for those questions and the session is closed and I suppose dinner is at 7 so thank you very much yeah